Chapter 1 — Definition, Landscape and Stakes

A brand is not a logo but what people expect of a business and remember about it, and this chapter defines the brand, its parts and its value, describes the landscape in which a small business builds one in 2026, and sets out what is at stake.

What You'll Be Able to Do

• Define a brand as the gut feeling a person holds about a product, a service or a company, and distinguish it from the logo, the slogan and the advertising campaign that are often mistaken for it.
• Separate the three layers of a brand — the platform the business decides, the identity it shows and the image people form — and explain why only two of the three are under the business's direct control.
• Name the parts of a brand, from the platform and the verbal identity to the visual identity and the experience at every touchpoint, and say which chapter of this book treats each one.
• Describe the landscape of 2026, including the number of touchpoints, the accessibility requirements that apply to text and interface elements, the stable format design tokens received in October 2025, and the Québec rules on French on signs in force since June 1, 2025.
• Explain what customer-based brand equity means for a small business, and why a known and trusted name receives a better response to the same offer.
• Run a first inventory of your own touchpoints, count how many of them are consistent with one another, and keep the list for the work of Chapter 7.

Why It Matters

Most small businesses already have a brand, and most of them have never decided what it is. The brand exists because people hold an opinion about the business whether or not anybody wrote one down. A customer who says the technician arrived when promised, a property manager who says the invoices never surprise them, a supplier who says the owner is straight about what the work will cost — each of those statements is a piece of a brand, formed from experience rather than from a document. The business did not choose those words, and it did not choose the feeling behind them. It earned them, one visit at a time, and the same mechanism that earned them can lose them.

The problem is that the earning is invisible until it is counted. A business of seven people can hold three versions of its logo, five different blues and four typefaces without anybody noticing, because each version was reasonable on the day it was made. A proposal can go out with one heading and a monthly report with another, and both can be perfectly professional. What is missing is not effort but a place where the pieces are added up, and the cost of that absence is not aesthetic. It is the cost of being forgettable in a market where the buyer has four other names in front of them, and of being unrecognizable at the exact moment a prospect is trying to remember who did the good work on the building next door.

This chapter sets the terms for the rest of the book. It defines what a brand is and what it is not, separates the layers that are often confused, names the parts that will each receive a chapter of their own, and describes the landscape a small business is working in during 2026. It then sets out what is at stake, in the language of customer-based brand equity rather than in the language of taste, and closes with the vocabulary that the remaining twelve chapters use. The running case, Nordec Entretien inc., appears here for the first time with the inventory that started the work: thirty-six touchpoints, three logos, and thirteen of those touchpoints agreeing with one another.

In This Chapter

Section 1.1 defines a brand and separates it from the logo, the slogan and the campaign. Section 1.2 distinguishes the platform, the identity and the image, and places Figure 1.1. Section 1.3 sets out the parts of a brand and the chapter that treats each one. Section 1.4 describes the landscape of 2026, from the number of touchpoints to accessibility, design tokens and the Québec rules on signs. Section 1.5 explains what is at stake for a small business through customer-based brand equity. Section 1.6 gives the vocabulary of branding in twelve terms. Section 1.7 sets out the laws that apply to a brand in Canada, Québec and the United States. Section 1.8 introduces the running case and the plan of the book.

1.1 What a Brand Is, and What It Is Not

A brand is a person's gut feeling about a product, a service or a company, and that feeling is not a file, a color or a mark on a page (Neumeier, 2005). The definition is deliberately small and deliberately human. It places the brand in the mind of the person who holds it rather than in the hands of the business that made it, and it explains why two businesses with identical offers can be remembered in completely different ways. One is the company that answers the phone, and the other is the company whose name nobody can quite place. The difference is not the quality of the work, which may be equal. The difference is what has accumulated in the customer's head.

Neumeier's book treats the brand as the bridge between strategy and creative work, and it names five disciplines that hold the bridge up: differentiate, collaborate, innovate, validate and cultivate (Neumeier, 2005). Read as a sequence, those five words describe the whole of this book. A business differentiates when it decides what it stands for, collaborates when it brings the people who make the materials into the same room as the people who decide, innovates when it finds a way to say something that the competitors are not saying, validates when it tests whether the work lands, and cultivates when it keeps the whole thing alive over years rather than weeks. None of the five is a design task, and all five produce design work as a by-product.

The brand is also an asset of the business in a sense that accountants and strategists both recognize. Brand equity is treated as a set of assets linked to a brand's name and symbol, and names, symbols and slogans are treated as intangible assets of the business (D. A. Aaker, 1991). That framing matters for a small business because it moves the brand out of the category of decoration and into the category of things that can be built, damaged, protected and eventually sold. It also carries a warning that is easy to ignore in a slow month: short-term moves such as price promotions can erode brand equity (D. A. Aaker, 1991). A discount that brings a customer in today can teach that customer to wait for the next discount, and the lesson is remembered longer than the visit.

A trademark is a different thing again, and the difference is worth stating at the outset because the two words are used interchangeably in ordinary speech. A trademark is a legal right over a sign used to distinguish the goods or services of one business from those of another, and it is registered, renewed and defended according to rules that have nothing to do with how people feel. Chapter 3 treats naming and the trademark search in full, and section 1.7 of this chapter sets out the statutes that apply. A business can hold a registered trademark over a name that nobody remembers, and it can be remembered warmly under a name it never registered. The two are related, and they are not the same.

What a brand is not can be stated just as plainly, and the list is short. A brand is not a logo, although a logo is one of the places where a brand becomes visible. A brand is not a slogan, although a good tagline can carry a promise in six words. A brand is not an advertising campaign, although a campaign is one of the ways a brand is put in front of people. Each of those three is a material, a tool or an occasion. The brand is the residue they leave behind, and a business that mistakes the material for the thing itself will spend its money on the material and wonder why nothing changed.

DEFINITION Brand, brand identity, brand image and brand equity

A brand is the gut feeling a person holds about a product, a service or a company, formed from everything that person has seen, heard and experienced (Neumeier, 2005). Brand identity is what the business aspires the brand to be: the meanings, the values and the personality the strategist wants the brand to stand for (D. A. Aaker, 1996). Brand image is what people actually hold, which is the result of their experience and of everything the business has shown them, and which the business influences but does not control. Brand equity is the value that attaches to the brand's name and symbol, treated as a set of assets of the business (D. A. Aaker, 1991); in its customer-based form it is the differential effect of brand knowledge on how customers respond to the marketing of the brand, and brand knowledge has two components, brand awareness and brand image (Keller, 1993). A trademark is the legal right over the sign, and it is a separate matter from all four.

WATCH OUT Starting with the logo

The most common way to begin a branding project is also the most expensive way to begin it, and the mistake is to start with the logo. A logo is a container. It holds whatever meaning the business has put into it, and it holds nothing at all if the business has not decided what it stands for, what it promises and how it speaks. A business that commissions a mark before it has settled those questions will receive a competent piece of design and will then discover that it cannot say what the mark means, cannot brief a printer or a sign maker on how to use it, and cannot tell whether a proposed change is an improvement or a loss. The order that works is the order this book follows: the platform first, then the name and the voice, then the visual identity, then the system that holds it together. A logo designed in that order takes less time to approve, because the arguments were settled before the drawing began.

1.2 Brand, Identity and Image

Three layers sit between a business and the person it serves, and confusing them is the source of most of the frustration in branding work. The first layer is the platform, which is what the business decides: its purpose, its promise, its values and its personality. The second layer is the identity, which is what the business shows: the name, the words, the logo, the colors, the typefaces and the templates. The third layer is the image, which is what people remember. The business controls the first two directly and the third only indirectly, through the touchpoints where the other two meet a real person.

The distinction between identity and position belongs to the strategy literature and is worth keeping precisely. Brand identity is what the strategist aspires the brand to be, and brand position is the part of the identity that is actively communicated (D. A. Aaker, 1996). The difference is one of scope and of audience. A business may hold a rich identity that includes the way it treats its technicians, the way it keeps its records and the way it answers the phone at seven in the morning, and it will communicate only a part of that to the market, because a market can hold only so much. The position is the part that is said out loud, in a proposal, on a website, on the side of a van. The identity is the part that must be true for the position to survive contact with a customer.

The same author offers four perspectives from which a brand can be looked at, and they are useful as a checklist rather than as a theory. A brand can be considered as a product, as an organization, as a person and as a symbol (D. A. Aaker, 1996). The product perspective asks what the offer does and how well it does it. The organizational perspective asks what kind of business stands behind it, which matters enormously to a customer choosing a contractor who will be in their building. The personal perspective asks what character the brand has, which is where the language of personality comes in. The symbolic perspective asks what the brand's marks and images carry, which is where the visual identity lives. A business that can answer all four has a brand with depth; a business that can answer only the first has a product.

The third layer, the image, is the one that decides whether any of the work paid. Keller's account of customer-based brand equity places the whole question in the customer's response: equity is the differential effect of brand knowledge on how consumers respond to the marketing of the brand, and it is positive when customers respond more favorably to the branded offer than they would to an unbranded one (Keller, 1993). The same account describes brand knowledge as having two components, brand awareness and brand image, and it says that equity arises from familiarity with the brand and from favorable, strong and unique associations (Keller, 1993). Read against the three layers, that is a precise statement of the business's job: build awareness, build the right associations, and let the response follow.

Figure 1.1 draws the three layers and the traffic between them. The figure shows three panels from left to right. The left panel is the platform, holding what the business promises. The middle panel is the identity, holding what the business shows. The right panel is the image, holding what people remember. Arrows run from the platform to the identity and from the identity through a band labeled touchpoints to the image, and a return arrow runs from the image back to the platform, labeled as measured in Chapter 9. Below the three panels, a strip shows Nordec's inventory before the refresh: thirty-six touchpoints, three logo versions, five blues and four typefaces, with thirteen of the thirty-six agreeing with one another.

Figure 1.1 — A brand in three layers. The business decides the platform and the identity; people form the image at every touchpoint, and the image is the only layer the business does not control directly.

The return arrow is the part of the figure that a small business most often leaves out. The image is not a verdict to be accepted; it is information to be read. A business that never asks what people remember is working from the platform and the identity alone, which means it is designing in a room with the door closed. The measurement does not have to be elaborate, and Chapter 9 sets out what a small business can do with a telephone and sixty names. What matters at this point in the book is that the loop exists, that it is drawn, and that the business intends to close it.

KEY CONCEPT The brand lives in people's heads, not in the logo file

The central idea of this book is that the brand is held by other people, and that the business's materials are only the means by which it is placed there. This is why two businesses can buy the same design work and end up with different brands, and why a business can improve its brand without changing a single visual element. It is also why the work in the chapters that follow is ordered the way it is. The platform comes first because a business cannot be consistent about something it has not decided. The verbal and visual identities come next because they are the instruments of consistency. The system and the guidelines come after that because consistency at scale needs a container. The touchpoints come next because that is where the image is actually formed. And the measurement comes at the end because the only test of the whole exercise is what a customer says when somebody asks them who did the work on their building.

1.3 The Parts of a Brand

A brand has four working parts, and each of them receives a chapter of its own in this book. The first is the platform: the purpose the business serves, the promise it makes, the values it holds and the personality it shows. The second is the verbal identity: the name, the tagline, the voice and the words the business uses and avoids. The third is the visual identity: the logo, the color palette, the typography and the imagery. The fourth is the experience: everything that happens at the touchpoints where a customer meets the business, from the first search result to the invoice that arrives after the work is done.

The platform is the part that most small businesses skip, and it is the part that makes the other three possible. A purpose is a sentence about what the business is for, written so that it can be used to decide something. A promise is what the customer can expect every time, stated in words a customer would recognize. Values are the few principles that change decisions, and a value that does not change a decision is a poster rather than a value. Personality is the character the brand shows, described in a handful of traits that a writer can use. Chapter 2 builds all four with the team, in workshops rather than in a document written alone.

The verbal identity is the part that travels furthest, because words go everywhere that a logo cannot. A name has to be pronounceable, memorable and legally available, and the research on suggestiveness gives a business one useful warning: a name that suggests a benefit improves recall of claims consistent with that benefit but lowers recall of claims about other benefits (Keller et al., 1998). A tagline carries the promise in a few words. A voice is a small set of rules about how the business speaks, and a glossary of the words to use and to avoid is what makes the voice repeatable by seven people rather than by one. Chapter 3 treats naming, the trademark search, the tagline, the voice and the glossary.

The visual identity is the part that people recognize before they read anything. A logo, a palette, a pair of typefaces and a way of handling images are the four elements, and the research on logos gives a business something concrete to aim at. In a study of 195 logos rated on 13 design characteristics, logos meant for high recognition were found to be very natural, very harmonious and moderately elaborate, while logos with a lower investment behind them should be less natural and very harmonious, and a professional look calls for moderate elaboration and natural elements (Henderson & Cote, 1998). The finding is useful precisely because it is modest: it does not promise that a good mark will sell anything, and it does say what kind of mark tends to be recognized. Chapter 4 treats the visual identity, and Chapter 5 turns it into a system of tokens, templates and a guide.

The experience is the part that decides whether the other three are believed. A brand is formed at every point where a person meets the business, and those points are more numerous than most owners expect. A preventive visit, a monthly report, a van in traffic, a uniform in a lobby, a proposal in an inbox, an invoice, a review left online, a receptionist's greeting and a search result are all touchpoints, and each of them either confirms the promise or contradicts it. Chapter 7 treats the touchpoints in full, and the inventory described in this chapter is the raw material for that work.

The following table sets out the parts of a brand, what each one contains, who decides it in a small business and the chapter that treats it. Read it as a map of the book rather than as a hierarchy: the parts are listed in the order in which they are built, and each row names the person who holds the decision in a business of a handful of people.

Table 1.1 — The parts of a brand and where this book treats them

METHOD A first inventory of your touchpoints

First, list every place a customer meets the business, walking through the year in order: the search, the referral, the first call, the proposal, the first visit, the visit sheet, the monthly report, the invoice, the van, the uniform, the sign, the website, the review page. Second, collect one sample of each: a photograph of the van and the sign, a saved copy of the report and the invoice, a screenshot of the website and the email signature, a photograph of a uniform. Third, note against each sample which logo version it carries, which colors appear, which typefaces are used and what tone the words take. Fourth, mark each touchpoint as consistent or not, using one question: would a stranger know that this came from the same business as the last one? Fifth, count the total and the consistent ones, write the percentage down, and keep the list, because it is the working document for Chapter 7. The whole exercise takes an afternoon and produces the number that starts the work.

CASE IN POINT Nordec Entretien inc.: thirty-six touchpoints, three logos

Nordec Entretien inc. is a fictional business created for this encyclopedia; its owner, its employees, its customers, its competitors, its figures and the events described are invented, and any resemblance to a real business is coincidental. Sophie Nadeau founded Nordec in 2022 and had a friend draw the first logo on a template site: a blue wrench inside a circle, with the name in capitals. Four years later, an inventory of the business's thirty-six touchpoints found three versions of that logo still in circulation, five different blues, four typefaces, and only thirteen of the thirty-six touchpoints agreeing with one another, which is 36.1%. The new service tiers and the fixed-price assessment had no names in French, no templates and no identity of their own. In the six months before the inventory, fifteen proposals were decided and eight were signed, which is 53.3%. Asked how they had first heard of Nordec, prospects answered a referral by a customer six times, another building of a current manager three times, the assessment letter three times, a web search twice and a Nordec van once, and five of the fifteen knew Nordec before the first contact.

1.4 The Landscape in 2026

A small business in 2026 builds its brand in a landscape that is wider and faster than the one its owner learned in. The number of places where a customer can meet a business has grown, and most of the growth is digital. A search result, a map listing, a review page, a social profile, an email signature, a proposal sent as a file, a monthly report sent as a file, a website that works on a telephone and a van parked on a street are all touchpoints, and a business of seven people can hold all of them at once. The consequence is not that branding has become harder in principle; it is that inconsistency has become easier, because there are more surfaces on which to be inconsistent.

One part of the landscape has become a requirement rather than a preference, and that is accessibility. The Web Content Accessibility Guidelines 2.2 are a W3C Recommendation, first published as a Recommendation on October 5, 2023 and issued in their 2.2 form on December 12, 2024 (World Wide Web Consortium, 2024). The guidelines set success criteria that a brand's materials must meet if they are to be readable by everyone. Success criterion 1.4.3, at level AA, requires a contrast ratio of at least 4.5:1 for normal text and at least 3:1 for large text, defined as 18 points or 14 points bold; text that is part of a logo or a brand name has no contrast requirement. Success criterion 1.4.6, at level AAA, raises those to 7:1 and 4.5:1 for large text, and success criterion 1.4.11, at level AA, requires graphical objects and interface components needed to understand content to reach at least 3:1 against adjacent colors (World Wide Web Consortium, 2024). For a brand, this means that a palette is not only a matter of taste: a color chosen for a heading has to carry text at a stated ratio, and a color chosen for a button has to stand out from what surrounds it. No legal obligation in Canada or Québec is claimed here, and the practical point stands on its own.

A second change is technical and quiet, and it affects how a visual identity is stored and repeated. On October 28, 2025, the Design Tokens Community Group published version 2025.10, the first stable version of the design tokens specification (Design Tokens Community Group, 2025). The document is a W3C Community Group report and not a W3C Standard, and it is worth saying so plainly. What it offers a business is a way of naming the small decisions of a visual identity — a color, a size, a spacing, a typeface choice — so that one file of named values can produce code for several platforms, with support for theming and multi-brand work, modern color spaces and aliases between tokens (Design Tokens Community Group, 2025). For a small business the practical effect is that the palette and the type scale can be written down once and applied everywhere, rather than being re-entered by hand in each document.

A third change concerns images. Generative image tools are now a family of tools that a small business may encounter, and they raise a question of rights rather than a question of taste. Chapter 12 treats the functions such tools offer and the questions to ask about the rights attached to what they produce, and this chapter only notes that the question exists. A business that uses an image in a proposal, on a van or on a website is making a claim about where that image came from, and the answer has to be written down somewhere.

In Québec, the landscape changed on a specific date. Rules in force since June 1, 2025, from the amended Regulation respecting the language of commerce and business published on June 26, 2024, govern how a trademark appears on a sign visible from outside a premises (Office québécois de la langue française, 2025a). The Office québécois de la langue française explains that a recognized trademark wholly or partly in another language, and a business name with a non-French specific element, must be accompanied by French terms — a generic term, a description or a slogan — so that French is markedly predominant, which means a much greater visual impact, at least twice the space, permanently, and at least as visible and legible; if a French version of the trademark is registered, it must be displayed (Office québécois de la langue française, 2025a). A separate page explains the rules for trademarks on products, in force since the same date, under which the generic or descriptive words inside a trademark on a product must appear in French while product and company names may stay in another language, with some products made before June 1, 2025 still sellable until June 1, 2027 (Office québécois de la langue française, 2025b). This book explains these rules and does not give legal advice.

Finally, the landscape includes a school of thought that has changed how marketers talk about being remembered. The work that stresses mental and physical availability argues from empirical regularities in buying data that a brand grows by being easy to notice and easy to buy (Sharp, 2010). The related concept of brand salience is defined as the propensity of a brand to be noticed or to come to mind in buying situations, reflecting the quantity and quality of memory associations, and it is treated as distinct from awareness and from attitude (Romaniuk & Sharp, 2004). For a small business, the practical translation is direct: the business that comes to mind first when a property manager has a problem is the business that gets the call, and coming to mind is a matter of being seen often enough, in the same recognizable way, in the places where the buyer already looks.

WHAT CHANGED Design tokens get a stable format

On October 28, 2025, the Design Tokens Community Group published version 2025.10 of the design tokens specification, the first stable version of the work (Design Tokens Community Group, 2025). The document is a W3C Community Group report and not a W3C Standard, and it should be described that way. What changed for a small business is smaller than it sounds and more useful than it looks. Before a stable version, a business that wanted its colors and type sizes to travel from a document to a website to an application had to keep the values in several places and hope they stayed equal. With a stable format, the values can be named once and referenced elsewhere, with support for theming and multi-brand work, modern color spaces and aliases between tokens, and one token file producing code for several platforms (Design Tokens Community Group, 2025). The practical consequence for the work of Chapter 5 is that a palette can be written down as a set of named values rather than as a page of instructions, and a named value is harder to get wrong than an instruction.

QUÉBEC & CANADA NOTE French on signs since June 1, 2025

In Québec, a sign visible from outside a premises is governed by rules in force since June 1, 2025, from the amended Regulation respecting the language of commerce and business published on June 26, 2024. The Office québécois de la langue française explains that a recognized trademark wholly or partly in another language, and a business name with a non-French specific element, must be accompanied by French terms — a generic term, a description or a slogan — so that French is markedly predominant (Office québécois de la langue française, 2025a). Markedly predominant is defined in visual terms: a much greater visual impact, at least twice the space, permanently, and at least as visible and legible. If a French version of the trademark is registered, that version must be displayed. For a business with a depot sign, a van or a storefront, the practical work is to decide what French words accompany the mark, to give them at least twice the space, and to keep them there rather than treating them as a temporary addition. This book explains the rules and does not give legal advice; the Office's own pages and a lawyer are the right sources for a particular sign.

1.5 What Is at Stake for a Small Business

The stake is not that a brand makes a business look better. The stake is that a brand changes how people respond to the same offer, and that change can be measured in the ordinary numbers of a small business. Keller's account of customer-based brand equity defines it as the differential effect of brand knowledge on how consumers respond to the marketing of the brand, and it says that equity is positive when customers respond more favorably to the branded offer (Keller, 1993). The same account locates the source of that response in familiarity with the brand and in favorable, strong and unique associations (Keller, 1993). Translated into the life of a business of seven people, the sentence reads: the same technician, the same ninety-minute response and the same monthly report will be received differently by a customer who knows the name than by one who does not.

That difference shows up in three places that a small business can watch. The first is the referral. A customer who can describe what the business does, in words the business itself supplied, is a customer who can pass the name on, and the referral arrives with the description attached. The second is the proposal that starts warm. A prospect who has already seen the van, the report or the review page reads the same document differently from a prospect who has never heard the name, and the difference is visible in how many questions are about the work rather than about the business. The third is the price conversation. A business that is known for something specific is not compared on price alone, because the buyer is comparing two different things rather than two prices.

The scale of the population in which this happens is worth stating, because it explains why the book is written for small businesses rather than for large ones. In December 2024, Canada had about 1.1 million employer businesses, of which 98.2% were small, meaning 1 to 99 employees, and 99.7% were small or medium, meaning 1 to 499 employees; in 2024, small businesses employed 5.8 million people, 46.6% of the private-sector workforce, and produced 34.2% of private-sector GDP on average in 2018–2022 (Innovation, Science and Economic Development Canada, 2026). Almost every business that will read this book is in the small band, and the small band is where the constraint described below bites hardest.

The constraint is money, and it shapes the whole method. A small business cannot buy awareness with advertising, because the cost of reaching enough people to be remembered is beyond what a business of seven people can spend. What it can do is use the exposures it already pays for. Every preventive visit, every monthly report, every van on the road and every uniform in a lobby is an exposure that the business is already funding through its ordinary operations, and the only question is whether those exposures carry a recognizable identity or a different one each time. This is the reason the book spends its effort on consistency rather than on reach, and the reason the case's rollout budget is spent on wraps, uniforms, templates and a sign rather than on media.

The cost of inconsistency is the mirror image, and it is easier to see in a list than in an argument. A business with three logo versions pays for the confusion three times: once when a customer fails to connect two encounters, once when a prospect cannot confirm that the business they were told about is the business in front of them, and once when the business itself cannot tell which of its materials is current. None of those costs appears on an invoice, and all of them are paid in the currency that matters most to a small business, which is the attention of a buyer who has four other names to consider.

DO THE MATH How consistent was Nordec?

Illustrative figures; the case is fictional. Step 1 — The inventory listed 36 touchpoints where a customer meets the business, from the search result to the invoice. Step 2 — Of those, 13 were consistent with one another, so 13 ÷ 36 = 36.1% of the touchpoints agreed, and 23 did not. Step 3 — The same inventory found 3 versions of the logo, 5 different blues and 4 typefaces in circulation, which means a customer meeting three touchpoints could meet three different businesses. Step 4 — Of the 15 prospects whose proposals were decided in the six months before the inventory, 5 knew Nordec before the first contact, which is 5 ÷ 15 = 33.3%. Step 5 — Of those 15, 1 named a Nordec van as the way they had first heard of the business, which is 6.7%. The van is the most visible touchpoint the business owns, seen on the road every working day, and it produced the fewest first contacts of any source named, because a van that is not recognizable is a van that is not remembered.

WHO DOES THIS The owner owns the brand

In a small business, the brand belongs to the owner, and this is not a statement about legal ownership. It is a statement about who can decide. Only the owner can settle what the business promises, because only the owner can commit the business to keeping it. Only the owner can decide that the ninety-minute commitment comes before a more profitable job, or that no invoice goes out for work the customer has not heard about. Only the owner can approve a change to the name, the promise or the identifiers, and only the owner can decide when a designer's proposal is right. What the owner does not have to do is the work of maintaining the brand. The templates, the asset folder, the checks on new materials and the inventory of touchpoints can all be held by somebody else, and in the case followed through this book they are held by the administrator, who becomes the brand keeper. The division is simple: the owner decides what the brand means, and the brand keeper makes sure that what goes out matches it.

1.6 The Vocabulary of Branding

The words of branding are used loosely in ordinary conversation, and the looseness costs a business real time, because two people can agree on a sentence while meaning two different things by it. This section fixes the twelve terms the rest of the book uses, and it fixes them in the sense in which this book uses them rather than in the sense of any particular school. Where a term has a legal meaning as well, the legal meaning is noted and the chapter that treats it is named.

A brand is the gut feeling a person holds about a product, a service or a company (Neumeier, 2005). Brand equity is the value that attaches to the brand's name and symbol, treated as a set of assets of the business (D. A. Aaker, 1991), and in its customer-based form it is the differential effect of brand knowledge on how customers respond to the marketing of the brand (Keller, 1993). Brand identity is what the business aspires the brand to be, and brand position is the part of that identity which is actively communicated (D. A. Aaker, 1996). Brand image is what people actually hold. Positioning, in the sense used by Book A3 (Strategy and Competitive Positioning), is the place the business chooses to occupy in the market relative to its competitors, and the brand is how that place is made visible and memorable.

A brand platform is the set of decisions that hold the brand together: purpose, promise, values and personality. A promise is the sentence that states what a customer can expect every time, and it is the part of the platform that a customer would recognize if they heard it. A touchpoint is any place where a person meets the business, whether that meeting is physical, digital or human. A distinctive brand asset is an element — a logo, a color, a typeface, a character, a tagline or a jingle — that a buyer associates with the brand and that helps the brand be noticed and recalled; the field is researched, and the work examines which elements succeed and which fail over the long term (Romaniuk, 2018). Brand architecture is the organizing structure of the brand portfolio, specifying the roles of the brands in it and the relationships between them (Aaker & Joachimsthaler, 2000). Brand guidelines are the document that records the rules of the identity and shows them in use. A trademark is the legal right over the sign, and it is registered for goods and services.

The following table sets out the twelve terms, the meaning each one carries in this book and the chapter that treats it. Read it as a glossary to return to rather than as a passage to memorize: the terms are used consistently from here on, and a reader who keeps the table open will find that the later chapters do not shift their meanings.

Table 1.2 — The vocabulary of branding

TOOLS OF THE TRADE What a first brand inventory needs

A first inventory of touchpoints needs four ordinary things and no specialist product. The first is a spreadsheet or a table with one row per touchpoint and columns for the place, the sample collected, the logo version, the colors, the typefaces, the tone and a consistent-or-not mark. The second is a camera or a telephone with a camera, because half the touchpoints are physical: the van, the uniform, the depot sign, the sticker on a panel, the folder handed to a customer. The third is a shared folder where the samples are kept in a fixed order, so that the next person can open the folder and see the same set of materials the first person saw. The fourth is a checklist of the places to look, written once and reused, so that the second inventory a year later covers the same ground as the first and the two can be compared. The functions are what matter: a place to record, a way to capture, a place to keep and a list to follow. No brand name belongs in this list, and no price.

1.7 The Rules That Apply

A brand is made of words, images and promises, and each of those three is touched by law. This section sets out the statutes that apply in Canada, in Québec and in the United States, in the terms in which they were read for this book. It explains the law and does not replace the advice of a lawyer or a trademark agent, and a business that is about to register a name, sign a design contract or put a claim in front of the public should take advice on its own facts.

The Trademarks Act, R.S.C. 1985, c. T-13, governs the registration of a mark in Canada. Section 30(1) provides that a person may apply to register a trademark for goods or services if they use it or propose to use it and are entitled to use it in Canada; section 30(2)(a) requires that the goods or services be stated in ordinary commercial terms; and section 30(3) requires that they be grouped by class of the Nice Classification, each group preceded by its class number. Section 46(1) provides that a registration lasts ten years from the date of registration and is renewable for further periods of ten years. In general terms, the owner of a registered trademark controls its use for the goods and services of the registration. A trademark agent assesses registrability and the risk of confusion, and that assessment is a different exercise from the naming work of Chapter 3, which produces candidates for the agent to search.

The Copyright Act, R.S.C. 1985, c. C-42, governs who owns a logo and the other creative work a business commissions. Section 13(1) provides that the author of a work is its first owner, and section 13(3) provides that the employer is the first owner of a work made by an employee in the course of employment, unless the parties agree otherwise. Section 13(4) provides that an assignment or grant of copyright is valid only if it is in writing and signed by the owner of the right. Section 14.1 provides that moral rights — the integrity of the work and the right to be associated with it — cannot be assigned but can be waived, in whole or in part, and that an assignment of copyright does not by itself waive them. The practical consequence for a business commissioning a logo from an independent studio is that the contract must assign the copyright in writing and should address the waiver of moral rights, because neither happens by default.

The Charter of the French language, CQLR c C-11, requires French in commerce and business in Québec, including commercial documents, signs and product inscriptions. This book states its rules only through the explanations of the Office québécois de la langue française and of the Registraire des entreprises, and it does not cite an article number of the Charter or of its regulation. The rules in force since June 1, 2025, on signs visible from outside a premises and on trademarks on products are set out in section 1.4 of this chapter, from the Office's own pages (Office québécois de la langue française, 2025a; Office québécois de la langue française, 2025b). Book H4 (Consumer Protection, Advertising Law, Competition and Language Requirements) treats the language requirements in full.

The Competition Act, R.S.C. 1985, c. C-34, governs what a business may say to the public. Section 52 makes it a criminal offense to knowingly or recklessly make a representation to the public that is false or misleading in a material respect, and section 74.01(1)(a) makes the same conduct reviewable conduct on the civil side; the provision applies to sales to businesses as well as to consumers. A brand promise or a tagline made to the public is a representation, and a promise the business cannot keep is a representation that is misleading in a material respect. Book A6 (Designing a Value-Added Offer) treats the offer side of this question, and Book H4 treats the advertising side.

In the United States, the Lanham Act, 15 U.S.C. § 1051, governs federal trademark applications. Section 1051(a) provides for an application by the owner of a mark used in commerce, and section 1051(b) provides for an application on a bona fide intention to use the mark. Section 1051(d) provides that, after a notice of allowance, a statement of use is due within six months, extendable by six months and then by further extensions totalling no more than 24 months for good cause. A Canadian business that sells or intends to sell in the United States meets this statute as well as its Canadian counterpart, and the two filings are separate.

THE RULE A trademark is registered for goods or services

The Trademarks Act, R.S.C. 1985, c. T-13, s. 30(1), provides that a person may apply to register a trademark for goods or services if they use it or propose to use it and are entitled to use it in Canada. Section 30(2)(a) requires the goods or services to be stated in ordinary commercial terms, and section 30(3) requires them to be grouped by class of the Nice Classification, each group preceded by its class number. Section 46(1) provides that a registration lasts ten years from the date of registration and is renewable for further periods of ten years. Two consequences follow for the work of this book. A trademark is registered for particular goods and services rather than for a name in general, so a business must decide what it is registering the mark for. And a registration is a renewable right rather than a permanent one, so the renewal date belongs in the same calendar as every other recurring obligation. This book explains the law and does not replace the advice of a lawyer or a trademark agent.

1.8 The Case and the Plan of This Book

The business followed through this book is Nordec Entretien inc., a fictional company created for this encyclopedia. It provides preventive maintenance and on-call technical service to commercial buildings on Montréal's north shore: light heating and ventilation work, minor plumbing, service electrical and loading doors. Sophie Nadeau founded it in 2022 after fourteen years as a building systems technician. The business has seven people — Sophie, four technicians, one of them the lead technician, a dispatcher and an administrator — and two vehicles. Its customers are owner-operators of buildings and property managers who run several buildings for their owners, and its competitors are the independent trades, billed by the call, and the regional integrator, a large facility-services firm that sells cleaning, security and maintenance together.

The strategy was settled in Book A3 (Strategy and Competitive Positioning): two territories, mid-size buildings of 2,000 to 9,000 square metres, owner-operators and small portfolios, with refusals of one-off emergency calls for buildings without a contract, of new buildings outside the two territories and of bundled tenders. The differentiators are the ninety-minute response to urgent contract calls, day and night; the building health record kept for every contract building; the one-page monthly report; and the capital planning note, a yearly note listing the equipment approaching the end of its life with a replacement window. The offer was settled in Book A6 (Designing a Value-Added Offer): three tiers, two add-ons, the ninety-minute guarantee and the Building Check, a fixed-price assessment for buildings not under contract.

This book begins six months after that offer refresh. Nordec has 52 buildings under contract, revenue of $37,251 a month and a run-rate of $447,012 a year. In the six months since the refresh, 138 urgent calls were received and 5 were missed, and 11 Building Checks were sold, of which 4 converted into contracts. The trigger for the brand work was an inventory of the business's 36 touchpoints, which found 3 versions of the logo, 5 different blues, 4 typefaces and only 13 of the 36 touchpoints consistent with one another. The new tiers and the Building Check had no names in French, no templates and no identity. The regional integrator's trucks, uniforms and reports look the same everywhere, and Nordec's do not.

The decision that follows from that inventory is the subject of the remaining chapters, and the plan is worth stating in full. Chapter 2 builds the brand platform in two half-day workshops with the seven people: the purpose, the promise, the three values and the four personality traits. Chapter 3 settles the names and the voice, including the French and English names of the tiers and of the Building Check, the tagline and the glossary of words to use and to avoid. Chapter 4 builds the visual identity with an independent design studio: the logo, the palette, the two typefaces and the eleven templates. Chapter 5 turns that work into a system of design tokens, templates and a guide, and names the brand keeper. Chapter 6 settles the brand architecture, which for Nordec is a branded house. Chapter 7 works through the touchpoints, from the 624 preventive visits a year to the vans on the road. Chapter 8 plans the refresh as an evolution rather than a revolution. Chapter 9 sets out how the brand is measured, and Chapter 10 to Chapter 13 cover the rhythm of the work, the roles and the help a small business may need, the functions a tool must offer, and a sixty-day plan.

IN LA MACHINE What the platform holds of a brand today

Two pages of the application copy read for this book touch the brand directly, and it is worth being plain about what each one does (La Machine, 2026). The Branding page at business/branding offers a logo drop zone, a brand name, a tagline and a website address, a primary and a secondary color with ten preset swatches, a choice among five fonts, an email footer and three social links. It computes the contrast of the primary color and shows a pass or fail badge at the 4.5:1 threshold, with a live preview of a client invitation card and a client email. Its values are not saved: the save button shows a confirmation and no other page uses what was entered. The page is a place to try a palette and see its contrast, not a brand register. The My Website page at business/website holds the site's own theme and branding panel, with a gallery of themes, a brand name, a tagline, a logo, a favicon and a banner image, five color overrides and a heading and body font, kept in the browser. La Machine has no saved brand kit that other pages use, no brand portal or guidelines publisher, and no brand consistency audit.

Costly Mistakes

• Starting with the logo. A mark commissioned before the platform is settled produces a competent drawing that the business cannot explain, cannot brief others on and cannot judge, and the arguments that were skipped return at every approval.
• Treating the brand as a design project rather than a business decision. A brand is decided by the owner, because only the owner can commit the business to keeping the promise, and a designer cannot make that decision on the business's behalf.
• Assuming the brand is what the business says it is. The image is held by other people and formed at every touchpoint, so a business that never asks what customers remember is designing with the door closed.
• Letting the touchpoints multiply without a count. Three logo versions, five blues and four typefaces accumulate one reasonable decision at a time, and none of them is visible until somebody lists all thirty-six places a customer meets the business.
• Choosing a palette without checking contrast. A color that looks well on a screen may fail the 4.5:1 ratio required for normal text, and the failure is discovered by a reader rather than by the designer.
• Commissioning creative work without a written assignment of copyright. Under the Copyright Act, an assignment is valid only if it is in writing and signed, and moral rights are not waived by an assignment of copyright on its own.
• Making a promise the business cannot keep. A brand promise is a representation to the public, and a representation that is false or misleading in a material respect is addressed by the Competition Act on both the criminal and the civil side.

Key Takeaways

• A brand is a person's gut feeling about a product, a service or a company, and it is not a logo, a slogan or an advertising campaign (Neumeier, 2005).
• A brand has three layers: the platform the business decides, the identity it shows and the image people form; the business controls the first two directly and the third only through its touchpoints.
• Brand identity is what the business aspires the brand to be and brand position is the part of it actively communicated (D. A. Aaker, 1996), while customer-based brand equity is the differential effect of brand knowledge on how customers respond to the same offer (Keller, 1993).
• A small business cannot buy awareness with advertising, so its brand is carried by the exposures it already pays for: its people, its documents, its vehicles and its service.
• The landscape of 2026 adds touchpoints, adds accessibility requirements for text and interface elements (World Wide Web Consortium, 2024), gives design tokens a stable format (Design Tokens Community Group, 2025), and in Québec adds rules on French on signs in force since June 1, 2025 (Office québécois de la langue française, 2025a).
• Nordec's inventory found 36 touchpoints, 3 logo versions, 5 blues, 4 typefaces and 13 of 36 touchpoints consistent with one another, which is 36.1%, and that number is where the work of this book begins.

Your Action Plan

1. Write your own definition of the brand in one sentence. Describe what you want a customer to feel about the business, in words a customer would recognize, and keep the sentence to one line. Deliverable: one written sentence. Verification: the sentence names a feeling or an expectation rather than a visual element, and it does not contain the word logo.
2. List every touchpoint where a customer meets the business. Walk through a year in order, from the search and the referral to the invoice and the review page, and write each place on its own row. Deliverable: a numbered list of touchpoints. Verification: the list includes at least one physical touchpoint, one document and one digital touchpoint.
3. Collect one sample of each and mark it consistent or not. Photograph the van, the sign and a uniform, save a copy of the report and the invoice, and take a screenshot of the website and the email signature; then mark each sample against the question of whether a stranger would know it came from the same business. Deliverable: a folder of samples with a consistent-or-not mark on each. Verification: every row of the list has a sample or a written note explaining why none exists.
4. Count the total and the consistent ones, and write the percentage down. Divide the consistent touchpoints by the total and record the figure beside the date, so that the same measurement can be repeated later. Deliverable: one percentage with its date. Verification: the arithmetic is shown, and the list is kept for Chapter 7.
5. Check one color against the contrast requirement before the next document goes out. Take the color used for headings and for body text and confirm the ratio against the background, using the thresholds of 4.5:1 for normal text and 3:1 for large text. Deliverable: a recorded ratio for the colors in use. Verification: the ratio is written down with the colors it applies to, and any color that fails is marked for replacement.

In La Machine

Two pages of the application copy read for this book touch the brand, and neither of them is a brand register (La Machine, 2026). The Branding page at business/branding takes a logo file, a brand name, a tagline, a website address, a primary and a secondary color chosen from ten preset swatches, one of five fonts, an email footer and three social links; it computes the contrast of the primary color and shows a pass or fail badge at the 4.5:1 threshold, with a live preview of a client invitation card and a client email. The page is a place to try a palette and see its contrast, and that is a genuinely useful thing to do before a document goes out. What it is not is a place where a brand is kept: its values are not saved, and no other page uses what was entered. The My Website page at business/website, with its builder and preview routes, holds the site's own theme and branding panel — a gallery of themes, a brand name, a tagline, a logo, a favicon and a banner image, five color overrides and a heading and body font — and that configuration is kept in the browser.

What the platform does not have belongs in the same breath, because a business should not plan around a function that is not there. La Machine has no saved brand kit that other pages use, no brand portal or guidelines publisher, no digital asset management with versions, usage rights or expiry dates, and no brand consistency audit. Its other pages do not read a stored palette or logo: the visual editor, the image library and the montage pages each work from their own materials, and the message templates hold no brand colors, logo or footer. The one place where the platform applies a stored identity is the styles page used by its OEM partners for white-label sites, and that is not available to an ordinary account. The practical conclusion for this chapter is that the inventory of touchpoints and the record of the brand live outside the platform, and the platform is where some of the materials are made.

Self-Check

1. What is a brand, according to the definition this book uses, and what three things is it commonly mistaken for?
2. What are the three layers of a brand, and which of them does the business control directly?
3. What is the difference between brand identity and brand position, and what is customer-based brand equity?
4. Name four changes in the landscape of 2026 that affect how a small business builds a brand, and say what each one requires.
5. What did Nordec's inventory find, and why does the number of consistent touchpoints matter more than the number of touchpoints?

Answers. 1. A brand is a person's gut feeling about a product, a service or a company (Neumeier, 2005), and it is commonly mistaken for a logo, a slogan and an advertising campaign, each of which is a material, a tool or an occasion rather than the brand itself. 2. The three layers are the platform, which is what the business promises; the identity, which is what it shows; and the image, which is what people remember; the business controls the platform and the identity directly and the image only through its touchpoints. 3. Brand identity is what the strategist aspires the brand to be, and brand position is the part of the identity that is actively communicated (D. A. Aaker, 1996); customer-based brand equity is the differential effect of brand knowledge on how consumers respond to the marketing of the brand, with brand knowledge having two components, brand awareness and brand image (Keller, 1993). 4. The growth in the number of touchpoints, most of them digital, which makes inconsistency easier; the accessibility criteria of WCAG 2.2, which set a 4.5:1 contrast ratio for normal text and 3:1 for large text and for interface components needed to understand content (World Wide Web Consortium, 2024); the first stable version of the design tokens specification, published on October 28, 2025 as a W3C Community Group report and not a W3C Standard (Design Tokens Community Group, 2025); and, in Québec, the rules in force since June 1, 2025 on French on signs visible from outside a premises, where French must be markedly predominant and take at least twice the space (Office québécois de la langue française, 2025a). 5. The inventory found 36 touchpoints, 3 versions of the logo, 5 blues, 4 typefaces and 13 of 36 touchpoints consistent with one another, which is 36.1%; the consistent count matters because a customer meets a handful of touchpoints rather than all of them, and each inconsistent one breaks the connection to the others.

Chapter References

Aaker, D. A. (1991). Managing brand equity: Capitalizing on the value of a brand name. Free Press.

Aaker, D. A. (1996). Building strong brands. Free Press.

Aaker, D. A., & Joachimsthaler, E. (2000). The brand relationship spectrum: The key to the brand architecture challenge. California Management Review, 42(4), 8–23. https://doi.org/10.2307/41166051

Charter of the French language, CQLR c C-11 (Québec). LégisQuébec, updated to June 10, 2026 — as read for Book C2 (Websites, Landing Pages and Conversion); its rules on trademarks, signs, products and business names are stated in this book through the explanations of the Office québécois de la langue française and of the Registraire des entreprises.

Competition Act, R.S.C. 1985, c. C-34, ss. 52 and 74.01 (Canada). Justice Laws Website, current to July 21, 2026 (last amended March 26, 2026) — as read for Book A6 (Designing a Value-Added Offer) on September 22, 2026.

Copyright Act, R.S.C. 1985, c. C-42, ss. 13 and 14.1 (Canada). Justice Laws Website, current to September 3, 2026 (last amended November 7, 2024) — read September 25, 2026.

Design Tokens Community Group. (2025, October 28). Design tokens specification reaches first stable version. World Wide Web Consortium. Retrieved September 25, 2026, from https://www.w3.org/community/design-tokens/2025/10/28/design-tokens-specification-reaches-first-stable-version/

Henderson, P. W., & Cote, J. A. (1998). Guidelines for selecting or modifying logos. Journal of Marketing, 62(2), 14–30. https://doi.org/10.1177/002224299806200202

Innovation, Science and Economic Development Canada. (2026). Key small business statistics 2025 (page modified April 23, 2026). Retrieved September 25, 2026, from https://ised-isde.canada.ca/site/sme-research-statistics/en/key-small-business-statistics/key-small-business-statistics-2025

Keller, K. L. (1993). Conceptualizing, measuring, and managing customer-based brand equity. Journal of Marketing, 57(1), 1–22. https://doi.org/10.1177/002224299305700101

Keller, K. L., Heckler, S. E., & Houston, M. J. (1998). The effects of brand name suggestiveness on advertising recall. Journal of Marketing, 62(1), 48–57. https://doi.org/10.1177/002224299806200105

La Machine. (2026). Documentation — route reference inventory. Internal product documentation, read September 25, 2026, together with the code of the pages this book describes: business/branding and business/website.

Lanham Act, 15 U.S.C. § 1051 (United States). Legal Information Institute, Cornell Law School — read September 25, 2026.

Neumeier, M. (2005). The brand gap (Rev. ed.). New Riders.

Office québécois de la langue française. (2025a). Affichage des marques de commerce et des noms d'entreprises (updated August 27, 2025). Retrieved September 25, 2026, from https://www.oqlf.gouv.qc.ca/francisation/entreprises/affichage-marques-noms.html

Office québécois de la langue française. (2025b). Marque de commerce sur les produits (updated August 27, 2025). Retrieved September 25, 2026, from https://www.oqlf.gouv.qc.ca/francisation/entreprises/marque-commerce-produits.html

Romaniuk, J. (2018). Building distinctive brand assets. Oxford University Press.

Romaniuk, J., & Sharp, B. (2004). Conceptualizing and measuring brand salience. Marketing Theory, 4(4), 327–342. https://doi.org/10.1177/1470593104047643

Sharp, B. (2010). How brands grow: What marketers don't know. Oxford University Press.

Trademarks Act, R.S.C. 1985, c. T-13, ss. 30 and 46 (Canada). Justice Laws Website, current to September 3, 2026 (last amended April 1, 2025) — read September 25, 2026.

World Wide Web Consortium. (2024). Web Content Accessibility Guidelines (WCAG) 2.2. W3C Recommendation, 12 December 2024 (first published as a Recommendation on 5 October 2023). Retrieved September 25, 2026, from https://www.w3.org/TR/WCAG22/

Chapter 2 — Brand Strategy

Brand strategy turns the positioning a business has chosen into a platform — purpose, promise, values and personality — that everyone in the business can apply, and it is written before anyone draws a logo.

What You'll Be Able to Do

• Distinguish the positioning chosen in Book A3 (Strategy and Competitive Positioning) from the brand strategy that follows it, and explain why the second cannot be skipped once the first is settled.
• Write a purpose, a mission and a vision in this book's sense, and test each one against the ordinary work of an ordinary week rather than against a website page.
• State a brand promise that a customer can check, on a call and over a year, and recognize the promises that sound warm and cannot be verified.
• Name three values that change decisions, and describe for each one the decision it changes and the way it shows to a customer.
• Choose four personality traits from the five dimensions of brand personality and write them into the voice the business uses in writing and on the phone.
• Assemble the whole platform on one page, with its proof points, so that a designer, a new employee or a proposal writer can apply it without asking what it means.

Why It Matters

A business that has chosen where it competes still has not said what it promises there. The positioning of Book A3 (Strategy and Competitive Positioning) is a decision about a market: which customers, which buildings, which services, which refusals. It is a map of the ground. What it does not contain is the sentence a technician says to a building owner at seven in the morning, the reason a prospect chooses one contractor over another when both can do the work, or the standard a dispatcher applies when two calls arrive at once. Those come from the brand platform, and a business that has not written one is not without a brand; it is without a decision about its brand, which means the decision is being made by whoever happens to be speaking.

The cost of that absence is not abstract, and it shows up in the ordinary arithmetic of selling. Nordec Entretien inc., the business followed through this book, is a fictional preventive maintenance and on-call service company on Montréal's north shore with seven people and two vehicles. In the six months before its platform was written, it decided fifteen proposals and signed eight, and when prospects were asked how they had first heard of the company, the answers were scattered across referrals, another building of a current manager, a letter, a web search and a van. Five of the fifteen knew the company before the first contact. The regional integrator it competes with looks the same everywhere — the same trucks, the same uniforms, the same reports — and Nordec did not. A platform is what makes a business look and sound like one thing rather than like seven people improvising.

There is a second reason, which is about the work that comes after this chapter. Everything in the rest of this book is built on the platform. The names and the voice of Chapter 3 are derived from the promise and the personality. The visual identity of Chapter 4 is a translation of the values into color, type and form. The design system of Chapter 5 is the platform made repeatable. The touchpoints of Chapter 7 are the platform made visible, and the measurement of Chapter 9 is the platform made checkable. A business that begins with a logo and works backwards will produce a mark that is pleasant and empty, and it will discover the emptiness at the first proposal, the first new hire and the first disagreement about a color. Writing the platform first is not a matter of taste; it is the order in which the decisions depend on one another.

In This Chapter

Section 2.1 explains how the positioning of Book A3 (Strategy and Competitive Positioning) becomes a promise, and why the identity a strategist aspires to is larger than the position actively communicated. Section 2.2 defines purpose, mission and vision in this book's words and tests each one against the daily work. Section 2.3 sets out the brand promise, the two tests it must pass, and the promise the case rejected. Section 2.4 explains why a value is useful only when it changes a decision, with the case's three values and the decision each one changes. Section 2.5 covers brand personality, the five dimensions and the archetype approach, and the case's four traits. Section 2.6 explains customer-based brand equity, the building blocks of the brand equity pyramid, salience, and the balance between brand building and short-term activation. Section 2.7 assembles the platform on one page, with its figure and its table. Section 2.8 tells the case's own story: two workshops, the customers' words, the rejected promise and the tagline in French first.

2.1 From Positioning to Promise

The positioning a business has chosen answers a question about the market, and the brand strategy answers a question about the business. Positioning says where the business competes and why it wins there: which customers it serves, which it refuses, what it does that others do not, and what it deliberately does not do. Nordec's positioning, settled in Book A3 (Strategy and Competitive Positioning), is precise. The business is the preventive partner of mid-size commercial buildings — those of 2,000 to 9,000 square metres — in two territories on Montréal's north shore, serving owner-operators and property managers who run several buildings for their owners. It is faster and better informed than the independent trades billed by the call, and more attentive than the regional integrator that sells cleaning, security and maintenance together. It refuses one-off emergency calls from buildings without a contract, new buildings outside its two territories, and bundled tenders.

That is a strong position, and it is not yet a brand. Nothing in it tells a technician how to speak to an owner whose rooftop unit has failed at the start of a heat wave, or what a dispatcher should say when a contract customer calls about a leak that is not urgent but is worrying. Nothing in it tells a designer what the company should feel like on a page, or a proposal writer which sentence should come first. Positioning is a decision about where to stand; brand strategy is a decision about what to promise from there, what to stand for, how to behave and how to sound. The two are sequential, and the second is not a restatement of the first in warmer words.

The distinction that makes this concrete comes from the work on brand identity and brand position. Brand identity is what the strategist aspires the brand to be, while brand position is the part of that identity which is actively communicated (D. A. Aaker, 1996). The two are not the same thing, and treating them as the same thing is one of the most common errors in a small business. The identity is the whole of what the business intends to mean: its purpose, its promise, its values, its personality, the way it treats a customer whose equipment has failed, the way it writes an invoice. The position is the slice of that identity the business chooses to put in front of the market now, in this territory, against these competitors, with the budget it has. A business can hold an identity that is broader than its current position, and it usually should, because the position will change as the business grows while the identity is meant to hold.

The same source offers four perspectives from which a brand can be considered — as a product, as an organization, as a person and as a symbol (D. A. Aaker, 1996). The four are useful to a small business because they force a wider question than the offer alone. A brand seen only as a product is a list of services and a price. Seen as an organization, it is also the kind of employer it is, the way it answers the phone at night, the way it treats a supplier. Seen as a person, it has a temperament: calm or urgent, formal or plain. Seen as a symbol, it has a mark, a color and a name that carry memory. Nordec's platform is written across all four, which is why the promise it settles on mentions both knowledge and warning rather than only speed.

The bridge between the strategy and the creative work is the platform, and it is worth naming what the platform is for. A brand is a person's gut feeling about a product, a service or a company, and it is not a logo; the work of building one runs through a small set of disciplines that connect the strategy to the creative work rather than leaving each side to guess at the other (Neumeier, 2005). That is exactly the function of the platform in a business of seven people. It is the document that lets an independent design studio produce three routes that are all recognizably Nordec, and it is the document that lets a new dispatcher answer the phone in a way that a customer would recognize as the same company that fixed the loading door in March.

The platform has four parts, and this chapter treats them in order. The purpose is the reason the business exists beyond making money, stated so that it is true of the daily work. The promise is what the customer can count on, stated so that it can be checked. The values are the principles that change decisions, which is the only test that separates a value from a slogan. The personality is the human temperament the business presents, expressed in four traits and carried into the voice. A tagline sits at the end of the platform as a compression of the promise, not as a substitute for it, and the proof points sit beside the promise as the evidence that it is kept.

Two warnings belong at the end of this section. The first is that the platform is not a marketing document. It is a management document that happens to be useful to marketing, and its most important readers are the seven people who do the work. The second is that the platform is short. A page is enough, and a page is what a person will actually use; a platform of twelve pages will be read once, admired and forgotten. The rest of this chapter builds the page, one element at a time, and section 2.7 puts it together.

DEFINITION Purpose, promise, values, personality and platform

The brand platform is the one-page statement of what a business promises, what it stands for, how it behaves and how it sounds, written so that everyone in the business can apply it. Its four parts are distinct. The purpose is the reason the business exists beyond profit, stated so that it is true of the daily work rather than of a website. The promise is what a customer can count on, stated so that it can be checked on a specific occasion. The values are the principles that change decisions, which means that a value which has never changed a decision is a preference rather than a value. The personality is the human temperament the business presents, expressed in a small number of traits and carried into the words the business uses. The tagline compresses the promise into a line that can be remembered, and the proof points are the evidence that the promise is kept. The platform is written before the logo, because the logo is a container and the platform is what it contains.

2.2 Purpose, Mission and Vision

Three words are used loosely in business writing, and this book gives each of them a specific job. The purpose is the reason the business exists beyond making money, and it is stated in a way that a person can recognize in the work of an ordinary Tuesday. The mission is what the business does, for whom and how, in the present tense, and it is close to a description of the offer. The vision is the future state the business is working toward, described far enough ahead to guide a decision and near enough to be imaginable. A business can hold all three at once, and they answer different questions: why we are here, what we do now, and where we are going.

The purpose is the one that is most often written badly, and the failure has a recognizable shape. A purpose written for a website tends to be large, warm and unfalsifiable: to be the most trusted partner in our industry, to make a difference for our clients, to deliver excellence every day. None of those sentences can be wrong, which is exactly the problem. A purpose that cannot be wrong cannot be used. The test this book applies is simple and severe: can a technician, a dispatcher or an administrator point to the purpose in something they did last week? If the answer is no, the sentence is decoration.

Nordec's purpose passes that test because it is written from the work rather than from the ambition. The company keeps commercial buildings running, with no surprises for the people who own them. A technician who replaces a belt before it fails is keeping a building running. A dispatcher who calls an owner to say that a part has been ordered and will arrive on Thursday is removing a surprise. An administrator who sends the one-page monthly report on the first working day is doing the same thing in a different medium. The purpose is not a claim about being the best; it is a description of what the seven people already do, written down so that it can be repeated and protected.

The mission is the plainest of the three and the easiest to write, because it is a description of the business as it stands. Nordec provides preventive maintenance and on-call technical service to commercial buildings of 2,000 to 9,000 square metres on Montréal's north shore, covering light heating and ventilation, minor plumbing, service electrical and loading doors, under three service tiers and a fixed-price assessment. Every word of that sentence can be checked against the offer designed in Book A6 (Designing a Value-Added Offer), which is what makes it a mission rather than a hope. A mission that describes a business the company does not yet run is a vision wearing the wrong label, and the confusion costs a business clarity at exactly the moment it is briefing somebody new.

The vision is the one that requires the most care, because it is the only one of the three that is not yet true. A vision that is too large becomes a poster; a vision that is too small becomes a plan. The useful middle is a state that the business could plausibly reach and that would change how it works. For Nordec, the vision that fits is a business whose building health record is the reason owners in its two territories call it first, and whose two territories are served so consistently that a new customer recognizes the company before the first visit. That is not a claim about size. It is a claim about what the business would be known for, and it is close enough that the seven people can see the steps toward it.

The three statements do different work in the platform, and it is worth saying which one carries the weight. The purpose is the one that survives a bad quarter, because it does not depend on growth. The mission is the one that a new employee learns in the first week, because it describes the work they will do. The vision is the one that decides which opportunities to refuse, because it says what the business is trying to become and therefore what would distract it. A business that has only a vision will chase it; a business that has only a mission will repeat itself; a business that has only a purpose will be admired and unfocused. The three belong together, and they belong on the same page as the promise, the values and the personality.

One habit keeps the three honest, and it costs nothing. Once a year, at the same meeting where the business reviews its positioning, the owner reads the purpose, the mission and the vision aloud and asks two questions of each. Is this still true of the work we did last year? And would a person who joined us six months ago recognize it? A statement that fails either question is not a failure of the business; it is a statement that has fallen behind the business, and the annual reading is where it is corrected before it becomes a sentence nobody believes.

WHO DOES THIS The owner writes, the team tests

In a small business, the platform is written by the owner and tested by the team, and the order matters. The owner holds the positioning, the refusals and the long view, and the owner is the only person who can settle a disagreement about what the business promises. But an owner who writes the platform alone and presents it finished will produce a document that the team reads politely and does not use. The practice that works is a short draft from the owner, followed by a session in which the people who do the work attack it. A technician who says that the promise cannot be kept on a Friday afternoon in February is not being difficult; they are performing the most useful test the platform will receive. The owner writes the purpose, the promise and the values; the team supplies the proof points, the objections and the words customers actually use. In a business of one, the same two roles are played by the same person on two different days, which is why the second day matters.

2.3 The Brand Promise

The brand promise is the sentence that says what a customer can count on, and it is the element of the platform that does the most work, because it is the one a customer can hold the business to. A promise is not a slogan and not an aspiration. It is a commitment about behavior, and it is useful only if it can be checked on a specific occasion by a specific person. That is the standard this chapter applies, and it is stricter than most businesses expect when they first write one.

Two tests separate a promise from a pleasant sentence. The first is the competitor test: could the closest competitor say the same thing without changing a single word? If the answer is yes, the sentence describes the category rather than the business, and it gives a customer no reason to choose. The second is the proof test: is there an occasion on which the promise can be shown to have been kept or broken? A promise that cannot fail cannot be trusted, because there is nothing at stake in it. A promise that can fail is a promise a business can be held to, and being held to it is what builds the memory a customer carries.

Nordec's rejected promise is the clearest illustration. The first candidate was "Always there for you." It is warm, it is short, and it fails both tests. The regional integrator could say it without changing a word, and it could say it more convincingly, because it has more trucks and more people. And it cannot be checked: there is no occasion on which a customer could say that Nordec was not there for them, because "there for you" has no measure. A promise that cannot be broken is not a promise; it is a mood.

The chosen promise is "We know your building, and we tell you ahead." It passes both tests, and it passes them in a way that can be pointed to. The first half is checkable on every visit: the technician arrives knowing what equipment is in the building, what was replaced last year, what is approaching the end of its life, because the building health record is kept for every contract building. The second half is checkable on every urgent call and every year: the 90-minute response is a commitment with a credit attached to it, and the capital planning note is a yearly document that lists the equipment approaching replacement with a window. A customer who says the company did not tell them ahead has something specific to point at, which is exactly what makes the promise worth making.

The promise also has to be true of the whole business rather than of the best day. A promise that only the lead technician can keep is a promise the business cannot make, because the customer will meet whoever is on call. This is where the platform meets the operations, and it is why the promise is written before the templates rather than after them. If the promise is that the customer is told ahead, then the record has to be updated on site before the technician leaves, because a record updated on Friday evening from memory will not support the sentence on Thursday afternoon. The values in section 2.4 are the mechanism that makes the promise repeatable, and the voice in Chapter 3 is the way it sounds.

A promise made to the public is also a representation, and in Canada that has a legal dimension that a business should understand without being alarmed by it. The Competition Act addresses representations to the public that are false or misleading in a material respect, both as a criminal offense under section 52 and as reviewable conduct under section 74.01(1)(a), and the civil provision applies to sales to businesses as well as to consumers. A brand promise or a tagline is a representation to the public. This book explains the law and does not replace the advice of a lawyer or a trademark agent, and the practical consequence for a platform is modest: a promise should be written so that the business can keep it, and the proof points should be real.

The last point about the promise is that it should be short enough to be remembered and specific enough to be repeated. A promise of twenty words will be paraphrased differently by each person who repeats it, and the paraphrase is what the customer will hear. A promise of eight or ten words can be carried intact. Nordec's promise is eleven words in English, and it is written in French first for the reasons section 2.8 sets out, because the customers, the technicians and the signs are in Québec.

KEY CONCEPT A promise you can check

The central idea of this chapter is that a brand promise is worth making only when a customer can check it. A checkable promise has three properties. It names a behavior rather than a feeling, so that it can be observed rather than sensed. It has an occasion attached to it, so that there is a moment at which it is kept or broken. And it is specific enough that the closest competitor could not repeat it without changing a word. A promise that fails any of the three is not a weaker promise; it is a different kind of sentence, closer to an aspiration, and it will do none of the work a promise does. The work a promise does is to make a business predictable. A customer who can predict what will happen chooses with less anxiety, refers with more confidence and complains about a specific failure rather than a general disappointment. That predictability is what a small business can offer against a larger competitor, and it costs nothing to state once the business has decided to keep it.

THE RULE A promise is a representation

Under the Competition Act, R.S.C. 1985, c. C-34, section 52 makes it an offense to knowingly or recklessly make a representation to the public that is false or misleading in a material respect, and section 74.01(1)(a) treats a representation to the public that is false or misleading in a material respect as reviewable conduct, applying to sales to businesses as well as to consumers. A brand promise or a tagline made to the public is a representation, and the rule that follows for a platform is that a promise should be written so that the business can keep it and so that its proof points are real. This book explains the law and does not replace the advice of a lawyer or a trademark agent; Book H4 (Consumer Protection, Advertising Law, Competition and Language Requirements) treats the subject in full.

2.4 Values That Change Decisions

A value is useful only if it changes a decision, and that single test removes most of the values a business writes for itself. Integrity, excellence, teamwork, respect and customer focus are not values in this book's sense, because no decision has ever turned on them. A person who is deciding whether to take a more profitable job instead of honoring a response commitment is not helped by the word integrity; they are helped by a value that says what to do. The values that belong on a platform are the ones that settle a real choice, and the way to find them is to look at the choices the business has already had to make.

Nordec's three values were found that way, and each one is written with the decision it changes attached to it. The first is Punctual, and the decision it changes is the one that arrives whenever a more profitable job competes with the 90-minute commitment: the commitment comes first, and the more profitable job is scheduled around it. The second is Straight, and the decision it changes is the one that arrives when a job is about to exceed the work included in the contract: the technician or the dispatcher makes the two-hour call before the work continues, and no invoice reaches an owner who has not heard about it first. The third is Thorough, and the decision it changes is the one that arrives at the end of a long day: the building health record is updated on site, before the technician leaves, rather than reconstructed later.

The three values are deliberately few, and they are deliberately unglamorous. A business of seven people can hold three values in mind; it cannot hold nine. And the values are written in the language of the work rather than in the language of a poster, because the person who has to apply them is standing in a mechanical room with a flashlight, not reading a strategy document. Punctual, Straight and Thorough are words a technician would use. That is not an accident of style; it is the condition under which a value is remembered at the moment it is needed.

The values also do the work that connects the promise to the operations. The promise says that the customer is told ahead. The value Straight is what makes that true when the news is unwelcome, because the two-hour call is the mechanism by which the telling happens. The promise says that the company knows the building. The value Thorough is what makes that true, because the record is the knowledge. The promise says that the company is there within 90 minutes. The value Punctual is what makes that true when a better job is available. Read this way, the values are not a separate element of the platform; they are the promise broken into the decisions that keep it.

The following table sets out the three values with what each one means, the decision it changes and the way it shows to a customer. Read it as a test rather than as a description: a value whose row cannot be filled in with a specific decision and a specific customer-visible result does not belong on the platform, and the space it occupies is better left empty.

Table 2.1 — Nordec's values and the decisions they change

The values have a second function that is easy to miss, which is to make a business legible to the people it hires. A new technician who reads three values with three decisions attached learns more about the company in ten minutes than from a policy manual, because the values describe what will actually be rewarded. A business that says it values punctuality and then rewards the technician who took the more profitable job has taught a different lesson, and the lesson will be learned. The platform is a statement of what the business intends, and the decisions are the evidence. Where the two disagree, the decisions win.

The last point about values is that they should be reviewed less often than the promise and more often than the purpose. A purpose can hold for a decade. A promise can hold for several years. A value is tied to the decisions a business actually faces, and those decisions change as the business grows. A value that was found in the choices of a seven-person company may need restating when the company has twenty people and a supervisor between the owner and the technician. The annual reading described in section 2.2 is the occasion for that review, and the question to ask is the same one: which decision did this value change last year?

WATCH OUT Values nobody could disagree with

The most common failure in a brand platform is a list of values that no reasonable person could oppose. Integrity, excellence, teamwork, innovation, respect and customer focus appear on thousands of walls, and none of them has ever settled an argument. The trap is subtle, because these words feel like commitments and they are written in good faith. The test that exposes them is to ask what decision would be different if the value were removed. If the answer is none, the value is a preference. A second version of the same trap is the value that is true but not distinctive: a business may genuinely be honest, and so may every competitor, which means honesty cannot be the reason a customer chooses. The remedy is to write values from the decisions the business has already had to make, in the words the people use, and to attach the decision to the value on the page. A platform with three values that each name a decision is stronger than a platform with nine that name none.

2.5 Brand Personality and Archetypes

The personality of a brand is the human temperament it presents, and it is the element of the platform that most directly shapes the words a business uses. Customers describe businesses as though they were people — friendly, cold, careful, hurried — and those descriptions are built from small things: how quickly a call is answered, whether an email is signed by a person, whether a technician explains what was replaced or simply leaves. A personality is not a costume laid over the work. It is a decision about which of the many true things about a business to emphasize, made so that the business is remembered as one thing rather than as seven.

The research on brand personality gives the decision a structure. Five dimensions of brand personality have been identified — sincerity, excitement, competence, sophistication and ruggedness — and a measurement scale for them has been reported as reliable, valid and generalizable (J. L. Aaker, 1997). The five dimensions are useful to a small business as a vocabulary rather than as a test. They give a name to the difference between a company that feels competent and one that feels exciting, and they make it possible to say which of the two the business is choosing. The scale was developed in a particular research setting, and this book does not claim that it applies unchanged in Québec; what travels is the idea that personality can be described in a small number of dimensions rather than in an endless list of adjectives.

A second approach to the same question comes from the study of archetypes, which presents a twelve-archetype approach to building a brand identity from the stories already present in an organization's culture, and which cautions against the negative side of archetypes — the way a strong pattern, pushed too far, becomes a caricature (Mark & Pearson, 2001). The approach is useful to a small business for one reason: it starts from the stories people in the business already tell, rather than from a list of adjectives chosen in a meeting. Its caution is equally useful. A company that decides it is the reliable one can slide into being the slow one, and a company that decides it is the bold one can slide into being the careless one. The book presents the approach as a way of working, and it offers no evidence of performance; a business should treat it as a lens rather than as a formula.

A third model, presented in the strategic brand management literature, is the brand identity prism, which is offered among the models a business can use to think about what a brand is made of (Kapferer, 2012). This book presents the prism as a model that exists and is used, and it does not reproduce its facets, because they were not read at source. The reason to mention it at all is that a reader who encounters the prism elsewhere should recognize it as one of several ways of organizing the same material, and should not treat any one of them as the only correct one. The platform in this book is deliberately simpler than any of the models, because it has to be used by seven people on a Tuesday.

Nordec's personality is four traits: calm, precise, plain-spoken, neighborly. Each one was chosen against a real alternative. Calm rather than urgent, because a building owner with a failed system does not need a contractor who sounds alarmed. Precise rather than thorough-sounding, because the value of the building health record is that it names equipment and dates rather than describing a general condition. Plain-spoken rather than technical, because the owner of a building is not a technician and should not have to translate. Neighborly rather than corporate, because the business works in two territories where its customers know each other, and because the regional integrator already occupies the corporate position.

The four traits are written into the voice in Chapter 3, and the connection between the two is direct. A personality that is not written into the voice is a list of adjectives on a page. Calm becomes a rule about how an urgent call is described. Precise becomes a rule about numbers and dates. Plain-spoken becomes a glossary of words to use and words to avoid. Neighborly becomes a rule about how a report is signed and how a customer is addressed. The personality is the part of the platform that a customer experiences most often, because it is carried in every sentence the business writes and every call it answers, and it is the part that is easiest to lose when a business is busy.

2.6 Brand Equity and How It Is Built

Brand equity is the value a brand adds to an offer, and the most useful definition for a small business is the one that starts from the customer. Customer-based brand equity is the differential effect of brand knowledge on how consumers respond to the marketing of the brand; brand knowledge has two components, brand awareness and brand image; and equity is positive when customers respond more favorably to the branded offer than they would to an equivalent unbranded one (Keller, 1993). The definition matters because it locates the equity in the customer's mind rather than in the business's accounts, and because it names the two things a small business can actually work on: whether people know the brand, and what they associate with it.

The same body of work describes how equity is built, in four steps and six building blocks. The four steps run from identity, through meaning, to responses, and finally to relationships; the six building blocks are salience, performance, imagery, judgments, feelings and resonance, with resonance at the top of the pyramid and four kinds of resonance named — behavioral loyalty, attitudinal attachment, a sense of community and active engagement (Keller, 2001). The report presents the model as a yardstick for brand tracking, and it is conceptual rather than empirically validated, which is worth saying plainly. For a small business, the value of the pyramid is that it puts the elements in an order. A business cannot build resonance before it has salience, and it cannot build salience before it has decided what it is.

Salience is the block at the base, and it deserves its own definition because it is often confused with awareness. Brand salience is the propensity of a brand to be noticed or to come to mind in buying situations, and it reflects the quantity and quality of the memory associations a brand has; it is distinct from awareness and from attitude (Romaniuk & Sharp, 2004). The distinction is practical. A building owner may be aware of Nordec, in the sense of recognizing the name, and still not think of the company when a rooftop unit fails, because the association between Nordec and that moment has not been built. Salience is what makes a brand come to mind at the moment of need, and it is built by repetition of consistent cues rather than by a single campaign.

The balance between building a brand and activating a sale is the subject of a body of work that studies the tension between long-term brand building and short-term activation and warns against using very short-term metrics as the primary measures of performance (Binet & Field, 2013). The practical warning is easy to state and hard to follow in a small business: a month in which every proposal came from a referral looks like a good month, and it teaches nothing about whether the business is becoming known. A business that measures only the last step of the sale will keep doing what produced the last step, and it will not notice that the pool of people who know it is not growing.

The proportion that is often quoted in this connection comes from the same line of work as reported by the Institute of Practitioners in Advertising, which states the 60:40 split between brand building and activation associated with Binet and Field, and notes that a 2018 follow-up found about 62:38 (Hoad, 2023). That figure is a guideline reported for advertisers in the studies, and it is not a rule for a small business. A seven-person company does not have an advertising budget to divide, and it should not invent one. What the guideline offers is a way of thinking: some of the effort should go to being known, and some to being chosen, and a business that spends everything on being chosen will eventually have nobody left to choose it.

For Nordec, the pyramid translates into a small number of ordinary acts. Salience is built by the vans, the uniforms, the monthly reports and the consistency of the name across every touchpoint. Performance is the 90-minute response and the quality of the work. Imagery is the building health record and the capital planning note, which are the tangible things a customer can point to. Judgments and feelings are what the owner concludes about the company after a year of service. Resonance is what the case will look for in Chapter 9, where the measurement of brand health is treated, and where the four kinds of resonance give the survey its questions.

WHAT CHANGED The brand equity pyramid, from 1993 to 2001

The idea of customer-based brand equity was set out in 1993, when it was defined as the differential effect of brand knowledge on how consumers respond to the marketing of a brand, with brand knowledge divided into awareness and image (Keller, 1993). Eight years later, the same author published a report that turned the definition into a structure for building a brand: four steps running from identity through meaning and responses to relationships, six building blocks named salience, performance, imagery, judgments, feelings and resonance, and four kinds of resonance at the top of the pyramid (Keller, 2001). The change is one of use rather than of meaning. The 1993 work explains what equity is and how it can be measured; the 2001 report offers a sequence a business can follow and a yardstick it can track against. The report is conceptual and offers no empirical validation, so it should be read as a way of organizing the work rather than as a proven recipe, and the sequence it proposes is the reason this book treats salience before resonance.

2.7 The Brand Platform on One Page

The platform is assembled on one page, and the page has a fixed shape so that it can be read at a glance and applied without a meeting. At the top sits the purpose, because it is the sentence that explains why the rest exists. Below it sits the promise, because it is the commitment the customer can check. Below that sit the values, each with the decision it changes, because the values are the mechanism that keeps the promise. Then come the personality traits, because they shape the words. At the bottom sit the tagline in French and in English, and the proof points, which are the evidence that the promise is real. A page arranged this way answers the questions a designer, a new employee and a proposal writer will ask, in the order they will ask them.

Figure 2.1 shows that page for the case followed through this book, and it is worth describing it in words for a listener. The figure is drawn as a single column: the purpose sits at the top, the promise beneath it, then the three values with the decision each one changes, then the four personality traits, and the tagline in French and in English at the bottom. A listener should picture a page that can be read from top to bottom in about a minute, with no paragraph longer than a line and no element that requires a second document to understand.

Figure 2.1 — Nordec's brand platform on one page. Each value names the decision it changes, which is what separates a value from a slogan.

The page is used in four ordinary situations, and each one tests a different part of it. The first is a briefing: when the business hires an independent design studio, the page is what the studio reads before proposing routes, and it is the reason three routes can be compared against a standard rather than against taste. The second is writing: when a proposal, a report or a letter is drafted, the promise and the personality decide what the first sentence says and what the business does not claim. The third is onboarding: a new technician or dispatcher reads the page in the first week and can describe what the company promises by the end of it. The fourth is a disagreement: when two people in the business want to do different things, the page is the place to look, and the value with its decision attached is usually enough to settle it.

The following table sets out the elements of Nordec's platform with the wording the business uses and the test each element must pass. Read it as a working document rather than as a summary: the third column is the part that keeps the platform honest, because an element that fails its test is an element that will not survive the first busy month.

Table 2.2 — Nordec's brand platform

Two habits keep the page alive. The first is that the page is stored where the work happens rather than where the strategy lives: with the templates, in the folder the administrator keeps, and in the project the business uses to run the rollout. The second is that the page is dated. A platform without a date is a document that nobody knows whether to trust, and a platform with a date can be reviewed on the anniversary of that date, which is the practice section 2.2 describes.

METHOD Running a platform workshop in two half-days

First, read the positioning and the customers' own words aloud, because the platform is derived from both and neither can be invented in the room. Second, draft the purpose in one sentence and test it on daily work by asking each person to name something they did last week that the sentence describes; a purpose that nobody can illustrate is rewritten. Third, write three candidate promises and test each one for proof, asking for the occasion on which it could be shown to have been kept and for the competitor who could say it unchanged. Fourth, name the values by the decisions they change, working from choices the business has already had to make rather than from a list of virtues, and write the decision beside each value. Fifth, choose four personality traits and write them into the voice, turning each trait into a rule about words, numbers or tone. Deliverable: one page, dated, with the purpose, the promise, three values and their decisions, four traits and the tagline. Pitfall: finishing with more than three values, which is the sign that the room has stopped deciding and started listing.

IN LA MACHINE Where the platform can be written

The application read for this book has a Branding page at business/branding, and it is worth being precise about what it does. It offers a logo drop zone, a brand name, a tagline and a website address, a primary and a secondary color with ten preset swatches, a choice among five fonts, an email footer and three social links, and it computes the contrast of the primary color and shows a pass or fail badge at the 4.5:1 threshold, with a live preview of a client invitation card and a client email (La Machine, 2026). The tagline field is the only place on that page where a piece of the platform could be entered, and the values entered there are not saved: the save button shows a confirmation and no other page uses what was entered. The page is a place to try a palette and see its contrast, not a brand register. The practical answer is that the platform is best kept as a document attached to a project, and the Projects page at productivity/projects is where the brand work and its recurring checks can live. La Machine has no brand platform page and no saved brand kit that other pages read.

2.8 The Case's Brand Platform

Nordec's platform was written in two half-day workshops with all seven people, and the shape of those two days is worth describing because it is repeatable. The first half-day was spent on the material rather than on the wording. The seven people read the positioning from Book A3 (Strategy and Competitive Positioning) and the offer from Book A6 (Designing a Value-Added Offer), and then they read the customers' own words from the twenty value interviews that Book A6 had already conducted. Nothing was drafted in the first half-day. The second half-day was spent drafting, testing and cutting, and it ended with one page.

The customers' words did most of the work, and they are the reason the promise is what it is. In the twenty interviews, the words customers used about Nordec were counted: "on time" or "fast" was used by fifteen of the twenty, "knows my building" by eleven, "tells me ahead" by nine, "straight with me" by seven and "friendly" by five. Read as a list, the five phrases describe a company that is quick, informed, forthcoming, honest and pleasant. Read as a platform, they divide into two groups. Speed is already promised elsewhere, in the 90-minute guarantee designed in Book A6, and it is the one thing the independent trades also claim. Knowledge and warning are the two phrases that describe something the competitors do not offer in the same way, and they became the promise.

The purpose came from the same material, stated in the plainest words the room could agree on: keep commercial buildings running, with no surprises for the people who own them. It is true of a preventive visit, of a monthly report and of a capital planning note, and every person in the room could name something they had done that week which the sentence described. The three values followed from the decisions the business had already had to make, and each one carries the decision it changes: Punctual, because the 90-minute commitment comes before a more profitable job; Straight, because the two-hour call comes before work that exceeds what the contract includes, and no invoice reaches an owner who has not heard about it; Thorough, because the record is updated on site, before the technician leaves. The personality is four traits — calm, precise, plain-spoken, neighborly — and the tagline is written in French first: « Votre bâtiment, un an d'avance. », with "Your building, a year ahead." in English.

The rejected promise is the part of the story that other businesses find most useful. The first candidate was "Always there for you," and it was rejected for two reasons that are worth repeating. The regional integrator could say it without changing a word, and could say it with more trucks behind it. And it could not be checked: there is no occasion on which a customer could show that Nordec had not been there for them. The chosen promise, "We know your building, and we tell you ahead," can be checked on every urgent call, through the 90-minute response, and every year, through the capital planning note. That is the difference between a sentence that sounds like a brand and a sentence that behaves like one.

The tagline is written in French first for a reason that belongs to Québec and to the business's own market. In Québec, a slogan that accompanies a trademark on an outdoor sign counts among the French terms that make French markedly predominant on that sign, as the Office québécois de la langue française explains (Office québécois de la langue française, 2025a). Nordec's customers, technicians and signs are in Québec, so the French line is the primary one and the English line is a translation of it rather than the other way around. This book explains the language rules and does not replace the advice of a lawyer or a trademark agent.

The platform was approved by all seven people, and the approval mattered more than the wording. A platform that the owner writes and the team receives is a document; a platform that the team helped to cut is a decision. The administrator, who becomes the brand keeper in Chapter 5, was in the room when the values were written, which is why the templates that follow from them are checked against a standard the administrator helped to set. The next chapter takes the promise and the personality and turns them into names and a voice.

QUÉBEC & CANADA NOTE A tagline in French first

In Québec, a slogan that accompanies a trademark on an outdoor sign counts among the French terms that make French markedly predominant on that sign, as the Office québécois de la langue française explains (Office québécois de la langue française, 2025a). That is why Nordec writes its tagline in French first — « Votre bâtiment, un an d'avance. » — and treats the English line as a translation rather than as the original. The practical consequence for a platform is that the French wording is drafted by the people who will use it, tested on a sign and on a report, and settled before the English version is written, because a tagline translated from English into French rarely survives the trip intact. This book explains the language rules and does not replace the advice of a lawyer or a trademark agent; Book H4 (Consumer Protection, Advertising Law, Competition and Language Requirements) treats the requirements in full.

DO THE MATH What customers said about Nordec

Illustrative figures; the case is fictional. Step 1 — The twenty value interviews of Book A6 (Designing a Value-Added Offer) are the source, and every phrase below is counted against those twenty people. Step 2 — "On time" or "fast" was used by 15 of 20, which is 75%; it is the strongest signal in the set, and it is already promised by the 90-minute guarantee. Step 3 — "Knows my building" was used by 11 of 20, which is 55%; it is the first half of the chosen promise. Step 4 — "Tells me ahead" was used by 9 of 20, which is 45%; it is the second half of the chosen promise, and the two halves together describe something the competitors do not claim. Step 5 — The promise therefore uses the second and the third phrases, and leaves the first to the guarantee that Book A6 already designed, because a promise that repeats a guarantee adds nothing a customer can check.

CASE IN POINT Nordec Entretien inc.: the promise that was rejected

Nordec Entretien inc. is a fictional business created for this encyclopedia; its owner, its employees, its customers, its competitors, its figures and the events described are invented, and any resemblance to a real business is coincidental. The first promise the seven people wrote was "Always there for you," and it lasted about twenty minutes. The lead technician killed it with one question: could the regional integrator say that? The answer was yes, and the regional integrator could say it with more trucks. The second objection came from the dispatcher, who asked how anyone would know whether the promise had been kept. There was no occasion on which a customer could point to a failure, because "there for you" has no measure. The room then went back to the customers' own words and found two phrases that could be checked: the building is known, and the owner is told ahead. The chosen promise, "We know your building, and we tell you ahead," has been checkable ever since — on every urgent call through the 90-minute response, and every year through the capital planning note. The rejected sentence was not wrong. It was simply a sentence that no one could be held to.

TOOLS OF THE TRADE What a platform workshop needs

A platform workshop needs five ordinary things and no software. The first is a wall or a whiteboard, large enough that the whole platform can be seen at once, because a platform assembled on separate pages is a platform nobody can test as a whole. The second is the interview notes from the customer research, in the customers' own words, because the promise is derived from what customers already say rather than from what the business wishes they said. The third is a one-page template with the elements already laid out — purpose, promise, values with their decisions, personality, tagline, proof points — so that the room is filling in a shape rather than inventing one. The fourth is a timer, because the second half-day is where a room will happily spend two hours on a single adjective, and the discipline of the workshop is that the page is finished by the end of it. The fifth is a photograph of the finished wall, taken before anyone leaves, because the page will be typed up later and the argument that produced a wording is easier to remember from the wall than from the transcript.

Costly Mistakes

• Writing the platform after the logo. A business that commissions a mark before deciding what it promises is asking a designer to invent the strategy by accident, and the result is a logo that is pleasant and empty.
• Writing a purpose that cannot be wrong. A purpose that no reasonable person could oppose cannot be applied, because there is no decision it could change and no week's work it could describe.
• Making a promise the closest competitor could repeat. A promise that describes the category rather than the business gives a customer no reason to choose, and it wastes the strongest sentence the platform contains.
• Listing values that have never changed a decision. Integrity, excellence and teamwork are preferences rather than values, and a platform built on them will be admired on the wall and ignored in the mechanical room.
• Leaving the personality as a list of adjectives. Calm, precise, plain-spoken and neighborly do nothing until each one becomes a rule about words, numbers or tone in the voice the business actually uses.
• Writing the platform alone and presenting it finished. A platform the team receives is a document; a platform the team helped to cut is a decision, and only the second survives a busy month.
• Treating the platform as a marketing document. Its most important readers are the seven people who do the work, and a platform that is filed with the brochures rather than with the templates will not be applied.

Key Takeaways

• The positioning of Book A3 (Strategy and Competitive Positioning) says where the business competes; the brand strategy says what it promises there, what it stands for, how it behaves and how it sounds, and the second is written before anyone draws a logo.
• Brand identity is what the strategist aspires the brand to be, and brand position is the part of that identity actively communicated, so a business can hold an identity broader than its current position (D. A. Aaker, 1996).
• A promise is worth making only when a customer can check it: it names a behavior, it has an occasion attached to it, and the closest competitor could not repeat it unchanged, which is why "Always there for you" was rejected and "We know your building, and we tell you ahead" was kept.
• A value is useful only if it changes a decision, and Nordec's three values each carry the decision they change: Punctual puts the 90-minute commitment before a more profitable job, Straight puts the two-hour call before work that exceeds the contract, and Thorough puts the record update on site before the technician leaves.
• Customer-based brand equity is the differential effect of brand knowledge on how customers respond, with awareness and image as its two components (Keller, 1993), and the brand equity pyramid orders the work from salience through performance, imagery, judgments and feelings to resonance, with the report offered as a conceptual yardstick rather than a validated recipe (Keller, 2001).
• The platform fits on one page — purpose, promise, values with their decisions, personality, tagline and proof points — and it is used to brief a designer, to write a proposal, to onboard a new person and to settle a disagreement.

Your Action Plan

1. Read the positioning and the customers' words before drafting anything. Take the positioning from Book A3 (Strategy and Competitive Positioning) and the interview notes from Book A6 (Designing a Value-Added Offer), and list the phrases customers actually used, with the number of times each one appears. Deliverable: one page of the customers' own words, counted. Verification: every phrase on the page is quoted from an interview rather than paraphrased by the business.
2. Draft the purpose in one sentence and test it on last week's work. Write the purpose, then ask every person in the business to name something they did in the past week that the sentence describes. Deliverable: one purpose sentence that every person can illustrate. Verification: a person who cannot illustrate it either rewrites the sentence or explains why the work does not match it.
3. Write three candidate promises and test each one twice. For each candidate, ask whether the closest competitor could say it unchanged and name the occasion on which a customer could show it was kept or broken. Deliverable: one chosen promise with its two tests recorded. Verification: the chosen promise names a behavior, has an occasion, and cannot be repeated by the closest competitor without a change.
4. Name three values by the decisions they change. Work from choices the business has already had to make, write the decision beside each value, and describe how each one shows to a customer. Deliverable: three values, each with its decision and its customer-visible result. Verification: for each value, a person can name a decision from the past year that it settled.
5. Assemble the one page, date it and store it where the work happens. Put the purpose, the promise, the values with their decisions, four personality traits, the tagline in French and in English, and the proof points on a single page, and file it with the templates rather than with the brochures. Deliverable: one dated platform page. Verification: a person who has not read it can describe what the business promises after reading it once.

In La Machine

The Branding page at business/branding is the closest thing the application has to a place for a brand, and it is a place to try a palette rather than a place to keep a platform (La Machine, 2026). It offers a logo drop zone, a brand name, a tagline and a website address, a primary and a secondary color with ten preset swatches, a choice among five fonts, an email footer and three social links, and it computes the contrast of the primary color and shows a pass or fail badge at the 4.5:1 threshold, with a live preview of a client invitation card and a client email. The tagline field is the only element of the platform it can hold. Its values are not saved: the save button shows a confirmation and no other page uses what was entered, so the page is a place to try a palette and see its contrast, not a brand register.

The practical answer for a business writing its platform is to keep the page as a document and to attach it to a project. The Projects page at productivity/projects can carry the brand work and its recurring checks, which is where the annual reading of the purpose, the mission and the vision described in section 2.2 can live as a task rather than as an intention. What the platform does not have is equally worth stating plainly: there is no brand platform page, no saved brand kit that other pages read, no brand portal or guidelines publisher, and no brand consistency audit. The platform is a document the business writes and keeps, and the application is where the work around it is organized rather than where the platform itself is stored.

Self-Check

1. What is the difference between the positioning chosen in Book A3 (Strategy and Competitive Positioning) and the brand strategy that follows it, and why does the second come first in this book?
2. What are the two tests a brand promise must pass, and why was "Always there for you" rejected?
3. What makes a value useful, and what decision does each of Nordec's three values change?
4. What are the five dimensions of brand personality, and what does the archetype approach add to the choice of four traits?
5. What are the two components of brand knowledge in the definition of customer-based brand equity, and what does the brand equity pyramid add to that definition?

Answers. 1. The positioning says where the business competes — which customers, which buildings, which services and which refusals — while the brand strategy says what the business promises there, what it stands for, how it behaves and how it sounds; the second comes first in this book because the logo, the names, the templates and the touchpoints are all derived from the platform, and a business that starts with the mark is asking a designer to invent the strategy by accident. 2. A promise must pass the competitor test, meaning the closest competitor could not say it without changing a word, and the proof test, meaning there is an occasion on which a customer could show it was kept or broken; "Always there for you" failed both, because the regional integrator could say it with more trucks behind it and because there is no occasion on which anyone could point to a failure. 3. A value is useful only if it changes a decision, and a value that has never changed one is a preference; Punctual changes the decision between the 90-minute commitment and a more profitable job, Straight changes the decision to continue work that is about to exceed what the contract includes, and Thorough changes the decision to update the building health record on site rather than later. 4. The five dimensions are sincerity, excitement, competence, sophistication and ruggedness, and a measurement scale for them has been reported as reliable, valid and generalizable (J. L. Aaker, 1997); the archetype approach adds a way of starting from the stories already present in an organization's culture rather than from a list of adjectives, and it cautions against the negative side of a strong pattern (Mark & Pearson, 2001). 5. Brand knowledge has two components, brand awareness and brand image, and equity is positive when customers respond more favorably to the branded offer (Keller, 1993); the brand equity pyramid adds a sequence of four steps and six building blocks — salience, performance, imagery, judgments, feelings and resonance — with four kinds of resonance at the top, and it is offered as a conceptual yardstick for tracking rather than as a validated recipe (Keller, 2001).

Chapter References

Aaker, D. A. (1996). Building strong brands. Free Press.

Aaker, J. L. (1997). Dimensions of brand personality. Journal of Marketing Research, 34(3), 347–356. https://doi.org/10.1177/002224379703400304

Binet, L., & Field, P. (2013). The long and the short of it: Balancing short and long-term marketing strategies. Institute of Practitioners in Advertising.

Competition Act, R.S.C. 1985, c. C-34, ss. 52 and 74.01 (Canada). Justice Laws Website, current to July 21, 2026 (last amended March 26, 2026) — as read for Book A6 (Designing a Value-Added Offer) on September 22, 2026.

Hoad, A. (2023, September 13). The next chapter for The Long and the Short of It. Institute of Practitioners in Advertising. https://ipa.co.uk/knowledge/ipa-blog/the-next-chapter-for-the-long-and-the-short-of-it

Kapferer, J.-N. (2012). The new strategic brand management: Advanced insights and strategic thinking (5th ed.). Kogan Page.

Keller, K. L. (1993). Conceptualizing, measuring, and managing customer-based brand equity. Journal of Marketing, 57(1), 1–22. https://doi.org/10.1177/002224299305700101

Keller, K. L. (2001). Building customer-based brand equity: A blueprint for creating strong brands (Report No. 01-107). Marketing Science Institute.

La Machine. (2026). Documentation — route reference inventory. Internal product documentation, read September 25, 2026, together with the code of the pages this book describes: business/branding and productivity/projects.

Mark, M., & Pearson, C. S. (2001). The hero and the outlaw: Building extraordinary brands through the power of archetypes. McGraw-Hill.

Neumeier, M. (2005). The brand gap (Rev. ed.). New Riders.

Office québécois de la langue française. (2025a). Affichage des marques de commerce et des noms d'entreprises (updated August 27, 2025). Retrieved September 25, 2026, from https://www.oqlf.gouv.qc.ca/francisation/entreprises/affichage-marques-noms.html

Romaniuk, J., & Sharp, B. (2004). Conceptualizing and measuring brand salience. Marketing Theory, 4(4), 327–342. https://doi.org/10.1177/1470593104047643

Chapter 3 — Naming and Verbal Identity

A name is the most repeated part of a brand and the hardest part to change, so it is chosen by a process that screens for meaning, language and legal availability before anyone grows attached to it, and it is surrounded by a voice that makes every sentence sound like the same business.

What You'll Be Able to Do

• Explain why a name is the first distinctive brand asset, how repetition turns it into memory, and why changing it later is the most expensive decision in this book.
• Classify the kinds of names a small business can choose — descriptive, suggestive, invented, founder's name, acronym and place name — and state the trade-off each one carries.
• Run a naming process from the brief to the decision, with a long list, a language and meaning screen, a knockout search and a short list that reaches the owner.
• Screen a candidate name for pronunciation and meaning in French and English, for unwanted meanings, for domain names and social handles, and for close marks in the official databases.
• Describe how a trademark is applied for and protected in Canada and in the United States, what the official fees and the wait to examination are, and what a trademark agent does that no search can do.
• Write a voice with principles, a glossary and key messages by audience, so that a visit sheet, a monthly report and a proposal all sound like the same business.

Why It Matters

A brand is made of many parts, and one of them is spoken more often than all the others together. The name appears in every sentence a customer says about the business, in every search they type, in every invoice they receive, in every referral they make, on every van and on every uniform. A logo is seen; a name is said. The case followed through this book counts its own exposures in a year: 624 preventive visits, 576 monthly reports, about 276 urgent calls, two vans on the road every working day and seven people in uniform. Each of those is an occasion on which the name is either spoken or read, and almost none of them is an occasion on which the logo is studied. The name is the element that carries the brand through the greatest number of repetitions, which is why it is the first distinctive asset a business should get right.

The cost of getting it wrong is not paid in one moment but in a slow loss of recognition. The case's own inventory of 36 touchpoints found three versions of the logo in circulation, five different blues, four typefaces, and only 13 of 36 touchpoints consistent with one another, which is 36.1%. The name was the one element that had not drifted, and it was the one element that could not have been repaired cheaply if it had. A name that is hard to pronounce, easy to misspell or confusable with another business loses referrals silently, because a referral that never happens leaves no trace in any report. In the case, nine of the fifteen proposals decided in the six months after the offer refresh came from a customer referral or from another building of a current manager, and every one of those nine began with somebody saying a name out loud to somebody else.

The voice is the other half of the same work, and it is the cheapest part of the brand to fix. A name is spoken; the voice is what surrounds it in writing, on the visit sheet, in the monthly report, in the proposal and in the email that follows a missed appointment. The case's visit sheet used to read "Preventive maintenance performed as per contract scope; anomalies noted," which is a sentence that could have been written by any competitor in the region. It now reads "We checked the rooftop units and the loading doors today. One door spring is worn: we will replace it on the next visit, within the contract, unless you call us first." Nothing about the service changed between those two sentences. What changed is that the second one sounds like the business that wrote it, and it costs nothing to repeat.

In This Chapter

Section 3.1 explains why a name is the most repeated element of a brand, what the law can protect and what changing it later costs. Section 3.2 sets out the six kinds of names a small business can choose and the trade-off each one carries. Section 3.3 describes the naming process from the brief to the decision, with its funnel of screens. Section 3.4 covers the language, meaning and availability screens, including the knockout search and the layered clearance search a business needs if it may sell in the United States. Section 3.5 covers trademarks in Canada and the United States, the official fees, the wait to examination and the Madrid Protocol. Section 3.6 covers naming in Québec under the Charter of the French language, through the rules of the Registraire des entreprises and of the Office québécois de la langue française. Section 3.7 covers voice, tone, the glossary and key messages by audience. Section 3.8 follows the case through the naming of its entry offer and the rollout of its voice.

3.1 Why Names Matter

A brand is made of many parts, and one of them is spoken more often than all the others together. The name appears in every sentence a customer says about the business, in every search they type, in every invoice they receive, in every referral they make, on every van and on every uniform. A logo is seen; a name is said. The case followed through this book counts its own exposures in a year: 624 preventive visits, 576 monthly reports, about 276 urgent calls, two vans on the road every working day and seven people in uniform. Each of those is an occasion on which the name is either spoken or read, and almost none of them is an occasion on which the logo is studied. The name is the element that carries the brand through the greatest number of repetitions, which is why it is the first distinctive asset a business should get right.

Repetition is what turns a name into memory, and memory is what a customer needs in order to come back. A person who wants the same service again must be able to say the name, spell it, find it and pass it on. In the case, six of the fifteen proposals decided in the six months after the offer refresh came from a referral by a customer, and three more came from another building of a current manager. Those nine proposals began with somebody saying a name out loud to somebody else. A name that is hard to pronounce, easy to misspell or confusable with another business loses part of that movement, and it loses it silently, because a referral that never happens leaves no trace in any report and no line in any spreadsheet.

The name is also the part of the identity that the law can protect, and that fact separates it from everything else in this book. A logo can be redrawn, a palette can be replaced and a set of typefaces can be exchanged, and the business keeps its recognition because the name has not moved. A name can be registered as a trademark, and the owner of a registered trademark controls its use for the goods and services named in the registration. That control is what allows a business to stop another from trading on the recognition it has built, and it is the reason the naming decision is taken with a search rather than with a vote. The search is described in section 3.4 and the registration in section 3.5, and both of them come before the name is printed on anything.

The cost of changing a name is the reason the decision deserves the time it takes. A business that changes its name gives up the recognition it has accumulated and starts again with the people who knew the old one, and it does so at the moment when it is also paying for the new signs, the new documents and the new website. The case followed through this book faced that question in Chapter 8 and answered it by keeping the name: the refresh is an evolution, and the name, the blue family and the letter N are kept as the assets people already recognize, while the wrench, the five blues and the four typefaces go. The name survived because it was the element with the most repetitions behind it and the least reason to move. A business that changes its name pays twice, once for the new identity and once for the recognition it has to rebuild.

There is a second reason the name is treated first, and it is about what the name carries. A name is the shortest possible statement of the promise, and a good one holds a piece of the positioning inside it. The case's promise is "We know your building, and we tell you ahead," and the tagline that travels with the name is « Votre bâtiment, un an d'avance. » in French and "Your building, a year ahead." in English. Neither of those fits inside the name itself, and that is the point: the name is the anchor, and the tagline is the sentence that hangs from it. A business that tries to put the whole promise into the name ends with a descriptive name that cannot be registered, which is the trap described in section 3.2.

The name is also the element that a customer uses to find the business again, and the case shows how small that movement can be. Of the fifteen prospects who received a proposal in the six months after the refresh, two had found Nordec through a web search, and one had seen a Nordec van. Those three arrived because a name was visible somewhere and could be typed or remembered. The other twelve arrived through people. A name that is easy to say and easy to spell is what makes both routes work, and it is the only part of the identity that a customer carries away without being given anything at all.

Finally, the name is the element that a business cannot outsource to a template. A logo can be drawn on a template site, as the case's first logo was in 2022, and a palette can be assembled from a set of presets. A name has to be chosen, and the choice has consequences that no template can absorb: it decides what the business is called in a search, in a register, on a sign and in a customer's sentence. That is why the naming work in this chapter comes before the visual work of Chapter 4, and why the screens described in section 3.4 are run before anybody becomes attached to a candidate. Chapter 2 chose the positioning; this chapter chooses the words that will carry it for as long as the business exists.

DEFINITION Trade name, trademark, tagline, voice and tone

A trade name is the name under which a business carries on its activity, the name that appears on the register and on the contract, and it identifies the business itself. A trademark is a name, a symbol, a slogan or a combination of them used to distinguish the goods or services of one business from those of another, and it is the form of the name that can be registered and protected. A tagline is a short phrase that travels with the name and compresses the promise, such as « Votre bâtiment, un an d'avance. ». Voice is the constant personality of the writing, the way the business always sounds whatever the occasion. Tone is the adjustment of that voice to the situation: the same voice in a cheerful announcement and in a message about a missed appointment, at a different temperature. Voice is the rule; tone is the application of the rule to a particular reader on a particular day.

3.2 Kinds of Names

There are six kinds of names a small business can choose, and the choice among them is a strategic decision rather than a matter of taste. Each kind carries a different balance between being understood quickly and being owned exclusively, and no kind is best in every situation. The right question is not which kind sounds most professional but which kind does the work the positioning of Chapter 2 requires: whether the business needs to be found by people who do not know it yet, whether it needs to be protected from imitators, whether it intends to sell in two languages, and whether it expects to add services in the next five years.

A descriptive name says what the business does. It is understood in one second, it needs no explanation, and it is the first kind most owners reach for. It is also the hardest kind to protect, because the words describe the services rather than distinguish them, and a name that a trademark agent judges hard to register may leave the business with no registered right to stop another from using the same words in the same city. A suggestive name hints at a benefit without stating it. It is easier to protect than a descriptive name and easier to understand than an invented one, and it carries a trade-off that has been studied: a name that suggests a benefit improves recall of advertised claims consistent with that benefit but lowers recall of claims about other benefits (Keller et al., 1998). A business that names itself after speed and later wants to be known for thoroughness is working against its own name.

An invented name is a word that did not exist before it was chosen. It is the easiest kind to protect and the hardest kind to remember on first hearing, and the sound of it is not neutral. In a set of studies, participants chose between invented brand names that differed only in a vowel sound, and a name was preferred when what the sound connotes fitted the category and disliked when it did not (Lowrey & Shrum, 2007). The studies were conducted in English only, so the finding is a reason to test a candidate in both languages rather than a rule to apply. A founder's name is warm, personal and immediately credible in a service business where the customer meets the owner, and it ties the brand to a person who may one day sell the business. An acronym is compact and often meaningless to a stranger, and it works best when the full name is already known. A place name gives local credibility and a natural boundary, and it becomes a constraint the day the business wants to serve a territory that does not share the name.

The case followed through this book kept a short coined name, Nordec, which is neither descriptive nor a founder's name, and it works in French and in English without translation. That choice was made in 2022 without a process, and the business kept it in the refresh of Chapter 8 because the name had four years of repetitions behind it. What the refresh added was the discipline the first choice had lacked: the entry offer, the tiers and the voice were all named and written through the screens described in the next two sections, and the result is a set of names that can be said in both languages, found in a search and registered.

The following table sets out the six kinds of names with an invented example of each, the strength and the weakness each one carries, and the legal note that follows from its kind. Read it as a menu rather than a ranking: the columns are the kind of name, an example that belongs to no real business, what the kind does well, what it costs, and what a trademark agent will say about it.

Table 3.1 — Kinds of names

WATCH OUT The descriptive name nobody can own

A descriptive name says exactly what the business does, which is why it is the first kind most owners reach for and the hardest kind to protect. "Montréal Plumbing and Heating" tells a customer everything in four words, and it tells a trademark examiner that the words describe the services rather than distinguish them, which is the opposite of what a trademark is for. The practical consequences arrive in a specific order. The name is likely to be hard to register, which a trademark agent assesses, and without a registration the business has no registered right to stop another business from using the same words in the same city. The name cannot be found easily, because a search for it returns every competitor who chose the same description. And the name cannot grow, because a business that adds electrical work to its plumbing services now has a name that is half wrong. The remedy is not to avoid description altogether; it is to keep the descriptive words out of the name and put them in the sentence that follows it.

3.3 A Naming Process

A naming project is a funnel, and the funnel is what keeps the decision from becoming a matter of taste. The process has six steps: the brief, the long list, the language and meaning screen, the knockout search, the short list with the agent's opinion, and the decision. Each step removes names for a stated reason, and the reason is written down beside the name so that the same candidate is not reconsidered three weeks later by somebody who has forgotten why it was dropped. The whole process takes a few weeks in a small business, and most of that time is spent waiting for the agent rather than working on the list.

The brief is one page and answers six questions. What is the name for: the business, an offer, a tier or a service line? Who will say it: an owner-operator, a property manager, a supplier, a search engine? In which languages must it work, and which one comes first? What must it feel like, in the words of the personality chosen in Chapter 2? What must it not be confused with? And on what date is the decision taken? A brief that answers those six questions turns the long list from a brainstorm into a piece of work, because every candidate can be tested against the same page.

The long list is written rather than collected, and it is written from the material the business already has. The case followed through this book produced twelve candidates for its entry offer, and they came from four sources: the words customers had used about the business in the twenty interviews of Book A6, the promise and the values of Chapter 2, the words the technicians already used on site, and the plain descriptions that appeared in the proposals. A long list of twelve is a workable size. A list of sixty is a list nobody screens properly, and a list of three is a decision that has already been taken.

The language and meaning screen is the first filter, and it is performed by the business rather than by a professional. Each candidate is read aloud in French and in English, written down, and read again by somebody who has not seen the list. A name that has to be spelled out over the telephone fails. A name that means something unfortunate in either language fails. A name that cannot be typed from hearing it fails. In the case, seven of the twelve candidates passed this screen, and the five that did not were dropped for reasons that were written beside them: two were hard to say in English, two carried an unwanted second meaning, and one was too close to the name of a supplier.

The knockout search is the second filter, and it is described in full in section 3.4. It is a search of the official databases for the exact name and for close spellings, and it removes candidates that are already taken. In the case, four of the seven survivors passed, and the three that did not were dropped because a close mark existed in the same or an adjacent class of goods and services. The knockout search is not a clearance search and it does not clear a name; it removes the names that are plainly unavailable, so that the agent's time is spent on the ones that have a chance.

The short list is what reaches the owner, and it is short in the literal sense: three or four names, each with its screens recorded, its domain situation and its handles. The owner chooses, and the choice is recorded with the reason, because the reason is what allows the decision to be explained to the team and revisited if the agent's opinion changes the picture. The agent reads the short list and gives an opinion on registrability and on the risk of confusion, and that opinion is the last input before the decision. The case's funnel is drawn in Figure 3.1.

Figure 3.1 shows the funnel of the case's naming project for its entry offer. It begins at the top with twelve candidate names, narrows to seven after the language and meaning screen, narrows again to four after the knockout search of the trademark databases, and ends at the bottom with one chosen name, « Bilan Nordec » in French and "Nordec Building Check" in English. Beside each narrowing, the figure names the screen that removed the candidates and the reason the screen exists.

Figure 3.1 — The naming funnel for Nordec's entry offer. Each screen removes names for a stated reason, and the legal check comes before anyone falls in love with a name.

KEY CONCEPT Check the name before you love it

The order of the work in a naming project is the whole of the method, and the order is this: screen first, fall in love second. A name that a team has already chosen, printed on a mock-up and defended in a meeting is a name that will be kept even when the search finds a close mark, the domain is taken and the French reading is unfortunate. The screens are cheap and fast — a language and meaning check, a knockout search of the official databases, a domain and handle check — and they cost an afternoon. The attachment they prevent costs a rebrand. The discipline is to run every screen on the long list before the short list is written, so that the names reaching the owner are names that have already survived something. The case followed through this book ran the screens in that order, and the funnel it produced is the subject of Figure 3.1.

3.4 Screening: Language, Meaning and Availability

The screens in this section are the part of the naming work that a small business performs itself, and they are the part that most often gets skipped. They are cheap, they are fast, and they remove names before anybody has become attached to one. A screen is not a legal opinion and it does not clear a name; it removes the candidates that are plainly unusable, so that the professional opinion that follows is spent on names that have a chance.

The language screen begins with the ear rather than the eye. Each candidate is read aloud in French and in English, and the person reading it is asked to write it down from hearing it alone. A name that cannot be written down from hearing it will be misspelled by every customer who tries to find the business, and it will be misheard by every dispatcher who takes a call. The screen then reads the name in both languages for meaning, including the meanings that come from the parts of the word rather than from the word itself. A name built from a root that means one thing in French and another in English is a name that will be explained twice. In the case, seven of the twelve candidates passed this screen, and the reasons the other five failed were written beside them.

The meaning screen also looks for unwanted associations, and it looks for them in the places a customer will look. A name that resembles a competitor's, a name that resembles a word in a trade the business does not serve, and a name that resembles a place the business does not operate in are all names that will produce confusion rather than recognition. The screen is performed by reading the candidate beside the names of the businesses the customer already knows, which in the case are the independent trades and the regional integrator. A name that sits too close to either of them is a name that will be remembered as the wrong business.

The availability screen has three parts, and they are performed in the same afternoon. The first is the domain name: the address the business would want, in the form it would want it, and the fallback if that address is taken. The second is the social handles, because a name that cannot be reached on the platforms where the business will appear is a name that will be found on somebody else's page. The third is the knockout search of the official trademark databases, described in the method callout below. None of the three parts is conclusive on its own, and together they remove most of the candidates that would have caused a problem later.

A business that may sell in the United States needs a wider search than a knockout, and the advice of the United States Patent and Trademark Office is specific about what that search covers. The office does not search for the applicant, and it recommends a clearance search of its own database of registered and pending marks, of state registries, of use on the internet, where rights can arise from use alone, and of international sources, in order to avoid refusals, oppositions and infringement suits (United States Patent and Trademark Office, 2023). The four layers are the reason a knockout search is described as a screen rather than a clearance: it reads one of the four, and the other three are the agent's work.

The screens produce a file rather than a verdict, and the file is what the agent reads. It holds the candidate, the date of each search, the address of each search, what was found and the reason the candidate was kept or dropped. A file of that kind takes an hour to build and it is the difference between a decision that can be explained and a decision that can only be defended. It is also the record that a business will want if it is ever asked why it chose the name it chose, and it is the first document in the trademark file that section 3.5 describes.

METHOD A knockout search in one hour

First, search the Canadian Trademarks Database for the exact name and for close spellings, including the forms that drop or double a letter and the forms that change one vowel. Second, repeat the search inside the classes of goods or services the business actually sells, because a mark that is close in an unrelated class is a different question from a mark that is close in the same one. Third, search the Québec business register and run a general web search for the name, because rights in a name can arise from use without any registration at all. Fourth, check the domain names the business would want and the social handles it would use, since a name that cannot be reached is a name that will be found by somebody else's page. Fifth, write down what was found, with the date and the address of each search, and give the file to a trademark agent, whose opinion is the only one that counts. A knockout search removes names; it does not clear them.

TOOLS OF THE TRADE What a naming project needs

A naming project needs five functions and no more. The first is access to the official trademark databases, in Canada and in any country where the business intends to sell, because those are the registers that decide whether a name is available. The second is the provincial business register, where a name that is already in use without a trademark registration will appear. The third is a domain registrar's search, which answers in a minute whether the address the business wants is reachable. The fourth is a spreadsheet of candidates, one row per name, with columns for the language screen, the meaning screen, the knockout result, the domain, the handles and the reason the name was kept or dropped. The fifth is a written brief, one page, stating what the name must do, who will say it, in which languages and for which offer. No brand name belongs on this list; the requirement is the function.

3.5 Trademarks: Search, Registration and Protection

A trademark is the form of a name that the law can protect, and the protection is what allows a business to stop another from trading on the recognition it has built. In Canada, the Trademarks Act, R.S.C. 1985, c. T-13, sets out how an application is made and how long a registration lasts. Section 30(1) provides that a person may apply to register a trademark for goods or services if they use it or propose to use it and are entitled to use it in Canada. Section 30(2)(a) requires that the goods or services be stated in ordinary commercial terms, and section 30(3) requires that they be grouped by class of the Nice Classification, each group preceded by its class number. Section 46(1) provides that a registration lasts ten years from the date of registration and is renewable for further periods of ten years.

The official fees are published by the Canadian Intellectual Property Office and they are charged by class. In 2026, the online filing fee is $491.06 for the first class and $149.04 for each additional class, and those amounts rise on January 1, 2027, to $499.41 and $151.57 (Canadian Intellectual Property Office, 2026a). The online renewal of a registration in 2026 is $595.06 for the first class and $185.49 for each additional class, rising in 2027 to $605.18 and $188.64, and there is no separate registration fee for applications filed after June 17, 2019 (Canadian Intellectual Property Office, 2026a). The class matters because the fee is charged by class, so a business that writes its goods and services loosely pays for classes it does not need.

The wait between filing and the first examination is published as a forecast rather than as a service standard. For an application filed in October 2026, the forecast wait to examination is about 8.2 months (Canadian Intellectual Property Office, 2026b). The same source notes that the Canadian Trademarks Database can be searched free of charge, in a basic and an advanced search, which is what makes the knockout search of section 3.4 possible without any cost at all (Canadian Intellectual Property Office, 2026b). A business that plans to launch a name on a sign in six weeks should know that the examination will not have happened by then, and that the application is what gives the priority date rather than the registration.

A business that intends to sell outside Canada has a second route, and it begins at home. Canada became a Contracting Party to the Madrid Protocol on June 17, 2019, and the amendments to the Trademarks Act that made the change possible received royal assent on June 19, 2014 (Canadian Intellectual Property Office, 2019). An international application under the Protocol requires a Canadian application or registration first, so the home filing is not optional and cannot be skipped in favor of a single international route. A business that intends to sell in more than one country therefore begins at home, and the date of the Canadian application becomes the date on which the clock starts.

In the United States, the Lanham Act, 15 U.S.C. § 1051, provides two bases for an application. Under § 1051(a), the owner of a mark used in commerce may apply; under § 1051(b), an application may rest on a bona fide intention to use the mark. Where an application is made on the intention to use, § 1051(d) provides that after a notice of allowance a statement of use is due within six months, extendable by six months and then by further extensions totalling no more than 24 months for good cause. The book states no fee and no examination time for the United States, and a business that sells there should take advice from a trademark agent rather than from a chapter.

The following table sets out the steps of a trademark application in Canada and in the United States side by side. Read it as a comparison of what each system asks for at each step, and note where the book states a rule for one country and not for the other, because a comparison that fills a gap with a guess is worse than a comparison that leaves it open.

Table 3.2 — Registering a trademark in Canada and the United States

A trademark agent assesses registrability and the risk of confusion, and that judgment is the part of the work a search cannot perform. The agent reads the search results, considers the goods and services the business actually sells, advises on the classes to name and files the application. The book explains these provisions and does not replace the advice of a lawyer or a trademark agent. A business that files without an agent is not doing anything illegal, and it is taking on a judgment about confusion that it is not equipped to make, which is the reason the case followed through this book paid an agent's fee for a single-class application.

THE RULE An application groups goods and services by Nice class

The Trademarks Act, R.S.C. 1985, c. T-13, s. 30(3), requires that the goods or services named in an application be grouped by class of the Nice Classification, each group preceded by its class number, and s. 30(2)(a) requires that they be stated in ordinary commercial terms. The same Act, at s. 46(1), provides that a registration lasts ten years from the date of registration and is renewable for further periods of ten years. The two provisions explain why the class matters to a small business: the class is what the application is about, and the fee is charged by class, so a business that writes its goods and services loosely pays for classes it does not need. The book explains these provisions and does not replace the advice of a lawyer or a trademark agent, who assesses registrability and the risk of confusion.

WHAT CHANGED Canada joins the Madrid Protocol

Canada became a Contracting Party to the Madrid Protocol on June 17, 2019, and the amendments to the Trademarks Act that made the change possible received royal assent on June 19, 2014 (Canadian Intellectual Property Office, 2019). The practical effect for a small business is a rule about the order of the work: an international application under the Protocol requires a Canadian application or registration first, so the home filing is not optional and cannot be skipped in favor of a single international route. A business that intends to sell in more than one country therefore begins at home, and the date of the Canadian application becomes the date on which the clock starts. The change is worth knowing even for a business that never files abroad, because it explains why the Canadian application is the first step in every sequence rather than one option among several.

DO THE MATH What a Canadian application costs Nordec

Illustrative figures; the case is fictional. Step 1 — The online filing fee for the first class of goods or services in 2026 is $491.06 (Canadian Intellectual Property Office, 2026a). Step 2 — Each additional class costs $149.04, and Nordec files in one class only, Nice class 37 for building maintenance and repair services, so it pays nothing for additional classes. Step 3 — The trademark agent's fee is $1,200, an illustrative figure for the case and not a published price. Step 4 — The total for the application is therefore 491.06 + 1,200 = $1,691.06. Step 5 — The official fees rise on January 1, 2027, to $499.41 for the first class and $151.57 for each additional class, and the online renewal of one class after ten years costs $595.06 at the 2026 rate, or $605.18 at the 2027 rate.

3.6 Naming in Québec: The Charter of the French Language

In Québec, a name is not only a branding decision. The Charter of the French language, CQLR c C-11, requires French in commerce and business in Québec, including commercial documents, signs and product inscriptions, and its rules on trademarks, signs, products and business names are stated in this book through the explanations of the Office québécois de la langue française and of the Registraire des entreprises. The book explains those rules and does not give legal advice, and a business that operates in Québec should read the two sources directly before it prints anything.

The Registraire des entreprises sets out how a business name is composed. A business name's generic term must be French, and it is mandatory when the specific element is in another language; where the generic is the same in both languages it comes first, so the guide's own example is "TRANSPORT FASTER" and not "FASTER TRANSPORT" (Registraire des entreprises, 2025). A business may declare versions of its name in other languages. A trademark in another language may be declared as another name without a French generic only if it is a Canadian trademark registered with the Canadian Intellectual Property Office, with no French version, shown with its symbol. The Registraire is not the authority on signs; the Office québécois de la langue française is, and the guide is information rather than a legal interpretation (Registraire des entreprises, 2025).

The rules for signs changed on June 1, 2025, under the amended Regulation respecting the language of commerce and business published on June 26, 2024. On a sign visible from outside a premises, a recognized trademark wholly or partly in another language, and a business name with a non-French specific element, must be accompanied by French terms — a generic term, a description or a slogan — so that French is markedly predominant, which means a much greater visual impact, at least twice the space, permanent, and at least as visible and legible; if a French version of the trademark is registered, it must be displayed (Office québécois de la langue française, 2025a). The rule is about the sign a passer-by sees, and it applies to the depot sign as much as to a storefront.

The rules for products also took effect on June 1, 2025. Since that date, the generic or descriptive words inside a trademark on a product — its nature, its ingredients, its color, its scent — must appear in French, while the product and company names may stay in another language; until June 1, 2027, some products made before June 1, 2025 may still be sold (Office québécois de la langue française, 2025b). A service business that hands a customer a printed item is not selling a product in the sense of this rule, and a business that sells a physical good under a trademark should read the rule before it prints a label.

For the case followed through this book, the rules describe the arrangement the business already has. The legal name is Nordec Entretien inc., in which "Entretien" is the French generic term and "Nordec" is the specific element in another language, which is the composition the Registraire's guide sets out. The tiers take French names first — Essentiel, Préventif, Portefeuille — with Essential, Preventive and Portfolio in English, and the entry offer is « Bilan Nordec » in French and "Nordec Building Check" in English. The depot sign carries the name with French terms beside it, and the vans carry the same arrangement, which is what the markedly predominant rule asks for. None of this was a legal question the business had to answer on its own; it was a design constraint that the naming decision had to respect from the beginning.

The practical lesson is that a name is chosen once and displayed in several regimes. A business that sells only in Québec chooses a name that works under the Charter. A business that sells in Québec and elsewhere chooses a name that works under the Charter and in the other market, which is why the case's names exist in two versions rather than one. A business that sells only outside Québec still needs to know the rules if it has a sign, a product or a document that crosses the border, and it needs to know them before the sign is made rather than after.

QUÉBEC & CANADA NOTE A French generic term in the business name

In Québec, a business name's generic term must be French, and it is mandatory when the specific element is in another language; where the generic is the same in both languages it comes first, so the guide's own example is "TRANSPORT FASTER" and not "FASTER TRANSPORT" (Registraire des entreprises, 2025). A business may declare versions of its name in other languages, and a trademark in another language may be declared as another name without a French generic only if it is a Canadian trademark registered with the Canadian Intellectual Property Office, has no French version and is shown with its symbol. The Registraire is not the authority on signs; the Office québécois de la langue française is. The guide is information and not a legal interpretation, and this book explains the rules rather than giving legal advice. For Nordec Entretien inc., the generic term "Entretien" is French and the specific element "Nordec" is not, which is the arrangement the rules describe.

3.7 Voice, Tone and Messaging

A name is spoken, and a voice is what surrounds it in writing. The distinction between voice and tone is the first thing to fix, because a business that confuses the two ends up either sounding the same in every situation, which is cold, or sounding different in every situation, which is unrecognizable. Voice is constant: it is the personality of the writing, chosen once and applied everywhere. Tone is the adjustment of that voice to the reader and the moment: the same voice in a cheerful announcement and in a message about a missed appointment, at a different temperature. Voice is the rule; tone is the application of the rule.

The case followed through this book wrote its voice as four principles, and each principle is a rule about what a sentence must contain rather than a description of how it should feel. The first is say the time: every message that concerns a visit states when it will happen or when it happened. The second is say the cost: every message that concerns work states whether it is inside the contract or outside it, and what the outside work will cost. The third is say what comes next: every message ends with the next step and who takes it. The fourth is no jargon without its plain word: a technical term is allowed when it is followed by the words that explain it. The four principles are short enough to be remembered and specific enough to be checked, which is what separates a voice from a wish.

The voice is supported by a glossary of twelve words to use and to avoid, and the glossary is the part of the work that makes the principles enforceable. A word to use is a word the customer already uses: "rooftop unit" rather than a model number, "loading door" rather than a part code, "we will come back" rather than "a follow-up will be scheduled." A word to avoid is a word that hides the answer: "as per," "utilize," "at your earliest convenience," "performed as required." The glossary is twelve lines long, it fits on one page, and it is the document that a new technician reads in their first week.

The rewrite of the visit sheet shows what the four principles do to a sentence. The old line read "Preventive maintenance performed as per contract scope; anomalies noted." It contains no time, no cost, no next step and two pieces of jargon, and it could have been written by any competitor in the region. The new line reads "We checked the rooftop units and the loading doors today. One door spring is worn: we will replace it on the next visit, within the contract, unless you call us first." The new line states when the work happened, states that the replacement is inside the contract, states what comes next and who decides, and uses no word that needs a glossary. The service is identical in both versions; only the writing changed.

Key messages are the third layer, and they are written by audience rather than by channel. An owner-operator of a single building asks what is wrong with the building and what it will cost, so the message for that reader leads with the finding and the price. A property manager who runs several buildings asks what is happening across the portfolio and what needs a decision this quarter, so the message for that reader leads with the portfolio and the decision. The two messages come from the same promise and the same four principles, and they differ in what they put first. The case's one-page monthly report and its consolidated portfolio report are the same voice applied to those two readers.

The voice is then applied to every touchpoint, and the list is longer than most businesses expect. In the case it covers the visit sheet, the one-page monthly report, the capital planning note, the consolidated portfolio report, the proposal, the invoice, the Bilan Nordec report, the letter, the email signature, the slide and the social media post. Eleven templates, one voice. The work of applying it is not writing eleven documents; it is writing the four principles and the twelve glossary lines and then rewriting each template once. Book B6 (Copywriting and Persuasive Communication) treats the craft of the sentences themselves, and this chapter treats the rules those sentences must obey.

The voice is also the cheapest part of the brand to fix, and that is worth stating plainly. A new logo costs a design fee, a new palette costs a rollout, and a new set of typefaces costs a license and a re-layout of every document. A voice costs an afternoon and a decision, and it changes every sentence the business writes from that day forward. A business that has no budget for a visual identity and a real need to be recognized has one lever it can pull for nothing, and the voice is that lever.

WHO DOES THIS The owner chooses, the agent clears

The naming work divides along a line that is easy to state and often crossed. The owner, or the small team around the owner, writes the brief, produces the long list, runs the language and meaning screen and the knockout search, and chooses the name from the short list. The trademark agent does the part that requires a professional judgment: reading the search results, assessing registrability, assessing the risk of confusion with an existing mark, advising on the classes to name and filing the application. A designer is not asked to clear a name, and an agent is not asked to choose one. In the case followed through this book, the owner and the administrator produced twelve candidates and ran the screens, and the agent read the four survivors and filed the application for the word mark NORDEC in Nice class 37. The division is not a hierarchy; it is a division between a decision about the business and an opinion about the law.

3.8 The Case's Names and Voice

The case followed through this book entered this chapter with a name it had chosen in 2022 without a process and an offer that had no names at all. The refresh of Book A6 had produced three tiers, two add-ons, a guarantee and a fixed-price assessment, and none of them had a name in French, a template or a voice. The naming work therefore had two jobs: to name the entry offer and the tiers, and to write the voice that would carry them. The name Nordec itself was kept, and the reason is the one given in section 3.1: it had four years of repetitions behind it and no reason to move.

The entry offer was named through the funnel of section 3.3. Twelve candidates were written from the words customers had used in the twenty interviews of Book A6, from the promise and the values of Chapter 2, from the words the technicians used on site and from the plain descriptions in the proposals. Seven passed the language and meaning screen because they worked in French and in English, and the five that failed were dropped for reasons written beside them. Four passed a knockout search of the trademark databases, and the three that failed were dropped because a close mark existed in the same or an adjacent class. The choice was « Bilan Nordec » in French and "Nordec Building Check" in English, and the reason it won was that "bilan" is the word a property manager already uses for a statement of condition, while "check" is the word an owner-operator uses for an inspection.

The tiers took French names first — Essentiel, Préventif, Portefeuille — with Essential, Preventive and Portfolio in English, and the order matters. A business that names its tiers in English first and translates them afterwards ends with names that sound like translations, and a business that names them in French first ends with names that work in both languages because the French version was the one that had to be right. The same order was applied to the entry offer, and the same order is what the Registraire's rules describe for a business name in Québec.

The knockout search found two registered marks close to "Nordec" in unrelated classes of goods, which is a result that changes the shape of an application rather than the choice of a name. A close mark in an unrelated class is a different question from a close mark in the same class, and the answer is to file in the class the business actually sells in rather than as broadly as possible. Nordec filed an application for the word mark NORDEC in Nice class 37, for building maintenance and repair services, through a trademark agent, at an illustrative agent's fee of $1,200 on top of the official fee of section 3.5. The application gives the business a priority date and a registered right if it proceeds to registration, and it does not by itself stop another business from using a similar name in a different class.

The voice was rolled out on the visit sheet first, because the visit sheet is the document the customer reads most often. The old line and the new line are quoted in section 3.7, and the difference between them is the whole of the voice work. The four principles — say the time, say the cost, say what comes next, no jargon without its plain word — were applied to the eleven templates of Chapter 5, and the glossary of twelve words was written on one page and given to every technician. The administrator, who becomes the brand keeper in Chapter 5, holds the glossary and checks every new material against it, and Sophie approves any change to the promise or to the identifiers.

The result is a set of names that can be said in both languages, found in a search and registered, and a voice that makes every sentence sound like the same business. The names cost an afternoon of screening and a single-class application; the voice cost an afternoon and a decision. Neither of them required a design budget, and both of them will still be in use when the palette of Chapter 4 has been replaced. That is the argument of this chapter in one sentence: the words are the part of the brand that lasts, and they are the part that costs the least to get right.

CASE IN POINT Nordec Entretien inc.: naming the Building Check

Nordec Entretien inc. is a fictional business created for this encyclopedia; its owner, its employees, its customers, its competitors, its figures and the events described are invented, and any resemblance to a real business is coincidental. The entry offer had no name, which meant that the fixed-price assessment at $950 was described differently by every person who sold it. Twelve candidates were written, seven passed the language and meaning screen because they worked in French and in English, and four passed a knockout search of the trademark databases. The choice was « Bilan Nordec » in French and "Nordec Building Check" in English, and the reason it won was that "bilan" is the word a property manager already uses for a statement of condition, while "check" is the word an owner-operator uses for an inspection. The tiers took French names first — Essentiel, Préventif, Portefeuille — with Essential, Preventive and Portfolio in English. The knockout search had found two registered marks close to "Nordec" in unrelated classes of goods, which is why the application was filed in Nice class 37 for building maintenance and repair services rather than more broadly.

IN LA MACHINE Names, voice and what the platform does not check

The Branding page at business/branding offers a field for a brand name and a field for a tagline, alongside a logo drop zone, a website address, a primary and a secondary color and a choice among five fonts (La Machine, 2026). It computes the contrast of the primary color and shows a pass or fail badge at the 4.5:1 threshold, and it previews a client invitation card and a client email. Its values are not saved: the "Save branding" button shows a confirmation and no other page uses what was entered, so the page is a place to try a palette and see its contrast rather than a brand register. The Templates page at business/marketing/templates stores message templates by channel with variables and a preview, which is where a voice can be applied and repeated, but it holds no brand colors, logo or footer. La Machine has no trademark search and no naming tool, and it does not check a name against any register.

Costly Mistakes

• Choosing a name before checking it. A name that a team has already printed on a mock-up is a name that will be kept even when the search finds a close mark, the domain is taken and the French reading is unfortunate.
• Reaching for a descriptive name. A name that says what the business does is understood in one second, is likely to be hard to register, cannot be found in a search and cannot grow when the offer changes.
• Skipping the language screen. A name that has to be spelled out over the telephone or that carries an unwanted second meaning in one of the two languages loses referrals silently and is discovered by a customer rather than by the business.
• Treating a knockout search as a clearance. A knockout search reads one of the four layers the United States Patent and Trademark Office recommends, and the other three — state registries, use on the internet and international sources — are the agent's work (United States Patent and Trademark Office, 2023).
• Filing in classes the business does not need. The official fee is charged by class, so an application that names goods and services loosely pays for classes it will never use and invites a question about what the business actually sells.
• Writing a voice that nobody can follow. Four principles and a twelve-line glossary can be remembered and checked; a paragraph describing how the business should feel cannot, and it will be ignored by the second person who has to write a customer email.
• Leaving the name out of the touchpoints. A name that appears on the van and not on the visit sheet, the report and the invoice is a name that the customer sees once and reads nowhere, which is the opposite of the repetition that makes a name memorable.

Key Takeaways

• A name is the most repeated element of a brand, spoken in every referral, typed in every search and printed on every document, and it is the one element that cannot be repaired cheaply once it has been chosen.
• The six kinds of names — descriptive, suggestive, invented, founder's name, acronym and place name — trade understanding against protection, and a name that suggests a benefit improves recall of claims consistent with that benefit while lowering recall of other claims (Keller et al., 1998).
• A naming project is a funnel: a one-page brief, a long list of about twelve candidates, a language and meaning screen, a knockout search of the official databases, a short list with the agent's opinion, and a decision recorded with its reason.
• In Canada, an application may rest on use or proposed use and groups the goods and services by class of the Nice Classification, and a registration lasts ten years and is renewable; the 2026 online filing fee is $491.06 for the first class and $149.04 for each additional class (Canadian Intellectual Property Office, 2026a).
• In Québec, a business name's generic term must be French and is mandatory when the specific element is in another language, and since June 1, 2025 a sign visible from outside must carry French terms so that French is markedly predominant (Registraire des entreprises, 2025; Office québécois de la langue française, 2025a).
• Voice is constant and tone adapts to the situation; four principles and a twelve-line glossary are enough to make a visit sheet, a monthly report and a proposal sound like the same business, and they cost an afternoon rather than a design budget.

Your Action Plan

1. Write the one-page naming brief. State what the name is for, who will say it, in which languages it must work, what it must feel like, what it must not be confused with, and the date of the decision. Deliverable: a one-page brief signed by the person who will decide. Verification: every candidate on the long list can be tested against the same six questions.
2. Write a long list of about twelve candidates and screen them for language and meaning. Read each one aloud in French and in English, write it down from hearing it alone, and record the reason beside every name that is dropped. Deliverable: a screened long list with a written reason for each removal. Verification: a person who has not seen the list can write down the survivors from hearing them.
3. Run the knockout search and record the file. Search the Canadian Trademarks Database for the exact name and close spellings, repeat the search in the classes of the goods and services the business sells, check the Québec business register and the web, and check the domains and handles. Deliverable: a dated search file with the address of each search. Verification: every surviving candidate has a recorded result in each of the four checks.
4. Take the short list to a trademark agent and file in the right class. Give the agent the search file and the brief, ask for an opinion on registrability and on the risk of confusion, and name the goods and services in ordinary commercial terms in the class the business actually sells in. Deliverable: a filed application with its class and its date. Verification: the class named in the application matches the services the business invoices for.
5. Write the voice and apply it to one document. Write four principles that state what a sentence must contain, a glossary of twelve words to use and to avoid, and rewrite one customer-facing document against them. Deliverable: four principles, a twelve-line glossary and one rewritten template. Verification: a reader who does not know the business can say what it promises and what happens next after reading the rewritten document.

In La Machine

The Branding page at business/branding is the only place in the application where a name and a tagline can be entered, and it is worth understanding exactly what it does with them (La Machine, 2026). It offers a logo drop zone that accepts PNG, JPEG, SVG or WebP, a field for a brand name, a field for a tagline and a field for a website address, a primary and a secondary color with ten preset swatches, a choice among five fonts, an email footer and three social links. It computes the contrast of the primary color and shows a pass or fail badge at the 4.5:1 threshold, with a live preview of a client invitation card and a client email. Its values are not saved: the "Save branding" button shows a confirmation and no other page uses what was entered. The page is therefore a place to try a palette and see its contrast, and it is not a brand register, a naming tool or a record of the voice.

The Templates page at business/marketing/templates is where a voice can actually be applied and repeated, because it stores message templates by channel with variables and a preview (La Machine, 2026). It holds no brand colors, no logo and no footer, so a template written there carries the words and not the identity, which is enough for the voice and not enough for the visual work of Chapter 4. What the platform does not have belongs in the same breath: there is no saved brand kit that other pages use, no brand portal or guidelines publisher, no digital asset management with versions and usage rights, no brand consistency audit, no trademark search and no naming tool. A business that needs a name checked against a register, a voice written down and a set of templates kept in one place does that work outside the platform, and it uses the platform to repeat the words once they exist.

Self-Check

1. Why is a name treated before the logo, the palette and the typefaces, and what does the law protect that nothing else in the identity can be protected by?
2. What are the six kinds of names, and what trade-off does a suggestive name carry?
3. What are the six steps of a naming project, and what does each screen remove?
4. What does a knockout search cover, and what does the United States Patent and Trademark Office recommend that a knockout search does not cover?
5. What are the four voice principles of the case, and what does the rewrite of the visit sheet show?

Answers. 1. A name is treated first because it is the most repeated element of the brand — spoken in every referral, typed in every search and printed on every document — and because it is the only element that can be registered as a trademark, which gives the owner control of its use for the goods and services named in the registration; a logo can be redrawn and a palette replaced, but a changed name gives up the recognition it has accumulated. 2. The six kinds are descriptive, suggestive, invented, founder's name, acronym and place name; a suggestive name improves recall of advertised claims consistent with the benefit it suggests but lowers recall of claims about other benefits (Keller et al., 1998). 3. The six steps are the brief, the long list, the language and meaning screen, the knockout search, the short list with the agent's opinion, and the decision; the language screen removes names that cannot be said or written down in both languages or that carry an unwanted meaning, and the knockout search removes names that are already taken in the official databases. 4. A knockout search covers the exact name and close spellings in the official trademark databases, the classes of the goods and services the business sells, the provincial business register and the web, and the domains and handles; the United States Patent and Trademark Office recommends a clearance search of its own database of registered and pending marks, of state registries, of use on the internet, where rights can arise from use alone, and of international sources (United States Patent and Trademark Office, 2023). 5. The four principles are say the time, say the cost, say what comes next, and no jargon without its plain word; the rewrite shows that the old line "Preventive maintenance performed as per contract scope; anomalies noted" contained no time, no cost, no next step and two pieces of jargon, while the new line states when the work happened, that the replacement is inside the contract, what comes next and who decides.

Chapter References

Canadian Intellectual Property Office. (2019). The Madrid Protocol in Canada: A summary of the general principles. https://ised-isde.canada.ca/site/canadian-intellectual-property-office/en/trademarks/practice-notices/madrid-protocol-canada-summary-general-principles

Canadian Intellectual Property Office. (2026a). Fees for trademarks (page modified July 31, 2026). Retrieved September 25, 2026, from https://ised-isde.canada.ca/site/canadian-intellectual-property-office/en/trademarks/fees-trademarks

Canadian Intellectual Property Office. (2026b). Trademarks (page modified October 5, 2026). Retrieved October 29, 2026, from https://ised-isde.canada.ca/site/canadian-intellectual-property-office/en/trademarks

Charter of the French language, CQLR c C-11 (Québec). LégisQuébec, updated to June 10, 2026 — as read for Book C2 (Websites, Landing Pages and Conversion); its rules on trademarks, signs, products and business names are stated in this book through the explanations of the Office québécois de la langue française and of the Registraire des entreprises.

Keller, K. L., Heckler, S. E., & Houston, M. J. (1998). The effects of brand name suggestiveness on advertising recall. Journal of Marketing, 62(1), 48–57. https://doi.org/10.1177/002224299806200105

La Machine. (2026). Documentation — route reference inventory. Internal product documentation, read September 25, 2026, together with the code of the pages this book describes: business/branding and business/marketing/templates.

Lanham Act, 15 U.S.C. § 1051 (United States). Legal Information Institute, Cornell Law School — read September 25, 2026.

Lowrey, T. M., & Shrum, L. J. (2007). Phonetic symbolism and brand name preference. Journal of Consumer Research, 34(3), 406–414. https://doi.org/10.1086/518530

Office québécois de la langue française. (2025a). Affichage des marques de commerce et des noms d'entreprises (updated August 27, 2025). Retrieved September 25, 2026, from https://www.oqlf.gouv.qc.ca/francisation/entreprises/affichage-marques-noms.html

Office québécois de la langue française. (2025b). Marque de commerce sur les produits (updated August 27, 2025). Retrieved September 25, 2026, from https://www.oqlf.gouv.qc.ca/francisation/entreprises/marque-commerce-produits.html

Registraire des entreprises. (2025). Les noms d'entreprises au Québec. Gouvernement du Québec. Retrieved September 25, 2026, from https://cdn-contenu.quebec.ca/cdn-contenu/adm/min/emploi-solidarite-sociale/registraire_entreprises/IN-531_REQ_noms_ENTR_QC.pdf

Trademarks Act, R.S.C. 1985, c. T-13, ss. 30 and 46 (Canada). Justice Laws Website, current to September 3, 2026 (last amended April 1, 2025) — read September 25, 2026.

United States Patent and Trademark Office. (2023). Comprehensive clearance search for similar trademarks (last modified November 30, 2023). Retrieved September 25, 2026, from https://www.uspto.gov/trademarks/search/comprehensive-clearance-search-similar-trademarks

Chapter 4 — Visual Identity

A visual identity is a small set of signs — a logo and its variants, colors, typefaces, imagery and icons — chosen to express the brand platform and to be recognized, used the same way everywhere, legible for everyone, and owned by the business.

What You'll Be Able to Do

• Describe what a visual identity does for a business — recognition, differentiation and meaning — and how the five phases of an identity process fit together.
• Build a logo system rather than a single logo, with the lock-ups, the one-color and reversed versions, the clear space and the minimum sizes that a small business actually needs.
• Choose a palette of one primary color, a small number of accents and a neutral or two, and write down the codes that print and screen work require.
• Select one or two typefaces for reasons that can be stated, and confirm that the license covers every use the business will make of them.
• Test every text-and-background pair in the palette against the contrast thresholds of the Web Content Accessibility Guidelines before the palette is published anywhere.
• Commission design work with a written brief and a signed assignment, so that the business holds the master files and the copyright in what it paid for.

Why It Matters

A brand platform is a set of sentences, and sentences are invisible until somebody gives them a form. The purpose, the promise, the values and the personality that Chapter 2 assembles are read by the owner, by the seven people who work in the business and by a handful of customers who ask. Everybody else meets the brand through signs: a van in traffic, a logo on a report, a blue that appears on an invoice and then on a website, a photograph of a technician on a roof. The visual identity is the part of the brand that travels without its author, and it either says the same thing every time or it says something slightly different every time.

The cost of saying something slightly different every time is measurable, and the case followed through this book measured it. An inventory of Nordec's 36 touchpoints found 3 versions of the logo in circulation, 5 different blues, 4 typefaces, and only 13 of 36 touchpoints consistent with one another, which is 36.1%. Nothing in that list is a disaster on its own. A second blue on a letterhead is not a business problem. But the same business that cannot keep one blue on its own documents is a business whose customers have to work to recognize it, and recognition is the cheapest thing a small business can buy. The regional integrator that competes with Nordec does not have a better logo; it has one logo, on every truck, every uniform and every report, and that uniformity is what makes a large firm look large.

There is a second reason the visual identity deserves a chapter of its own, and it is about ownership rather than appearance. A logo, a palette and a set of typefaces are assets of the business in the plainest sense: they cost money, they last for years, they appear on everything, and they can be lost. A business that commissions a logo without a written assignment may not own it. A business that uses a typeface without a license that covers the intended use may not be able to reproduce its own materials. A business whose palette fails a contrast test has made a choice that some of its customers cannot read. None of these problems is visible on the day the design is approved, and all of them are expensive later. This chapter treats the visual identity as what it is: a small, durable, owned set of signs, chosen once and then used for years.

In This Chapter

Section 4.1 explains what a visual identity does and sets out the five phases of an identity process. Section 4.2 covers the logo and the system of variants that surrounds it, including what the research on logo design says about recognition. Section 4.3 covers color, from the primary to the accents and the neutrals, and what is known about color and perceived brand personality. Section 4.4 covers typography, weights, sizes and licenses. Section 4.5 covers imagery, icons and illustration, with the rights that come with a photograph. Section 4.6 covers accessibility and legibility, with the contrast thresholds and the pairs a palette must be tested against. Section 4.7 covers commissioning design and owning the result. Section 4.8 follows the case through its own identity project, from the brief to the signed contract.

4.1 What a Visual Identity Does

A visual identity is the visible half of a brand, and it has three jobs that are easy to state and hard to do at the same time. The first is recognition: a person who has seen the business once should be able to tell, at a glance and without reading, that the thing in front of them comes from the same business as the thing they saw last month. The second is differentiation: the signs should make the business distinguishable from the businesses that sell something similar, including the ones that look more polished. The third is meaning: what a customer sees should match what the business says about itself, so that the promise and the picture reinforce each other rather than pulling apart.

Recognition is not produced by a beautiful logo. It is produced by the same logo, the same blue and the same typeface arriving again and again, in places a customer already looks. Nordec's touchpoint inventory is the clearest illustration in this book. The business had a logo, and it had three versions of it; it had a blue, and it had five; it had typefaces, and it had four. Each of those variations was introduced by a person solving a small problem on a Tuesday, and each of them made the business slightly harder to recognize. The number that matters is not the number of versions but the number of consistent touchpoints: 13 of 36, or 36.1%, which means that a customer moving through the business's materials met a different-looking business more often than not.

Differentiation is the job that a small business finds hardest, because the temptation is to look like the category. A maintenance business is expected to look like a maintenance business, and the safest route is to use the symbols the category uses: a wrench, a gear, a house outline, a blue that every competitor also uses. The result is a set of businesses that are individually competent and collectively indistinguishable. The regional integrator that competes with Nordec has solved this problem by scale rather than by design: its trucks, uniforms and reports look the same everywhere, and the uniformity itself is the message. A small business cannot match that scale, and it does not have to. It can match the consistency, which costs almost nothing once the signs are chosen.

Meaning is the job that connects this chapter to the one before it. The brand platform of Chapter 2 gives Nordec a purpose — to keep commercial buildings running, with no surprises for the people who own them — and a promise, "We know your building, and we tell you ahead." The visual identity does not repeat those sentences; it gives them a form. The route the studio eventually presented keeps the letter N and draws a horizon line through it, which is a picture of the promise: the building seen a year ahead. A customer who never reads the promise still sees the horizon. That is what it means for a visual identity to carry meaning rather than to be decorated with it.

The work of getting there has a shape, and a reference work on the subject presents it in five phases: research, strategy, design, launch and governance (Wheeler & Meyerson, 2024). The same book is explicit that governance — managing the assets after the launch — is a phase of the work and not an afterthought, and it supplies checklists, tools and case studies for each phase. The five phases are worth reading slowly, because most small businesses begin at the third one and stop at the fourth, which is exactly the combination that produces a logo nobody uses correctly.

Research, in a business of seven people, is not a market study. It is the material the business already has: the twenty value interviews of Book A6, in which urgent response within 90 minutes was named in the top three by 18 of 20 people and the capital planning note by 12; the words customers used about Nordec, "on time" or "fast" 15 times, "knows my building" 11, "tells me ahead" 9, "straight with me" 7, "friendly" 5; and the touchpoint inventory that counted the 36 places where the brand appears. Strategy is the platform of Chapter 2. Design is the studio's three routes. Launch is the rollout across the touchpoints of Chapter 7. Governance is the guide, the templates and the brand keeper of Chapter 5.

A business that skips the first two phases gets a logo it cannot explain, and the explanation matters more than the logo. When a technician asks why the van is blue, the answer "the designer liked it" ends the conversation, and the next person who needs a color for a new form will choose their own. When the answer is "because blue is the calm and precise part of who we are, and amber is the part that says look here", the choice is repeatable. The visual identity is not a set of preferences; it is a set of decisions with reasons attached, and the reasons are what allow seven people to make the same decision without asking.

The scope of the identity is worth naming before the elements are examined one by one. A complete visual identity contains a logo and its variants, a palette, one or two typefaces with their weights and sizes, rules for photography and illustration, a set of icons, and the rules that hold all of it together. Chapter 5 treats the system and the guide that make those elements usable by people who are not designers. This chapter treats the elements themselves: what each one is for, what is known about it, and how to choose it without spending more than the decision is worth.

Nordec's starting point is the ordinary one. Sophie Nadeau founded the business in 2022 and had a friend draw the first logo on a template site: a blue wrench inside a circle, with "NORDEC" in capitals beneath it. It cost nothing, it was ready in a week, and it did the job for four years. What it did not do was survive growth. By the time of the refresh, the business had three tiers and an entry offer, and none of them had names in French, templates or an identity. The new offer had been designed in Book A6 and priced in Book A7, and it was arriving in front of customers wearing the visual identity of a two-person business.

The trigger for the project was not aesthetic. In the six months since the offer refresh, Nordec decided 15 proposals and signed 8 of them, which is 53.3%. Asked how they had first heard of Nordec, the prospects answered: a referral by a customer 6 times, another building of a current manager 3 times, the Building Check letter 3 times, a web search 2 times, and a Nordec van once. Only 5 of 15 knew Nordec before the first contact. Those answers describe a business that is found by word of mouth and then has to prove itself from a standing start, and a visual identity is one of the few tools that can shorten that start.

The van deserves a sentence of its own, because it is the asset a service business gets for free. Nordec has two vehicles on the road every working day, in two territories, on the north shore of Montréal. A van is a moving sign that is seen by property managers, by owner-operators and by the trades who work in the same buildings, and it is seen without being read. Two vans, seven people in uniform and 624 preventive visits a year add up to a large number of exposures that cost nothing beyond the design itself, which is why the vehicle wrap appears in the rollout budget of Chapter 7 and why the identity has to work at the size and speed at which a van is actually seen.

What a visual identity is not is worth stating plainly, because the misunderstanding is common. It is not a logo, although it contains one. It is not a matter of personal preference, although taste is involved. It is not a one-time purchase, because the signs have to be applied, maintained and occasionally repaired. And it is not a substitute for the offer: a business with a beautiful identity and a poor service will be recognized and then avoided. The identity's job is to make the business recognizable and to say something true about it, and the rest is done by the work.

DEFINITION Logo, lock-up, palette, typeface and clear space

A logo is the mark that identifies the business: a name set in type, a symbol, or the two together. A lock-up is a fixed arrangement of the parts of a logo — the symbol and the name placed in a stated relationship, at a stated distance, so that the arrangement is reproduced rather than reinvented each time. A palette is the set of colors the business uses, usually one primary color, one or two accents and one or two neutrals, each written with its codes for screen and for print. A typeface is the design of a set of letters, figures and symbols; a font is the file that carries it, and the two words are used interchangeably in most business documents. Clear space is the area around a logo that nothing else may enter, measured against a part of the logo itself so that it scales with it. Minimum size is the smallest size at which the logo may be reproduced, stated separately for print and for screen.

KEY CONCEPT A few signs, always the same

The central idea of this chapter is that a visual identity works through repetition rather than through invention, and that repetition requires a small number of signs. A business with one logo, one blue, one typeface and one photographic rule can apply them without thinking, and the result is that every touchpoint reinforces every other one. A business with three logos, five blues and four typefaces has to make a decision every time it produces anything, and each decision is an opportunity to drift. The research on distinctive brand assets treats logos, colors, fonts, characters, taglines and jingles as a field that has been studied over the long term, examining which branding elements succeed and which fail (Romaniuk, 2018). The practical reading of that work for a small business is not that any particular element is magic; it is that an element earns its place by being used consistently and by becoming linked to the brand in the customer's memory. A sign used once is decoration. A sign used a thousand times is an asset.

4.2 The Logo and Its Variants

Logos come in four broad kinds, and the choice between them has consequences that a small business feels for years. A wordmark is the name set in type, with no separate symbol; it must be read, so it needs space and it works best when the name is short. A symbol is a mark without the name; it is compact and it can be reproduced anywhere, but it has to be learned before it means anything. A combination is a symbol and a name locked together, which is the most common choice for a business that is still building recognition, because it delivers both the mark and the name in one glance. An emblem places the name inside the mark, which produces a compact and often handsome result that becomes difficult to reproduce at small sizes.

Nordec chose a combination. The chosen route keeps the letter N and draws a horizon line through it, and the name sits beside it in the humanist sans-serif. The letter is a natural form that every customer already knows, the horizon line is a single addition that carries the promise, and the two together are legible at the size of a van and at the size of a favicon. The generic wrench that had served since 2022 was dropped, and that decision is the one that made the mark distinctive rather than merely tidy.

A logo is not one file, and this is the point at which most small businesses stop too early. What a business needs is a system: a primary horizontal lock-up for the places where there is room, a stacked lock-up for the square spaces, the symbol alone for the small ones, a one-color version for the processes that cannot print two colors, and a reversed version for dark backgrounds. Each of those exists because a real situation demands it, and a business that does not have them will improvise them badly at the moment of need.

The van roof is the clearest example. A roof panel is a small, curved, awkward surface, and the horizontal lock-up with the name beside the symbol will not fit on it in a way that can be read from the ground. The symbol alone fits, and it is recognized because the same symbol appears on the report, the invoice and the website. The one-color version exists for the same reason in a different medium: an embroidered uniform, an engraved plate, a stamped document and a single-color print job cannot carry a two-color mark, and a business without a one-color version will either pay for a second color it does not need or accept a mark that looks wrong.

Clear space is the rule that keeps a logo from being crowded, and it is measured against the logo itself so that it scales automatically. Nordec's clear space is equal to the height of the letter N. Nothing enters that area: not a photograph, not a rule, not another logo, not the edge of the page. The rule sounds fussy and it is not, because the alternative is a logo that competes with whatever the layout places beside it, and a logo that competes is a logo that is not seen. The same logic applies to the minimum size, which is the smallest size at which the mark still reads: 20 millimetres in print and 24 pixels on screen for the symbol. Below that size the horizon line closes up and the letter becomes a smudge, and a smudge on a business card is worse than no mark at all.

The research on logo design gives the choice of a mark more structure than taste alone provides. A study of 195 logos rated on 13 design characteristics produced three recommendations that a small business can use directly (Henderson & Cote, 1998). Logos meant for high recognition should be very natural, very harmonious and moderately elaborate. Logos with a lower investment behind them should be less natural and very harmonious. A professional look calls for moderate elaboration and natural elements. The three recommendations are not a formula, and the study does not say that any particular logo will succeed; it says that the characteristics were rated and that the recommendations follow from the ratings.

The three characteristics are worth translating into ordinary language. Natural means that the mark contains recognizable objects and shapes rather than abstract geometry: a letter, a horizon, a building, a tool. Harmonious means that the parts belong together, that the proportions are balanced and that nothing in the mark fights anything else. Elaborate refers to the amount of detail and design content in the mark. The recommendation that a logo with a lower investment behind it should be less natural and very harmonious is the one that speaks most directly to a business of seven people, because it says that a mark which is simple, balanced and slightly abstract will hold up better than one that tries to depict the whole business in a single image.

Nordec's chosen route sits comfortably inside those recommendations. The letter N is natural, because it is a letter that every customer reads every day. The horizon line is a single, simple addition rather than a scene. The whole is harmonious, because the line is drawn at the same weight as the letter and crosses it at a point that balances the two halves. The elaboration is moderate: there is enough in the mark to be interesting and not enough to disappear when it is small. The wrench, by contrast, was natural in the wrong way: it is a recognizable object, but it is the object that every trade in the category uses, and it says "tools" without saying anything about buildings, contracts or a year ahead.

It is worth saying what the logo does not do, because the expectations placed on it are usually too high. A logo does not explain the business, persuade a customer or carry the promise on its own. It is a signature, and a signature is recognized rather than read. The identity is the signature plus everything around it: the paper, the ink, the handwriting, the photograph on the wall. A business that spends its whole design budget on the mark and nothing on the templates, the photography and the guide has bought a signature and no stationery.

Testing a logo before it is adopted is cheap and it is rarely done. The mark should be looked at small, at the size it will appear on a business card and a favicon. It should be looked at in one color, printed on an ordinary office printer. It should be looked at reversed, on the blue that will carry it. It should be placed on a photograph, on a van, on a report and on a phone screen. Each of those views answers a question that the presentation on a large white screen does not, and a mark that survives all of them is a mark that will survive the next five years.

The files matter as much as the drawing. A logo should exist as a vector master, which is a file that can be enlarged to the side of a building without losing its edges, and as a small set of exports for the places that cannot use a vector: an image file with a transparent background for documents and slides, a version for the website, and a version for the sign maker. The vector master is the asset; the exports are conveniences that can be remade from it. A business that holds only an image file has lost the ability to have its own logo reproduced properly, and it will discover this at the sign maker's counter.

The following table sets out the six versions of the logo that Nordec received, with the use each one is meant for, its minimum size and the backgrounds it may sit on. Read it as the answer to a practical question: when a person in the business needs a logo for a new material, which version do they reach for, and at what size does it stop working.

Table 4.1 — Nordec's logo system

CASE IN POINT Nordec Entretien inc.: the wrench that had to go

Nordec Entretien inc. is a fictional business created for this encyclopedia; its owner, its employees, its customers, its competitors, its figures and the events described are invented, and any resemblance to a real business is coincidental. The first logo, drawn in 2022 by a friend on a template site, was a blue wrench inside a circle with "NORDEC" in capitals. It was free, it was ready in a week, and it worked for a business of one. Four years later, the studio presented three routes, and the discussion in the second workshop turned on the wrench. The argument for keeping it was recognition: customers knew it, and it had been on two vans. The argument against it was the one that won. A wrench says that the business owns tools, which every competitor also owns; it says nothing about buildings, about contracts, about a record kept for every building or about a note that lists the equipment approaching the end of its life. The regional integrator's mark is abstract and its trucks are identical; the independent trades use tools and clip art. Nordec's differentiators are the 90-minute response, the building health record, the one-page monthly report and the capital planning note, and none of them is a wrench. The letter N with a horizon line kept the recognition the old mark had earned and dropped the part that said nothing.

4.3 Color

A palette has three jobs, and a business that gives each of them a color will find the palette easy to use. The primary color carries the brand: it is the one that appears on the van, the report header and the website, and it is the color a customer would name if asked what color the business is. The accents mark attention: they are used sparingly, for the things that must be noticed, and their scarcity is what gives them force. The neutrals carry the text and the backgrounds, and they are the colors that do most of the work while being noticed least.

Nordec's palette has five colors and white. Nordec Blue, #1F3A5F, is the primary. Signal Amber, #E8A33D, is the accent. Slate, #5B6770, is the secondary neutral for text that should not shout. Mist, #F4F6F8, is the background neutral. White is the fifth, and it is a color in the sense that matters here, because it is the background against which most of the others are read. Five colors is a small palette, and smallness is a virtue in this context: a palette that a person can remember is a palette that a person will use correctly.

Every color in the palette is written as a family of notations rather than as a single value, because the same color has to survive several journeys. A hexadecimal code such as #1F3A5F describes a color for a screen, and it is the code that a website, an email signature and a slide will use. Print work needs a different notation, because ink on paper does not behave like light from a screen, and a print code describes a mixture of inks rather than a value of light. The practical consequence is that a palette is not finished until both notations exist for every color, and a business that has only the screen codes will find that its blue shifts every time it goes to a printer.

The research on color and brand personality is worth presenting carefully, because it is easy to overstate. One article reports four studies that related hue to the dimensions of brand personality and examined saturation and value as well, and its title pairs red with excitement and blue with competence (Labrecque & Milne, 2012). What the studies support is a claim about perception: in those studies, colors were associated with particular personality dimensions, and saturation and value amplified those associations. What they do not support is a claim that a color means one thing everywhere, that a color causes a purchase, or that choosing blue will make a business seem competent to every audience. The finding is an input to a decision, not the decision.

Read that way, the finding is useful. A business that has chosen calm, precise, plain-spoken and neighborly as its personality traits has a reason to prefer a blue over a red, and the reason is that the association runs in the same direction as the platform rather than against it. The same business should not conclude that the blue is doing the work. The blue is one of a dozen signals, and the technician who arrives within 90 minutes is a stronger one. Color is a supporting argument, and supporting arguments are worth having when they cost nothing.

The five blues that were in circulation at Nordec are the illustration of what happens without a palette. Five different blues is not a palette; it is an accident with a history. Each of the five arrived with a document, a template or a supplier, and each of them was close enough to the others to look like a mistake rather than a choice. The remedy is not to choose the best of the five but to choose one and to write it down with its codes, so that the next person who needs a blue has an answer that does not require a decision.

Amber's role in the palette is the most interesting one, because it is the color that had to be constrained. Signal Amber is a warm, high-visibility color, and its natural temptation is to be used everywhere. It appears on the van stripe, on safety details, on the small marks that draw the eye to a figure in a report, and on the accent elements of the website. It does not carry text on white, for a reason that section 4.6 sets out with numbers, and that single constraint is what turned a pleasant color into a disciplined one. A palette with a rule attached is a palette that produces consistent work.

The neutrals are the part of the palette that a small business most often forgets to choose. Slate carries secondary text: captions, table labels, the small print that should be readable without competing with the body. Mist carries backgrounds: the panel behind a table, the fill of a report section, the quiet area of a web page. A business that has no neutrals will use the primary color for everything, which produces documents that are loud in the wrong places, and it will use black and gray that belong to no palette at all.

The number of colors in a palette is a decision about discipline rather than about taste. A palette of five colors and white can be applied by anybody; a palette of twelve requires a designer every time. The same logic applies to the shades within a color: a business needs one blue, not four tints of blue, unless the tints are defined as part of the system and given their own codes. Every additional color is an additional decision, and decisions made under time pressure are the ones that produce the sixth blue.

Color and the platform should be read together. Nordec's purpose is to keep commercial buildings running with no surprises for the people who own them, and its personality is calm, precise, plain-spoken and neighborly. A deep blue reads as calm and precise; a warm amber reads as attention and as a warning that is friendly rather than alarming; a slate gray reads as plain; a mist background reads as quiet. None of those readings is guaranteed, and none of them is a rule. Together they form a palette that a person can describe in one sentence, which is the test that matters: if the palette cannot be explained to the person who applies it, it will not be applied.

The trap in color selection is choosing a color because it is liked. Liking is not nothing — a palette the owner dislikes will not survive — but it is not a reason that can be repeated to seven people or written into a guide. The reasons that can be repeated are the platform, the category, the contrast thresholds and the media the color has to survive. A color chosen for those reasons can be defended in a workshop, and a color that can be defended is a color that stays.

WATCH OUT A color that fails on white

The most common palette error in a small business is not an ugly color; it is a usable-looking color that cannot carry text. A warm amber, a light green, a bright yellow and a pale gray all look pleasant on a screen and all fail the contrast threshold when they are used for words on a white background. The failure is invisible to the person who chose the color, because they can read it; it is visible to a customer reading a phone in daylight, to a person with reduced vision, and to anybody over fifty. The remedy is a test rather than an eye: every text-and-background pair the business intends to use is computed against the threshold before the palette is published, and any pair that fails is moved to a role that does not involve text. In Nordec's palette, Signal Amber on white measures 2.16 to 1, which fails even the 3 to 1 threshold for large text, so amber carries accents, the van stripe and safety details, and never words on a white page. The color was not changed. Its job was.

DO THE MATH Nordec's contrast ratios

Illustrative figures; the case is fictional. Step 1 — The contrast ratio of two colors is (L1 + 0.05) ÷ (L2 + 0.05), where L1 is the relative luminance of the lighter color and L2 that of the darker, and the result is written as a number to one. Step 2 — White text on Nordec Blue, #1F3A5F, gives 11.48:1, which passes the 4.5 to 1 threshold for body text with a wide margin, so blue panels can carry white words. Step 3 — Slate, #5B6770, on white gives 5.80:1, which passes for body text, so Slate can carry captions and table labels. Step 4 — Signal Amber, #E8A33D, on white gives 2.16:1, which fails the 4.5 to 1 threshold for body text and also fails the 3 to 1 threshold for large text, so amber never carries words on a white background. Step 5 — Nordec Blue on Signal Amber gives 5.32:1, which passes, so amber can carry blue text — which is exactly what happens on the van stripe, where the name is set in blue on an amber band.

4.4 Typography

Typography is the part of a visual identity that carries the most words and is noticed the least, and it is therefore the part where a small business can gain the most for the least money. A typeface is not a decoration applied to text; it is the shape of every letter the business writes, and it decides how a report reads before a single sentence is understood. The rule that keeps typography manageable is the same rule that governs the palette: choose few, choose once, and write down what was chosen.

Nordec uses two typefaces. The first is a humanist sans-serif, used for everything: headings, body text, the name in the logo, the numbers in a report. The second is a monospaced face, used only for building record codes, because those codes appear in columns and a monospaced face makes the columns line up without any further work. Two families is the practical maximum for a business of this size. A third family would have to earn its place by doing something the first two cannot, and in most small businesses nothing does.

The choice of a humanist sans-serif rather than a geometric one or a serif is a decision with reasons. A humanist sans-serif has letterforms with a slight variation in stroke and open, generous shapes, which makes it comfortable to read at small sizes and in long passages, and it reproduces well in the conditions a maintenance business actually meets: a printed report, a phone screen, a van panel seen at an angle. A geometric sans-serif is more even and more modern in feel, and it is often less comfortable in a long document. A serif is excellent for long documents and less suited to a van. None of those statements is a law; they are the reasons behind a choice, and the reasons are what a business writes down.

Weights and sizes are the second half of typography, and they are where consistency is won or lost. A business needs a small set of weights — regular, medium and bold is enough for most work — and a scale of sizes that repeats across documents. Nordec's scale runs from the small caption size used in table labels up to the size used for the title of a monthly report, with a handful of steps between. The value of a fixed scale is that a person producing a new document does not have to invent a hierarchy; they choose from the steps that exist, and the result looks like the other documents without anybody comparing them.

Legibility is a property of the letterforms and of the setting, and it can be described without naming a typeface. The height of the lowercase letters relative to the capitals, the openness of the counters inside letters such as e and a, the amount of space between letters and between lines, and the length of a line of text all affect how easily a passage is read. A typeface with a generous lowercase height and open counters reads well at small sizes. A line of text that runs the full width of a page is harder to read than one that runs half of it. These are the things a business can check without being a designer, and they are the things that make a report readable on a phone.

The license is the part of typography that small businesses most often overlook, and it is the part with the clearest consequences. The SIL Open Font License permits using, studying, copying, embedding, modifying and redistributing the font software, and it allows the font to be sold only when it is bundled with other software; reserved font names may not be used by modified versions without permission (SIL International, 2007). The same license states something that surprises many businesses: documents and other works made with the fonts are not covered by the license, so a logo set in such a font is not itself under the license. The logo file belongs to the business; the font file does not.

That distinction has a practical edge. A business that commissions a logo set in an open-license typeface owns the logo and can use it forever, because the license does not reach the artwork. The same business cannot claim to own the typeface, cannot redistribute the font file to a supplier as if it were its own asset, and cannot modify the font and keep its name. Those limits are mild, and they are the reason open-license typefaces are a sensible default for a small business that wants to avoid a licensing conversation every time a new supplier asks for the font.

A paid license is a different arrangement and it must cover the uses the business will actually make of the typeface. A license for desktop use, which covers a person installing the font on a computer and setting type in documents, does not automatically cover embedding the font in a website, in a mobile application or in a document that is distributed as an editable file. A vehicle wrap is a print use and belongs to the print license; a web page is a web use and belongs to the web license. The practical rule is to list the media the business uses — print, web, app, vehicle, email — and to confirm that the license covers each of them before the typeface is adopted, because changing a typeface after it has been applied to two vans and eleven templates is expensive.

The four typefaces that were in circulation at Nordec are the illustration of what happens without a decision. Each arrived with a document, a template or a supplier, and each was close enough to the others to look like an accident. The remedy is the same as for color: choose one or two, write them down with their weights and sizes, and make the substitutes explicit. A substitute is the typeface a person uses when the chosen one is unavailable, and naming it in advance prevents the fifth typeface from arriving through a document that had to be finished on a Friday.

Typography also carries tone, although it does not speak. A typeface with tight spacing and heavy weights reads as loud; one with generous spacing and light weights reads as quiet; one with unusual letterforms reads as playful. Nordec's voice principles from Chapter 3 — say the time, say the cost, say what comes next, no jargon without its plain word — describe a business that is plain and unhurried, and the typeface should not contradict them. A calm typeface applied to a calm sentence is a small reinforcement, and small reinforcements are what a visual identity is made of.

Testing typography is as cheap as testing a logo and as rarely done. A page of body text should be printed at the size it will actually be read, and read on paper and on a phone. A table should be set in the chosen typeface and checked for the alignment of figures. A heading should be set at the largest size in the scale and checked for the way it breaks across two lines. The name should be set in the typeface and looked at from across a room. Each of those tests costs minutes and prevents the discovery that a typeface which looked well on a presentation screen is uncomfortable in a monthly report that 48 buildings receive twelve times a year.

4.5 Imagery, Icons and Illustration

Imagery is the part of a visual identity that shows the work rather than describing it, and for a service business it is often the most persuasive element. A photograph of a technician on a roof, in a mechanical room, with a gauge in hand, tells a property manager more about what the business does than a page of text. The rule that makes imagery useful is the same rule that governs the rest of the identity: it should be the business's own, and it should be used the same way every time.

Nordec's photographic rule is short and strict. Photographs show real Nordec technicians at work in real buildings, in natural light, and stock photographs of smiling models are not used. The reason is the promise. A business that says "We know your building, and we tell you ahead" cannot illustrate itself with a stranger in a hard hat standing in a building the business has never entered. The photograph is a claim, and a false claim in a photograph is as damaging as a false claim in a sentence.

The practical rules that make the photographic rule work are worth writing down, because a rule without a procedure is a preference. Photographs are taken during real visits, which means the business does not need a photo shoot; it needs a person who remembers to take three pictures on a Tuesday. The people photographed are asked for their consent, and the consent covers the uses the business intends. Customer equipment labels, unit numbers and anything that identifies a building or its owner are kept out of frame, because a photograph that publishes a customer's equipment list is a privacy problem rather than a marketing asset. And the pictures are kept in one place, with the ones that have been approved separated from the ones that have not.

The volume of opportunity is the argument for the rule. Nordec's exposures in a year include 624 preventive visits, about 276 urgent calls, two vans on the road every working day and seven people in uniform. A business with that many visits does not need a photographer; it needs a habit. Three photographs a week, taken by the technicians who are already on site, produce more usable images in a year than a single commissioned shoot, and they are images of the business's own work.

Icons are the second element of imagery, and they do a specific job: they mark the categories that repeat. Nordec uses line icons of the building systems it services — HVAC, plumbing, electrical and loading doors — drawn in Nordec Blue. Line icons are chosen for three reasons. They reproduce at small sizes, which matters on a report and on a phone. They sit beside text without competing with it, because a line drawing has less visual weight than a filled shape. And they can be drawn in one color, which means they work in the palette without introducing new ones.

The icon set should be small and closed. Four icons that cover the four systems the business services are more useful than twenty icons that cover everything a building contains, because the four will be used consistently and the twenty will be used inconsistently. The style rules are equally short: a consistent stroke weight, a consistent corner treatment, a consistent size on the page, and no icon that contains text. An icon set that follows those rules looks like a set; one that does not looks like a collection.

Illustration is the third element and the one a small business can most easily do without. Illustration is useful when the business needs to show something that cannot be photographed — a process, a timeline, an abstract relationship — and it is a liability when it is used to decorate a page that a photograph would serve better. Nordec's capital planning note is the one place where a simple diagram earns its place, because it shows equipment approaching the end of its life against a replacement window, and that is a relationship rather than an object. Everywhere else, the photograph is the better choice.

Image rights and licenses are the part of imagery that carries obligations. A photograph has an owner, and the owner is usually the person who took it, which means that a picture taken by a technician belongs to the technician unless the business has an agreement that says otherwise. A photograph bought from a stock library comes with a license that states what may be done with it, and the license has an address and often an attribution requirement. The practical discipline is to record, for every image the business uses, where it came from, what the license permits and what attribution it requires, and to keep that record with the image rather than in somebody's memory.

Consent and privacy belong in the same paragraph as rights, because a photograph of a person is personal information as well as a work. A business that publishes a picture of an employee or a customer's staff has to be able to say what consent it holds and what the picture is used for. Book H2 (Privacy and Personal Information) treats the obligations in full, and the practical rule for this chapter is that a photograph of a person is not used until the business can point to the consent that covers it.

Generative image tools are a family of tools that produce images from a description, and they raise questions about rights, about disclosure and about the use of a picture that depicts something that never happened. This book treats them in Chapter 12, section 12.7, where the questions are set out in full. For the purposes of a visual identity, the position is simple: a business whose promise is that it knows the building should illustrate itself with the building, and a picture of a building that does not exist cannot support that promise.

The test for any image is a single question: does this picture show our work, our people or our buildings? A photograph that passes the test is an asset that can be used for years. A photograph that fails it is a decoration that will be replaced, and replacing images is more expensive than choosing them well once. The same test applies to icons and to illustration, and it is the reason the icon set is limited to the four systems the business actually services.

WHO DOES THIS The studio designs, the owner decides

In a small business, the visual identity is produced by an outside studio and decided by the owner, and the boundary between the two is worth drawing clearly. The studio brings the craft: the drawing, the palette, the typefaces, the templates and the guide, and it is the studio's job to present options that are genuinely different rather than three versions of the same idea. The owner brings the knowledge the studio cannot have: what customers say, what the technicians do, which words appear in the interviews and which of the business's differentiators actually win work. The owner also brings the decision, and the decision should be taken once, in a room, with the reasons written down. What the owner should not do is design. What the studio should not do is decide what the business stands for. In the case followed through this book, the studio presented three routes and the owner chose the one that kept the letter N, and the choice was made in a workshop with the seven people present rather than by email over a week.

4.6 Accessibility and Legibility

Legibility is the part of a visual identity that is not a matter of taste, because it has thresholds that can be computed. A palette that looks well on a designer's screen may be unreadable for a customer reading a phone in daylight, and the difference between the two is a number. The numbers come from the Web Content Accessibility Guidelines, and they are the most useful technical constraint a small business can apply to its identity, because they are free, they are public and they settle arguments.

The guidelines state their requirements as success criteria at three levels. Success criterion 1.4.3, at level AA, requires text to have a contrast ratio of at least 4.5 to 1 against its background, and large text — 18 points, or 14 points bold — at least 3 to 1. Text that is part of a logo or a brand name has no contrast requirement. Success criterion 1.4.6, at level AAA, raises the requirement to 7 to 1, and 4.5 to 1 for large text. Success criterion 1.4.11, at level AA, requires graphical objects and interface components needed to understand content to have a contrast of at least 3 to 1 against adjacent colors (World Wide Web Consortium, 2024). The guidelines are a W3C Recommendation, first published as a Recommendation on 5 October 2023 and current in the version of 12 December 2024.

One point of caution belongs here, and it is a legal rather than a technical one. This book explains the guidelines and does not claim that a business in Canada or Québec is under a legal obligation to meet them. The guidelines are a technical standard, widely used and widely referenced, and a business that applies them is making a choice about who can read its materials. Whether a particular business is subject to a particular legal requirement is a question for a lawyer, and it is not answered by this chapter.

The contrast ratio is computed from the relative luminance of the two colors, and the formula is (L1 + 0.05) ÷ (L2 + 0.05), where L1 is the luminance of the lighter color and L2 that of the darker. The result is a number to one, and the number is compared with the thresholds. The computation is not something a business does by hand; it is done by a contrast checker, which is a small tool that takes two color codes and returns the ratio. What matters for this chapter is not the arithmetic but the habit: every pair the business intends to use is checked before the palette is published, and the pairs that fail are given a different job.

Nordec's five pairs were checked, and the results are the reason the palette is disciplined. White on Nordec Blue measures 11.48 to 1, which passes comfortably and makes blue panels safe for white text. Nordec Blue on Mist measures 10.60 to 1, which passes and makes the report background safe for body text. Slate on white measures 5.80 to 1, which passes and makes Slate safe for captions and table labels. Nordec Blue on Signal Amber measures 5.32 to 1, which passes and makes the amber van stripe safe for blue lettering. Signal Amber on white measures 2.16 to 1, which fails both the 4.5 to 1 threshold for body text and the 3 to 1 threshold for large text, and that single failure is what turned amber into an accent color.

The consequence of the amber result is worth stating as a rule, because it is the kind of rule that prevents a mistake rather than correcting one. Amber never carries text on white. It carries accents: the van stripe, the small marks that draw the eye to a figure in a report, the safety details on a uniform, the highlight on a web page. Where amber is used as a background, the text on it is Nordec Blue, which passes at 5.32 to 1. The color was not changed and the palette was not reduced; one role was reassigned, and the reassignment is written into the guide so that it does not have to be rediscovered.

The figure below shows the palette and the five pairs together, which is the form in which the test result is actually used. Figure 4.1 presents Nordec's five colors with their names and their codes, and beside them the five text-and-background pairs with their contrast ratios of 11.48, 10.60, 5.80, 5.32 and 2.16 to 1, each marked against the 4.5 to 1 threshold for body text, with the amber-on-white pair marked "accents only, never text".

Figure 4.1 — Nordec's palette and its contrast ratios. Four pairs pass the 4.5:1 threshold for body text; Signal Amber on white does not, so amber carries accents, never text.

The following table sets out the same five pairs in a form that can be read as a working rule. Its columns give the text color, the background, the measured ratio, whether the pair passes at level AA for body text, and the use the pair is approved for. Read it as the answer to the question a person asks when they are laying out a new document: which combinations am I allowed to use, and what is each one for.

Table 4.2 — Nordec's color pairs and their contrast

Legibility is more than contrast, and the rest of it is worth a paragraph because contrast alone will not make a document readable. The size of the text matters, and a body size that is comfortable on a desktop screen is often too small on a phone. The weight matters, and a light weight that looks elegant in a heading becomes faint in a table. The length of a line matters, and a line that runs the full width of a page is harder to follow than one that runs half of it. The spacing between lines matters, and a tight leading that saves a page makes a long report tiring. None of those is a threshold with a number attached, and all of them are visible in a test that takes ten minutes: print the document, read it, and look at it on a phone.

The logo exemption deserves a careful reading, because it is often misunderstood in both directions. Text that is part of a logo or a brand name has no contrast requirement under the guidelines, which means that a logo may legitimately use a color combination that would fail if it were body text. The exemption exists because a logo is a graphic mark rather than a passage to be read. The misunderstanding runs the other way as well: the exemption applies to the logo, not to the palette. A business cannot use the exemption to justify putting amber words on a white page, because those words are text and not a logo. The palette used for text is tested, and the logo is exempt.

Testing legibility in the places the business actually uses is the last step and the cheapest one. The report should be printed and read at arm's length. The van should be looked at from twenty metres, which is the distance at which it is actually seen. The website should be opened on a phone in daylight. The uniform should be looked at from across a room. Each of those tests answers a question that a screen cannot, and each of them takes minutes. A palette that has been tested in all four places is a palette that will not have to be changed after the wraps are printed.

METHOD Testing a palette for contrast

First, list every text-and-background pair the business intends to use, including the pairs that appear in documents, on the website, on signage and on a vehicle; a pair that is not on the list is a pair nobody has tested. Second, compute the contrast ratio of each pair with the formula (L1 + 0.05) ÷ (L2 + 0.05), using a contrast checker rather than an eye. Third, compare each result with the thresholds: 4.5 to 1 for body text and 3 to 1 for large text at level AA, and 7 to 1 and 4.5 to 1 respectively at level AAA. Fourth, move every failing color to a role that does not involve text — an accent, a rule, a stripe, a safety detail — and record the new role. Fifth, write the approved pairs into the guide as a table, with the ratio beside each one, so that the next person who lays out a document chooses from a list rather than from a color picker. Deliverable: a table of approved pairs with their ratios. Pitfall: testing the palette on a screen only, which hides the failures that appear in print and in daylight.

THE RULE Contrast at level AA

The Web Content Accessibility Guidelines 2.2 state, at success criterion 1.4.3 (level AA), that text must have a contrast ratio of at least 4.5 to 1 against its background, and large text — 18 points, or 14 points bold — at least 3 to 1; text that is part of a logo or a brand name has no contrast requirement. At success criterion 1.4.11 (level AA), graphical objects and interface components needed to understand content must have a contrast of at least 3 to 1 against adjacent colors (World Wide Web Consortium, 2024). The guidelines are a W3C Recommendation, first published as a Recommendation on 5 October 2023 and current in the version of 12 December 2024. Two readings of the rule matter for a small business. The first is that the thresholds apply to the pairs the business actually uses, not to the palette as a whole, so a color that fails in one role may be perfectly usable in another. The second is that the logo exemption covers the mark and not the text around it, so a business cannot lean on it to justify unreadable body copy. This book explains the guidelines; it does not claim that a business in Canada or Québec is legally obliged to meet them.

WHAT CHANGED WCAG 2.2

The Web Content Accessibility Guidelines reached version 2.2 as a W3C Recommendation on 5 October 2023, and the current version is the Recommendation of 12 December 2024 (World Wide Web Consortium, 2024). For a small business working on a visual identity, the change that matters is not a new threshold but a settled one. The contrast requirements at 1.4.3 and 1.4.11 have been stable across versions, which means that a palette tested against them today will still be compliant in five years, and that a business can write the ratios into its guide without expecting to rewrite them. The second consequence is practical rather than technical: because the guidelines are public and the thresholds are numbers, a business can settle a disagreement about a color with a measurement rather than with an opinion. That is worth more to a small team than any particular ratio, because it turns a matter of taste into a matter of fact, and matters of fact can be delegated.

4.7 Commissioning Design and Owning the Result

Most small businesses do not employ a designer, and the visual identity is therefore a commissioned project rather than an internal one. That changes the work in two ways. The business has to be able to describe what it needs in a form a studio can price and produce, and it has to end the project owning what it paid for. Both of those are ordinary commercial skills, and both are frequently skipped, which is why this section exists.

The brief is the document that makes the project possible, and it does not need to be long. A useful brief states what the business does, who its customers are, what it promises and what makes it different; it names the competitors whose look should be avoided; it lists the places the identity will appear, from the van to the monthly report; it states the languages the identity must work in; and it says what the business wants to be true of the result. Nordec's brief was built from the platform of Chapter 2 and the touchpoint inventory of Chapter 1, and it named the two territories, the owner-operators and property managers, the 90-minute response, the building health record and the capital planning note. A studio that receives that brief can present routes that are actually different from one another.

Routes are the second element of the commission, and their number matters. A studio that presents one route is presenting a decision that has already been taken. A studio that presents three routes is presenting three genuinely different directions, each with a reason, and the business chooses between them. Nordec's studio presented three, and the choice was made in a workshop with the seven people present. The number three is not a rule from anywhere; it is the number that gives a real choice without turning the project into a comparison of ten variations of the same idea.

Revisions are the third element, and they should be stated in the agreement rather than negotiated later. A fixed number of rounds of revision, applied to the chosen route, is the arrangement that keeps a project moving and keeps the fee predictable. Nordec's studio worked on a fixed fee of $7,800 for the logo, the palette, the typography, eleven templates and the guide, with the routes presented once and the revisions applied to the chosen one. A fixed fee with a stated scope is easier for a small business to approve than an hourly arrangement, because the business knows what it is buying before it starts.

Copyright is the element that most small businesses get wrong, and the law is clear enough to state plainly. Under the Copyright Act, the author of a work is its first owner (s. 13(1)); the employer is the first owner of a work made by an employee in the course of employment, unless there is an agreement to the contrary (s. 13(3)); and an assignment or grant of copyright is valid only if it is in writing and signed by the owner of the right (s. 13(4)). Moral rights — the integrity of the work and the right to be associated with it — cannot be assigned, but they can be waived, in whole or in part, and an assignment of copyright does not by itself waive them (s. 14.1).

The practical consequence for a business that commissions a logo from an independent studio is direct. The studio is the author, and therefore the first owner, unless a signed written assignment says otherwise. A business that pays an invoice and receives the files has bought the use of the work, not the ownership of it, and the difference appears years later when the business wants to modify the mark, register it, or use it after the studio has closed. The remedy is a clause, and the clause is short: the studio assigns the copyright in the deliverables to the business, in writing and signed.

Moral rights need their own sentence because they behave differently from copyright. They belong to the author and they cannot be handed over, so a business cannot acquire them; what it can obtain is a waiver, in whole or in part, and the waiver must be in writing. A waiver matters when the business intends to modify the work — to redraw the mark for a new medium, to change a color, to adapt a template — because a modification that damages the author's reputation is the kind of thing moral rights protect against. A studio that assigns the copyright and waives its moral rights has given the business everything it needs to use and adapt the identity without coming back for permission.

This book explains the law and does not replace the advice of a lawyer or a trademark agent. The clauses described here are ordinary commercial clauses, and a business that is commissioning a substantial identity project should have the agreement reviewed by a lawyer before it is signed. Book H5 (Intellectual Property) treats the wider subject, including the registration of a trademark in the name itself, and Chapter 11 of this book treats the question of when a business should bring in a specialist rather than doing the work itself.

Master files are the deliverable that a business most often fails to collect. The vector masters of the logo and its variants, the palette with its codes, the typeface files with their licenses, the templates in their editable form and the guide are all part of what was bought, and they should be handed over at the end of the project in a folder the business controls. A business that holds only the finished exports cannot produce a new sign, a new uniform or a new template without going back to the studio, which is a dependency rather than an asset. The handover is also the moment to confirm the licenses: which typefaces are used, under which license, and whether the license covers the media the business uses.

The handover checklist is short enough to state in a paragraph. The business should hold the vector masters of every logo version, the palette with screen and print codes, the typeface files with their license documents, the templates in editable form, the guide, and the written assignment with the moral rights waiver. It should also hold a record of what was paid and what the scope was, because the next project will be priced against it. A business that can produce all of those items has commissioned a visual identity. A business that cannot has commissioned a picture.

QUÉBEC & CANADA NOTE Who owns the logo you paid for

Under the Copyright Act, the author of a work is its first owner (s. 13(1)), an employer is the first owner of a work made by an employee in the course of employment unless agreed otherwise (s. 13(3)), and an assignment or grant of copyright is valid only if it is in writing and signed by the owner of the right (s. 13(4)). Moral rights cannot be assigned but can be waived, in whole or in part, and an assignment of copyright does not by itself waive them (s. 14.1). The consequence for a Québec business that hires an independent design studio is that the studio remains the owner of the logo, the palette artwork and the templates unless a signed written assignment says otherwise, and that the moral rights must be waived in writing if the business intends to modify the work. Paying the invoice is not the same as acquiring the rights, and the two are separated by a signature. This book explains the law and does not replace the advice of a lawyer or a trademark agent.

TOOLS OF THE TRADE What an identity project needs

An identity project is carried by five functions, and a business should look for the function rather than for a product. The first is a vector drawing application, used by the studio to draw the mark and its variants as scalable artwork rather than as a picture. The second is a contrast checker, which takes two color codes and returns the ratio, and which is the tool that turns the palette from a preference into a tested decision. The third is a color-code reference for print and for screen, so that every color in the palette exists in both notations and can be given to a printer, a sign maker or a web developer without a conversation. The fourth is a shared folder for master files, holding the vector masters, the exports, the typeface files with their licenses and the guide, in a place the business controls rather than the studio. The fifth is a written brief and a signed agreement, which are not software but are the tools that decide whether the project ends with an asset or with a dependency. No brand name belongs on this list, and no price; the requirement is the function.

4.8 The Case's New Identity

Nordec's identity project began with a brief and ended with a signed contract, and the sequence is worth following because it is repeatable. The brief was written by Sophie with the administrator, and it drew on the material the business already had: the twenty value interviews of Book A6, the words customers used about Nordec, the touchpoint inventory of 36 places, and the platform of Chapter 2 with its purpose, its promise, its three values and its four personality traits. The brief named the two territories, the buildings of 2,000 to 9,000 square metres, the owner-operators and the property managers, and the four differentiators: the 90-minute response, the building health record, the one-page monthly report and the capital planning note. It also named what the identity had to work on: two vans, seven uniforms, a depot sign, eleven templates, a website and a monthly report that 48 buildings receive twelve times a year.

The studio presented three routes. The first kept the wrench and modernised the circle around it. The second replaced the wrench with a building outline and set the name beneath it. The third kept the letter N and drew a horizon line through it, dropping the wrench entirely. The three were genuinely different, and the workshop that chose between them took half a day. The first route was rejected because the wrench says what every competitor says. The second was rejected because a building outline is the most common mark in the category and because it describes the customer's asset rather than the business's promise. The third was chosen because the letter N is a form every customer already reads, because the horizon line is a picture of the promise — the building seen a year ahead — and because the mark survives at the size of a favicon and at the size of a van.

The palette was built around the blue that customers already associated with the business, and it was reduced from five blues to one. Nordec Blue, #1F3A5F, is the primary. Signal Amber, #E8A33D, is the accent. Slate, #5B6770, is the secondary neutral for text that should not shout, and Mist, #F4F6F8, is the background neutral, with white as the fifth. The reduction is the part of the project that customers will notice least and that the business will feel most, because a person producing a document now has one blue to reach for rather than five to choose between.

The contrast test is what turned the palette from a set of colors into a set of rules. The five pairs were computed, and four of them passed the 4.5 to 1 threshold for body text: white on Nordec Blue at 11.48 to 1, Nordec Blue on Mist at 10.60 to 1, Slate on white at 5.80 to 1, and Nordec Blue on Signal Amber at 5.32 to 1. The fifth, Signal Amber on white, measured 2.16 to 1 and failed both the body-text threshold and the 3 to 1 threshold for large text. The consequence was not a change of color but a change of role: amber carries accents, the van stripe and safety details, and never text on a white background. Where amber is a background, the words on it are Nordec Blue, which passes at 5.32 to 1. The rule is written into the guide with the ratios beside it, so that the next person to lay out a document chooses from a table rather than from a color picker.

The typography followed the same discipline. Two open-license typefaces were chosen: a humanist sans-serif for everything, and a monospaced face for building record codes, because the codes appear in columns in the building health record and a monospaced face makes the columns line up. Both are open-license, which means the business can use, embed and modify them within the terms of the license, and the license does not reach the logo artwork, so the mark itself belongs to Nordec. The weights and sizes were fixed as a small scale, and the substitutes were named so that a document produced on a Friday does not introduce a fifth typeface.

The logo system that was delivered has six versions: the primary horizontal lock-up, the stacked lock-up, the symbol alone, the wordmark alone, the one-color version and the reversed version. The clear space is equal to the height of the letter N, and the minimum size is 20 millimetres in print and 24 pixels on screen for the symbol. The imagery rules were written at the same time: photographs of real Nordec technicians at work in real buildings, in natural light, with no stock photographs of smiling models, and line icons of the four building systems drawn in Nordec Blue.

The contract with the studio assigns the copyright in the deliverables to Nordec in writing and includes a waiver of moral rights, which is what allows the business to modify the mark for a new medium without going back for permission. The fee was a fixed $7,800 for the logo, the palette, the typography, eleven templates and the guide, and the master files were handed over in a folder the business controls. The administrator, who becomes the brand keeper in Chapter 5, holds the templates and the asset folder and checks every new material against the guide.

What the identity does not do is worth one sentence, because the case is often read as a story about a logo. The new identity did not sell a single contract. It made the business recognizable in the places where it was already being seen — two vans on the road every working day, seven people in uniform, 624 preventive visits a year and 576 monthly reports — and it made the business's own materials look like they came from one business rather than from three. The offer, the Bilan Nordec and the referrals are what moved the numbers, and Chapter 13 is explicit that the growth is not attributed to the brand. What the identity changed is the cost of being recognized, and that cost is paid every day.

IN LA MACHINE A contrast badge, a website theme and a visual studio

Three pages of the application copy read for this book touch the work of this chapter, and one of them does something genuinely useful (La Machine, 2026). Business → Branding, at business/branding, offers a logo drop zone for PNG, JPEG, SVG or WebP files, a brand name, a tagline and a website address, a primary and a secondary color with ten preset swatches, a choice among five fonts, an email footer and three social links; it computes the contrast of the primary color and shows a pass or fail badge at the 4.5 to 1 threshold, with a live preview of a client invitation card and a client email. Its values are not saved: the "Save branding" button only shows a confirmation, and no other page uses what was entered. The page is a place to try a palette and see its contrast, not a brand register. Business → My Website, at business/website, has a Theme & Branding panel with a gallery of 28 themes, a brand name, a tagline, a logo, a favicon and a banner image, five color overrides and a heading and body font chosen from a catalog of 200, kept in the browser. Assets → Visual Studio, at business/social/visual-studio, is an editor for social, video, print and web formats, including a business card and a poster, whose designs are saved and whose exports go to the Gallery. None of the three reads a stored brand palette or logo.

Costly Mistakes

• Treating the logo as the whole identity. A logo without a palette, a typeface, imagery rules and templates is a signature with no stationery, and it will be applied differently by every person who uses it.
• Keeping several versions of the logo in circulation. Three versions of a mark and five blues do not look like variety; they look like a business that cannot decide, and they make recognition harder every month they continue.
• Choosing a color for text without testing it. A warm amber or a pale gray that reads comfortably on a designer's screen may fail the contrast threshold on a phone in daylight, and the failure is invisible to the person who chose it.
• Using a typeface without a license that covers the use. A desktop license does not automatically cover a website, an application or a vehicle wrap, and discovering the gap after two vans have been wrapped is expensive.
• Commissioning a logo without a signed written assignment. The independent studio remains the author and the first owner of the work unless a signed assignment says otherwise, and paying the invoice does not transfer the rights.
• Forgetting the moral rights waiver. Moral rights cannot be assigned but can be waived in writing, and a business that intends to modify its own mark needs the waiver in the same agreement as the assignment.
• Failing to collect the master files. A business that holds only finished image exports cannot produce a new sign, a new uniform or a new template without returning to the studio, which turns an asset into a dependency.

Key Takeaways

• A visual identity is a small set of signs — a logo and its variants, colors, typefaces, imagery and icons — chosen to express the brand platform and used the same way everywhere, and it works through repetition rather than through invention.
• The identity process has five phases — research, strategy, design, launch and governance — and governance is a phase of the work rather than an afterthought (Wheeler & Meyerson, 2024); most small businesses begin at design and stop at launch.
• A logo is a system rather than a file: Nordec received six versions, with clear space equal to the height of the letter N and a minimum size of 20 millimetres in print and 24 pixels on screen for the symbol.
• The study of 195 logos rated on 13 design characteristics recommends that logos meant for high recognition be very natural, very harmonious and moderately elaborate, and that logos with a lower investment behind them be less natural and very harmonious (Henderson & Cote, 1998).
• Every text-and-background pair in a palette is tested before publication: level AA requires 4.5 to 1 for text and 3 to 1 for large text, level AAA requires 7 to 1 and 4.5 to 1, and graphical objects needed to understand content require 3 to 1 (World Wide Web Consortium, 2024); Nordec's Signal Amber on white measures 2.16 to 1 and therefore carries accents, never text.
• A commissioned identity is owned only when the copyright is assigned in writing and signed and the moral rights are waived, because the author is the first owner of a work and an assignment is valid only in writing (Copyright Act, ss. 13 and 14.1).

Your Action Plan

1. Inventory the signs the business is using today. Collect every version of the logo, every color and every typeface that appears on a document, a vehicle, a sign or a screen, and count the touchpoints where the brand appears. Deliverable: a list of versions, colors and typefaces with the touchpoint count. Verification: every item on the list can be traced to a file or a material that exists.
2. Write the brief before commissioning anything. State what the business does, who its customers are, what it promises, what makes it different, the places the identity must work and the languages it must work in. Deliverable: a brief of one or two pages. Verification: the brief names the differentiators and the media, and a studio could price the project from it.
3. Choose a palette and test every pair. Reduce the colors to one primary, one or two accents and one or two neutrals, write each with its screen and print codes, and compute the contrast ratio of every text-and-background pair the business intends to use. Deliverable: a palette with codes and a table of approved pairs with their ratios. Verification: no pair used for text measures below 4.5 to 1, and every failing color has been given a role that does not involve text.
4. Confirm the typeface licenses for every medium. List the media the business uses — print, web, application, vehicle, email — and check that the license for each typeface covers each of them, keeping the license documents with the font files. Deliverable: a license record per typeface. Verification: every medium on the list is covered by a license the business holds.
5. Sign the assignment and collect the master files. Have the agreement assign the copyright in writing and signed, include a waiver of moral rights, and hand over the vector masters, the exports, the typeface files, the templates and the guide into a folder the business controls. Deliverable: a signed agreement and a complete master folder. Verification: the business can produce a new sign, a new uniform and a new template without contacting the studio.

In La Machine

Three routes in the application copy read for this book touch the work of this chapter, and their limits are as informative as their functions (La Machine, 2026). Business → Branding, at business/branding, takes a logo file in PNG, JPEG, SVG or WebP, a brand name, a tagline and a website address, a primary and a secondary color with ten preset swatches, a choice among five fonts, an email footer and three social links, and it computes the contrast of the primary color and shows a pass or fail badge at the 4.5 to 1 threshold, with a live preview of a client invitation card and a client email. That badge is the one place in the platform where a palette decision is checked against a number, and it is worth using for exactly that. Business → My Website, at business/website, carries a Theme & Branding panel with a gallery of 28 themes, a brand name, a tagline, a logo, a favicon and a banner image, five color overrides and a heading and body font chosen from a catalog of 200, kept in the browser. Assets → Visual Studio, at business/social/visual-studio, is an editor for social, video, print and web formats, including a business card and a poster, whose designs are saved and whose exports go to the Gallery.

What the platform does not do belongs in the same breath, and it is most of what this chapter describes. The Branding page does not save its values: the "Save branding" button shows a confirmation and no other page uses what was entered, so the page is a place to try a palette and see its contrast rather than a brand register. My Website keeps its configuration in the browser and applies it to that site alone. Visual Studio, the Gallery and Montages do not read a stored brand palette or logo, and the Gallery records, for each stock image, its license, the license's address and the attribution text, plus a provenance form for the user's own images, which is a record rather than a system. La Machine has no saved brand kit that other pages use, no brand portal or guidelines publisher, no digital asset management with versions, usage rights or expiry dates, and no brand consistency audit. The palette, the typefaces, the logo system and the guide this chapter describes live outside the platform, in a folder the business controls, and the platform is where some of them are applied rather than where they are kept.

Self-Check

1. What are the three jobs of a visual identity, and what are the five phases of the identity process?
2. What is a logo system, and why does a business need more than one version of its logo?
3. What does the research on logo design recommend for a logo with a lower investment behind it, and what do the three characteristics mean in ordinary language?
4. What are the contrast thresholds at level AA and level AAA, and what does the logo exemption cover?
5. Who owns a logo that an independent studio has designed and the business has paid for, and what two things must the agreement contain?

Answers. 1. The three jobs are recognition, differentiation and meaning: the signs should let a customer tell at a glance that two materials come from the same business, should distinguish the business from its competitors, and should match what the business says about itself. The five phases are research, strategy, design, launch and governance, and governance — managing the assets after the launch — is a phase of the work rather than an afterthought (Wheeler & Meyerson, 2024). 2. A logo system is the set of versions of the mark that a business actually needs: a primary horizontal lock-up, a stacked lock-up, the symbol alone, the wordmark alone, a one-color version and a reversed version, each with its clear space and its minimum size. The versions exist because real situations demand them — a van roof needs the symbol alone, an embroidered uniform needs one color, a dark panel needs the reversed version — and a business without them will improvise them badly at the moment of need. 3. The study of 195 logos rated on 13 design characteristics recommends that logos with a lower investment behind them be less natural and very harmonious (Henderson & Cote, 1998). Natural means that the mark contains recognizable objects and shapes rather than abstract geometry; harmonious means that the parts belong together and the proportions are balanced; elaborate refers to the amount of detail and design content in the mark. 4. At level AA, success criterion 1.4.3 requires text to have a contrast ratio of at least 4.5 to 1 against its background and large text — 18 points, or 14 points bold — at least 3 to 1, and success criterion 1.4.11 requires graphical objects and interface components needed to understand content to have at least 3 to 1 against adjacent colors; at level AAA, success criterion 1.4.6 raises this to 7 to 1 and 4.5 to 1 for large text (World Wide Web Consortium, 2024). The logo exemption covers text that is part of a logo or a brand name, which has no contrast requirement, and it does not extend to the palette used for ordinary text. 5. The studio owns the logo, because the author of a work is its first owner and an assignment or grant of copyright is valid only if it is in writing and signed by the owner of the right (Copyright Act, s. 13). The agreement must therefore contain a written, signed assignment of the copyright in the deliverables, and a waiver of moral rights, in whole or in part, because moral rights cannot be assigned and an assignment of copyright does not by itself waive them (Copyright Act, s. 14.1).

Chapter References

Copyright Act, R.S.C. 1985, c. C-42, ss. 13 and 14.1 (Canada). Justice Laws Website, current to September 3, 2026 (last amended November 7, 2024) — read September 25, 2026.

Henderson, P. W., & Cote, J. A. (1998). Guidelines for selecting or modifying logos. Journal of Marketing, 62(2), 14–30. https://doi.org/10.1177/002224299806200202

La Machine. (2026). Documentation — route reference inventory. Internal product documentation, read September 25, 2026, together with the code of the pages this book describes: business/branding, business/website and business/social/visual-studio.

Labrecque, L. I., & Milne, G. R. (2012). Exciting red and competent blue: The importance of color in marketing. Journal of the Academy of Marketing Science, 40(5), 711–727. https://doi.org/10.1007/s11747-010-0245-y

Romaniuk, J. (2018). Building distinctive brand assets. Oxford University Press.

SIL International. (2007). SIL Open Font License (Version 1.1, February 26, 2007). Retrieved September 25, 2026, from https://openfontlicense.org/open-font-license-official-text/

Wheeler, A., & Meyerson, R. (2024). Designing brand identity: A comprehensive guide to the world of brands and branding (6th ed.). Wiley.

World Wide Web Consortium. (2024). Web Content Accessibility Guidelines (WCAG) 2.2. W3C Recommendation, 12 December 2024 (first published as a Recommendation on 5 October 2023). Retrieved September 25, 2026, from https://www.w3.org/TR/WCAG22/

Chapter 5 — Design Systems and Brand Guidelines

An identity becomes useful when it becomes a system — named decisions, reusable pieces, ready-to-fill documents and a guide that says who may change what — so that a technician, a printer and a web developer produce the same brand without a designer at their side.

What You'll Be Able to Do

• Explain why a set of delivered files is not a system, and name the five layers a small business's design system contains.
• Write a set of design tokens that names every color, type size, spacing and logo size by its role, and point every template at those names.
• Build a small library of components and templates from the pieces a business already uses, so that a document is filled rather than laid out.
• Assemble a brand guide in a full version and a one-page summary, and decide what belongs in each.
• Set the levels of change in a brand and say who approves each one, from filling a template to changing the promise.
• Keep an asset folder with masters, exports, retired files and recorded licenses, and run the yearly review that keeps the system alive.

Why It Matters

The work of the previous chapters produces decisions: a platform, a name, a voice, a logo, a palette, a set of typefaces. Every one of those decisions is correct on the day it is taken, and every one of them begins to erode on the day after, because the brand is not produced by the people who designed it. It is produced by a technician writing a visit sheet in a mechanical room, an administrator assembling a proposal between two telephone calls, a wrap shop that has never met the business, and whoever updates the website. Those people are competent and busy, and none of them will read a design rationale. What they will use is a template that is already correct and a guide that answers the two questions the template does not.

Without that, the decisions are re-taken, slightly differently, hundreds of times a year. The case followed through this book measured the result before the work began: an inventory of thirty-six touchpoints found three versions of the logo, five different blues, four typefaces, and only thirteen of the thirty-six touchpoints consistent with one another, or 36.1%. Nobody had decided to use five blues. Each one had arrived with a document, a supplier or a person, and each one had been reasonable on the day it appeared. The same erosion happens in every small business, and it is invisible because no single instance of it is a problem.

A system is what stops it, and a system is cheaper than it sounds. Nordec's system is twenty-one design tokens, a small set of components, eleven templates, a twenty-four-page guide with a one-page summary, one asset folder and one person who keeps it. It was built from the studio's files rather than from scratch, and it saves the administrator about twenty-five and a half hours a year on proposals and reports alone, before counting the monthly reports that were already templated. The larger gain is not the hours. It is that a decision taken once reaches every document, every vehicle and every page without anybody having to remember it.

In This Chapter

Section 5.1 explains why a set of files is not a system and places governance as the fifth phase of the identity process. Section 5.2 sets out design tokens as named decisions, with the vendor-neutral format that reached its first stable version in October 2025. Section 5.3 builds components and templates from small reusable pieces. Section 5.4 describes the brand guidelines document, in its full and one-page forms. Section 5.5 sets the levels of change and says who approves each one. Section 5.6 covers the asset library, masters and exports, versions and licenses. Section 5.7 explains how a system is kept alive, including the accessibility rules carried in the tokens. Section 5.8 shows how the studio's files became Nordec's system, and what the technicians did with the first template.

5.1 From Identity to System

A brand identity arrives as a set of files, and a set of files is not a system. The studio hands over a logo in several formats, a palette with its codes, two typefaces, a handful of layouts and a document that explains them. Every one of those files is the visible result of a decision — this blue rather than that one, this letterform rather than the one the business used before, this much space around the mark — and the decisions are what the business actually needs to keep. A file cannot explain itself. It cannot say why the blue is that blue, which color a heading may sit on, how much clear space the mark requires, or what to do when a document type appears that nobody designed. A system is the set of decisions written down in a form that people who are not designers can apply.

The distance between the two is easy to measure in the case followed through this book. An inventory of Nordec's thirty-six touchpoints, taken before the work of this book began, found three versions of the logo in circulation, five different blues, four typefaces, and only thirteen of the thirty-six touchpoints consistent with one another, or 36.1%. Nobody had decided to use five blues. Each one had arrived with a document, a supplier or a person, and each one had been reasonable on the day it appeared. What was missing was not a better logo. It was a place where the decisions lived, so that the next document did not have to invent them again.

The design literature treats that place as part of the work rather than as an afterthought. The five-phase identity process set out by Wheeler and Meyerson runs through research, strategy, design and launch across touchpoints, and then adds a fifth phase, governance, described as the managing of assets (Wheeler & Meyerson, 2024). The order matters. Governance is not what remains when the interesting work is finished; it is the phase that decides whether the four earlier phases survive their first year. A business that stops at launch has bought a set of files. A business that completes the fifth phase has bought a system, and the difference shows up in the second year, when the people who commissioned the work have moved on to other things and the documents keep being produced.

Figure 5.1 shows the shape of that system as four stacked layers. The bottom layer holds twenty-one design tokens, the named decisions about color, type, spacing and logo size. The second layer holds the components built from those tokens: the lock-ups, the header, the table style, the callout and the signature block. The third layer holds the eleven templates that assemble the components into documents a person can fill. The top layer holds the thirty-six touchpoints where the brand meets a customer. Beside the four layers, two boxes govern all of them: the twenty-four-page guide that records the decisions, and the brand keeper who holds the templates and checks every new material.

Figure 5.1 — From identity to system. A change made once in a token reaches every template and every touchpoint, which is the point of a system.

The layers are worth reading from the bottom up, because that is the direction in which a change travels. A decision taken at the level of a token — say, that the muted text color is Slate, #5B6770 — reaches every component that uses muted text, every template that assembles those components, and every touchpoint where a template is used. A decision taken at the level of a template reaches only that template. A decision taken at the level of a single document reaches nothing at all, which is why the five blues in the inventory were not a decision but an accumulation. The purpose of a system is to move decisions as far down the stack as they will go, so that the fewest possible decisions have to be repeated.

For a small business, the argument for a system is not tidiness. It is that the brand is produced by people who are not designers. In Nordec's case, the visit sheet is written by a technician in a mechanical room, the proposal is assembled by the administrator between two telephone calls, the vehicle wrap is produced by a wrap shop that has never met the business, and the website is maintained by whoever updates it. None of those people will read a design rationale, and none of them should have to. What they need is a template that is already correct and a guide that answers the two or three questions the template does not. A system is what allows a business to be consistent without a designer standing beside every document.

The cost of not having one is paid in small amounts that are easy to miss. Before the system existed, the administrator rebuilt the layout of each proposal and each report by hand, at about twenty minutes a document. The work was not wasted — the documents were produced and sent — but the twenty minutes were spent on a decision that had already been taken and then forgotten. Multiply that by the documents a year and the figure becomes a real one, which is the calculation set out in section 5.3. The point of the system is not that it saves the whole amount; it is that the amount is spent on the customer rather than on the layout.

A system also changes what a business can hand over. When the administrator is away, a colleague can fill a template without asking how it should look. When a new technician joins, the visit sheet is already the shape the business uses. When the business hires a printer, a wrap shop or a web developer, the guide answers the questions that would otherwise arrive by telephone over three days. And when the business decides to change something — a color, a tagline, a template — there is one place to change it and a record of who agreed. That is what the rest of this chapter builds, one layer at a time.

One caution belongs at the start, because it is the mistake that turns a system into a burden. A system is not a catalog of everything a business might ever produce. It is the small set of decisions that are repeated often enough to be worth naming. Nordec's twenty-one tokens and eleven templates cover the documents the business actually issues; they do not cover a trade-show banner that will be produced once, or a photograph taken for a single news item. A system that tries to anticipate every case becomes a document nobody reads, and the guide that nobody opens is the subject of section 5.4. The test is repetition: if a decision is taken more than a few times a year, it belongs in the system, and if it is taken once, it belongs in a conversation.

DEFINITION Design token, component, template, design system and brand guide

A design token is a named decision about a visual value — a color, a type size, a spacing, a minimum logo size — written once and referred to by name everywhere it is used. A component is a reusable piece of a document or a page built from tokens: a logo lock-up, a header, a table style, a callout, a signature block. A template is a document assembled from components, with the places a person fills left empty: a visit sheet, a proposal, a monthly report. A design system is the whole arrangement — the tokens, the components, the templates, the asset library and the rules that connect them — kept in one place and maintained over time. A brand guide is the document that records the system in words and pictures, so that a person who did not take part in the design can apply it correctly. The five terms describe one thing seen at five distances, and a small business needs all five in some form, however modest that form may be.

5.2 Design Tokens

A design token is a name for a decision. Instead of writing a color code into a document, the document refers to a name, and the name holds the code. The name is the important part, because it is what allows the decision to change without the document changing. If the business later adjusts its blue, the token keeps its name and the documents that refer to it follow. If the business had written the code into every document, the same adjustment would mean opening every file, which in practice means the adjustment never happens and the old blue survives in half the materials.

The naming rule is that a token is named by its role rather than by its look. Nordec's primary color token is color.brand.primary, not a name built from the word blue, because the day the business decides that its primary color is no longer blue, a token named after the color would have to be renamed everywhere it appears. The same rule applies to type and spacing. A token called type.size.body describes what the size is for; a token called after its measurement would have to be renamed the moment the measurement changed. Role names are slightly harder to write on the first day and much easier to live with afterwards, which is the trade a system always makes.

Nordec's system holds twenty-one tokens in four families: seven color tokens, six typography tokens, five spacing tokens and three logo sizes. The color family covers the primary blue, the amber accent, the muted text color, the subtle surface, white, and the two pairs that are allowed to carry text. The typography family covers the two typefaces, the body size in print and on screen, the heading size, the caption size and the line spacing. The spacing family covers the standard gaps between blocks, the page margin, the space above a heading, the space inside a table cell and the space around the logo. The logo family covers the minimum size of the symbol, of the horizontal lock-up and of the stacked lock-up.

The following table presents a selection of eight of Nordec's twenty-one tokens, with the value each one holds and the places where it is used. Read it as a sample rather than as the whole set: the point of the table is to show how a token ties a value to a use, so that a person filling a template never has to decide what color a heading should be.

Table 5.1 — A selection of Nordec's tokens

A token set is only useful if the tools that produce the documents can read it. That is the problem the vendor-neutral format addresses. On October 28, 2025, version 2025.10 of the design tokens specification was published as the first stable version of the format, by the Design Tokens Community Group of the World Wide Web Consortium (Design Tokens Community Group, 2025). It is a Community Group report and not a W3C Standard, and it should be described as such. What it offers a small business is a way of writing the token set once so that it can produce the values a design application needs, the values a web page needs and the values a print layout needs, with theming and multi-brand support, modern color spaces and aliases between tokens. A business with one brand and one website does not need the full format. It needs the idea, which is that the values live in one list rather than in every file.

The color tokens carry one more piece of information that is easy to leave out and expensive to leave out: the pairs that are allowed to carry text. Nordec's guide records five measured pairs. White text on Nordec Blue measures 11.48 to 1. Nordec Blue text on Mist measures 10.60 to 1. Slate text on white measures 5.80 to 1. Nordec Blue text on Signal Amber measures 5.32 to 1. Signal Amber text on white measures 2.16 to 1, which is why amber never carries text on a white background and serves instead for accents, the van stripe and safety details. Recording the pairs beside the color tokens is what stops a well-meaning person from setting a heading in amber on white, and it is the reason the contrast rule in section 5.7 is written into the tokens rather than into a paragraph of advice.

KEY CONCEPT Decide once, apply everywhere

The whole value of a design system is contained in one sentence: a decision is taken once, in one place, and every output follows it. That is what separates a system from a folder of files. In a folder of files, the same decision is taken again in every document, by whoever is producing it, with whatever they remember of the last one — which is how a business ends up with five blues and four typefaces without anybody choosing them. In a system, the decision is written down as a token, the templates refer to the token, and the person filling the template never sees the decision at all. The consequence is not only consistency. It is speed, because the twenty minutes spent rebuilding a layout become five minutes spent filling one, and it is control, because a change made once reaches every touchpoint instead of reaching the ones somebody remembered. The rule is worth stating in the negative as well: any decision that has to be repeated is a decision that has not yet been made.

METHOD Turning an identity into tokens

First, list every color, type size, spacing value and logo size that appears in the identity as it stands, including the ones that arrived by accident, because the list is the honest starting point. Second, give each value a name by its role rather than by its look, so that the name survives a change of value. Third, remove the near-duplicates: two blues that differ by a few points of shade are one blue, and the same is true of two body sizes that differ by half a point. Fourth, point every template at the names rather than at the values, so that no template holds a code of its own. Fifth, record the values and the pairs that are allowed to carry text in the guide, beside the tokens, so that the rule travels with the decision. The method takes an afternoon for a small identity, and it is the afternoon that makes every later change cheap.

WHAT CHANGED A stable format for design tokens

On October 28, 2025, the design tokens specification reached its first stable version, numbered 2025.10, published by the Design Tokens Community Group of the World Wide Web Consortium (Design Tokens Community Group, 2025). It is a Community Group report and not a W3C Standard, and it is not a Recommendation. What changed for a small business is narrower than it sounds and more useful than it sounds. Before a stable version existed, a token set was a private convention: each business wrote its list in whatever shape its tools happened to accept, and moving the list from one tool to another meant rewriting it. A stable, vendor-neutral format means the same list can produce the values a design application needs, the values a web page needs and the values a print layout needs, with theming and multi-brand support and aliases between tokens. A business with one brand and one website will not use most of that. It can still write its twenty-one tokens in the shared shape, which is what makes the list portable in five years.

5.3 Components and Templates

A system is built from small pieces upwards, and the vocabulary for that is well established. The approach known as atomic design describes five levels for building interfaces from reusable parts: atoms, molecules, organisms, templates and pages (Frost, 2016). An atom is the smallest piece that cannot usefully be divided further. A molecule is a small group of atoms working together. An organism is a larger group that forms a recognisable section. A template arranges organisms into a layout with the content left out. A page is that layout with real content in it. The names are borrowed from chemistry and they are not important in themselves; what matters is the direction, which is from the smallest reusable piece to the finished document, so that a change at the bottom reaches everything above it.

Mapped onto a small business, the levels are less abstract than they sound. Nordec's atoms are the color values, the type sizes and the letter N itself. A molecule is the logo lock-up: the symbol, the wordmark and the clear space around them, treated as one piece that may be placed but not rearranged. An organism is the header of the monthly report, which combines the lock-up, the building name, the month and the rule beneath them. A template is the monthly report with its fields empty. A page is that report filled in for one building in one month. The same five levels describe the proposal, the visit sheet and the website, which is why the vocabulary is worth learning once.

A small business's component library is short, and it is worth listing because the list is the whole of it. Nordec's components are the three logo lock-ups, the document header, the document footer, the table style, the callout, the signature block, the photograph treatment and the icon set. Nine components carry eleven templates and thirty-six touchpoints. None of them is complicated, and each of them exists because it appears in more than one place. The test for adding a component is the same test as for a token: if a piece appears in three documents, it is a component, and if it appears in one, it is a detail of that document.

The templates are where the system meets the work. Nordec has eleven: the visit sheet, the one-page monthly report, the capital planning note, the consolidated portfolio report, the proposal, the invoice, the Bilan Nordec report, the letter, the email signature, the slide and the social media post. Each one is a document with the components already placed and the fields left empty. The technician fills the visit sheet on site; the administrator fills the proposal from the notes of a call; the monthly report is produced for each of the forty-eight buildings on Preventive and Portfolio, which is five hundred and seventy-six reports a year. A template is not a form in the bureaucratic sense. It is a document that has already been laid out correctly, so that the person using it spends their attention on the content.

The reason templates beat instructions is worth stating plainly, because it is the reason this chapter exists. An instruction asks a person to remember a rule and apply it under time pressure. A template asks a person to fill a field. The first fails whenever the person is tired, new, in a hurry or working from a telephone, which is most of the time in a small business. The second fails only when the template itself is wrong, which is a problem that can be fixed once. This is also why the templates must be editable by the people who use them rather than locked: a template that cannot be adjusted for a real case will be abandoned, and the document will be rebuilt by hand, which is exactly the twenty minutes the system was meant to remove.

The templates also carry the language question, which in Québec is not optional. A business that serves customers in French and in English prepares its templates in French first and produces English versions where it serves English-speaking customers, because commercial documents must be available in French under the Charter of the French language. The practical consequence for a design system is that the French and English versions of a template must be built from the same components, so that a change to the header reaches both, and that the French version must be the one that is maintained first. A business that maintains only the English version and translates when asked will find that its French documents are the ones that drift.

QUÉBEC & CANADA NOTE Templates in both languages

A Québec business prepares its templates in French, with English versions where it serves English-speaking customers, because commercial documents must be available in French under the Charter of the French language (CQLR c C-11). The design consequence is specific and worth building into the system from the first day. The two language versions of a template are not two templates; they are one template with two sets of text, built from the same components, so that a change to the header, the table style or the footer reaches both. The French version is the one that is maintained first, and the English version follows, because a business that maintains only the English version discovers its French documents drifting within a year. Nordec's eleven templates exist in both languages, and the tier names follow the same rule: Essentiel, Préventif and Portefeuille in French, with Essential, Preventive and Portfolio in English. This book explains the law and does not replace the advice of a lawyer.

DO THE MATH What the system saves Nordec

Illustrative figures; the case is fictional. Step 1 — Before the system, the administrator rebuilt each proposal and report layout by hand, at about 20 minutes a document. Step 2 — With templates, the same document takes about 5 minutes, because the layout is already correct and only the content is filled. Step 3 — The business produces about 15 proposals and Bilan Nordec reports every six months, which is 30 a year, plus about 6 other one-off documents a month, which is 72 a year, for a total of 102 documents a year. Step 4 — At 15 minutes saved per document, 102 × 15 minutes = 1,530 minutes, or 25.5 hours a year. Step 5 — At the administrator's rate of $36 an hour, that is about $918 a year, and the figure excludes the monthly reports, which were already templated before the system was built and therefore save nothing further.

5.4 The Brand Guidelines Document

A brand guide is the document that records the system so that a person who did not take part in the design can apply it. It is written for its readers rather than for its authors, and its readers are not designers. They are the administrator filling a proposal, the technician writing a visit sheet, the printer setting a job, the wrap shop laying out a van, the web developer building a page and the supplier producing a garment. Each of them arrives with a question, and the guide's job is to answer that question in the place where they will look for it. A guide that answers the questions its readers actually ask is used; a guide that presents the design is admired once and then closed.

Nordec's guide runs to twenty-four pages, with a one-page summary that is printed and posted. The full version is organized by the question rather than by the design process, which is why the platform comes first and the approvals come last. The platform section states the purpose, the promise, the three values and the four personality traits, because a person who is about to write to a customer needs to know what the business is promising before they know which blue to use. The voice section carries the four principles — say the time, say the cost, say what comes next, no jargon without its plain word — together with the glossary of twelve words to use and twelve to avoid. The logo section shows the lock-ups, the clear space, the minimum sizes and the things that must never be done to the mark.

The following table sets out the nine parts of Nordec's guide, with the number of pages each one takes, what it says and who uses it most. Read it as a map of the document rather than as a table of contents: the page counts show where the weight sits, and the last column shows that no single part of the guide serves everybody, which is the reason the parts are separated at all.

Table 5.2 — The contents of Nordec's guide

The one-page summary is not a shortened guide. It is a different document with a different job. It carries the logo lock-up, the five colors with their codes, the two typefaces, the tagline and the name of the brand keeper, and it fits on a single sheet that can be pinned above a desk or taped inside a van. Its purpose is to answer the questions that arrive when nobody is going to open a twenty-four-page document: which blue, which typeface, who do I ask. A business that produces only the full guide will find that the questions are answered from memory, and a business that produces only the summary will find that the harder questions have nowhere to go. The two documents are a pair, and the summary is the one that gets used most.

The guide is also where the system admits what it does not cover. Every identity meets cases its designers did not anticipate: a co-branded document with a manufacturer, a sign that must carry a permit number, a photograph that must show a customer's building. A guide that pretends to cover everything will be ignored at the first case it gets wrong. A guide that names the cases it does not cover, and says who decides them, keeps its authority for the cases it does. Nordec's guide devotes part of its approvals section to exactly this: a short list of situations that go to the brand keeper, and a shorter list that goes to Sophie, so that a person facing an unusual case knows where to take it rather than guessing.

WATCH OUT The guide nobody opens

The most common failure of a brand guide is not that it is wrong. It is that it is never opened, and it fails for three predictable reasons. The first is that it is written for the designer rather than the reader, so that a person looking for the answer to a practical question has to read three pages of rationale to find it. The second is that it is too long for the questions it answers, so that a twenty-four-page document is consulted for a decision that a one-page summary would have settled. The third, and the most damaging, is that it is out of date: the guide still shows the old template, the old color or the old tagline, and the first person who notices stops trusting it. The remedy for all three is the same discipline. Write the guide around the questions people ask, produce a one-page summary for the common ones, and give one person the job of keeping it current, with a date on the cover that changes when the contents do.

5.5 Governance: Who Approves What

Governance is the fifth phase of the identity process (Wheeler & Meyerson, 2024), and in a small business it takes the form of a short list of levels. The purpose of the list is not to create approvals for their own sake. It is to make sure that the decisions which are expensive to reverse are taken by the person who can see their consequences, and that the decisions which are cheap to reverse are taken by whoever is closest to the work. A business that requires the owner's approval for every document will find that the documents are produced without approval. A business that requires no approval at all will find that its identifiers change by accident.

The levels are best described by what they change. At the first level, a person fills a template: a technician completes a visit sheet, the administrator completes a proposal from a call. Nobody approves this, and nobody should, because the template already carries the decisions. At the second level, a person creates a document that does not exist as a template but is assembled from existing components: a one-off letter to a customer, a page for a trade event. This is the brand keeper's decision, taken in minutes, and it is the level at which most small businesses either have a rule or have nothing. At the third level, a template itself changes: a field is added to the visit sheet, a section is removed from the monthly report. This is the brand keeper's decision with a note in the change log, and it is the level at which a system quietly improves or quietly decays.

At the fourth level, an identifier changes: a token value, a lock-up, the clear space around the mark, the typefaces. This is Sophie's decision, taken with the studio, because it reaches every template and every touchpoint at once. At the fifth level, the platform changes: the promise, the name, the tagline, the values. This is not a design decision at all; it is the decision described in Chapter 2, and it belongs to the owner because it changes what the business is promising rather than how the promise looks. A business that treats the fifth level as a design change will find that its documents and its promise have drifted apart, which is the failure the whole book is written to prevent.

The exceptions are as important as the levels, and a small business meets three of them. The first is a supplier's own rules: a manufacturer's certified-installer mark may appear on the website only under the manufacturer's own usage rules, which may conflict with the business's own guide, and the supplier's rules win because the mark is not the business's to control. The second is a legal requirement: a permit number, a registration statement or a required notice may have to appear in a place and a size the guide does not anticipate. The third is a co-branded document, where two identities share a page and neither one's guide covers the combination. Each of the three is a case for the brand keeper, and each is worth naming in the guide so that the answer does not have to be invented under deadline.

The brand keeper is the person who makes the system work, and the role is smaller than its name suggests. In Nordec's case the administrator holds it: they keep the templates and the asset folder, they check every new material against the guide, they answer the questions that arrive from the printer and the wrap shop, and they keep the change log. They do not decide what the brand is; they decide whether a document follows it. That distinction is what makes the role workable in a business of seven people, because it separates the question of what the business promises, which is Sophie's, from the question of whether a document matches the promise, which is a check anybody can learn to perform.

WHO DOES THIS The brand keeper runs the system

In a small business the brand keeper is usually not a designer and does not need to be. In Nordec's case the administrator holds the role, and the work is a set of small habits rather than a job title. They hold the templates and the asset folder, and they are the only person who writes in the folder, which is what keeps the masters from being overwritten by exports. They check every new material before it leaves the business: a proposal, a page, a garment, a sign, a social post. They answer the questions that arrive from outside, from the printer, the wrap shop and the web developer, using the guide rather than their memory. They keep the change log, so that the answer to "when did this change?" is a line in a file rather than a guess. And they know the short list of questions that go to Sophie, which is the list of things that touch the promise or the identifiers. The role takes a few hours a month, and it is the difference between a system that is used and a set of files that was delivered.

The levels connect to the rest of the business through the recurring work described in Chapter 10. A template change is a task with a verification criterion, which is the shape the platform gives to work that must be done the same way each time. A token change is a project with a date, because it reaches every template and every touchpoint and cannot be done in an afternoon by one person. The yearly review described in section 5.7 is where the levels are read back: how many changes were made at each level, how many exceptions were granted, and whether the guide still describes the system that is actually in use. A governance list that is never read is a list of intentions, and the yearly review is what turns it into a record.

5.6 Asset Libraries and Version Control

The asset folder is the least glamorous part of a design system and the part that causes the most damage when it is missing. It holds the files that the templates and the suppliers need: the logo in its several formats, the typefaces, the photographs, the icons, the templates themselves and the guide. Without a folder, those files live in email attachments, in the downloads of three different people and in the account of a supplier who has since changed hands. The result is that the business cannot answer the simplest question a printer can ask, which is which file is the current one.

Nordec's folder is organized by identifier rather than by document type. There is one folder for the logo, one for the colors, one for the typefaces, one for the photographs, one for the icons, one for the templates and one for the guide. Inside each folder, the masters are separated from the exports. A master is the editable original from which everything else is produced; an export is a copy in a format for a particular use, such as a file for print, a file for a screen or a file for a garment. The separation matters because an export cannot be edited without losing what makes it an export, and a business that edits an export produces a file that no longer matches the master it came from.

The folder also holds a retired section, and the rule about it is simple: nothing is deleted. A retired file is moved rather than removed, because old documents exist. A customer holds a proposal from three years ago, a building carries a sign installed in 2024, and a van may still be on the road with an older stripe. When somebody asks whether a document is current, the answer is a comparison with the folder rather than a memory, and the retired section is what makes the comparison possible. The same rule applies to templates: a template that is replaced is moved to the retired section with the date on which it was replaced, so that a document produced from it can be identified later.

File names carry the version and the date, because a folder without them becomes a set of files called final, final-two and final-latest. Nordec's convention puts the identifier first, then the variant, then the version, then the date: nordec-logo-primary-v03-2026-09. The order matters because it sorts the files by identifier and then by recency, so that the newest version of a piece is the last one in its group. The convention is written into the guide, in the approvals section, because a naming rule that lives only in one person's habit disappears when that person is away.

Licenses are recorded with the files, and this is the part of the folder that a small business most often leaves out. The two typefaces in Nordec's identity are open-license faces, and the license that covers them permits using, studying, copying, embedding, modifying and redistributing the font software, with selling permitted only when the font is bundled with other software; reserved font names may not be used by modified versions without permission (SIL International, 2007). One detail of that license is worth recording in the folder because it is widely misunderstood: documents and other works made with the fonts are not covered by the license, so a logo set in such a font is not itself under it. Photographs carry a different record: for each image, the folder notes its license, the address of the license and the attribution text, and for the business's own photographs it notes the source, the address, the license and the attribution in the same fields.

The question of when a folder becomes a system is answered by volume and by people. A business of two people with one folder and one person who writes in it does not need anything more. A business of forty people, several suppliers and a website that changes weekly will find that a shared folder is not enough, because it cannot record who downloaded which file, which version is approved, or when a license expires. That is the point at which a brand portal or a digital asset management system earns its place, and Chapter 12 sets out the functions to look for. The functions come before the product: versions, approvals, usage rights, expiry dates and a search that a supplier can use without an account.

TOOLS OF THE TRADE What a small design system needs

A small design system is carried by four things, and none of them is a specialist product. The first is editable templates in the applications the business already uses, one file per template, with the components already placed and the fields left empty, saved in a folder that the brand keeper controls. The second is a shared asset folder with versions, organized by identifier, with masters separated from exports, a retired section and file names that carry the version and the date. The third is a token list, kept either in a spreadsheet or in a token file in the vendor-neutral format, holding the name, the value and the places each token is used, with the allowed color pairs recorded beside the color tokens. The fourth is a change log, a single page with four columns — the date, what changed, who asked and who approved — which is what turns a set of decisions into a history. No brand name belongs on this list, and no price; the requirement is the function, and a business of seven people can meet all four with what it already owns.

IN LA MACHINE Where a system could live, and where it cannot yet

The application copy read for this book shows a platform that can hold parts of a system and not the system itself (La Machine, 2026). Business → Branding at business/branding offers a logo drop zone, a brand name, a tagline, a website address, a primary and a secondary color with ten preset swatches, a choice among five fonts, an email footer and three social links, and it computes the contrast of the primary color with a pass or fail badge at the 4.5 to 1 threshold. Its values are not saved: the save button shows a confirmation and no other page uses what was entered, so the page is a place to try a palette and see its contrast, not a brand register. Business → My Website at business/website keeps a Theme & Branding panel with a gallery of twenty-eight themes, a brand name, a tagline, a logo, a favicon, a banner image, five color overrides and a heading and body font chosen from a catalog of two hundred, held in the browser. Assets → Visual Studio at business/social/visual-studio saves designs and sends exports to the Gallery. For the platform's OEM partners only, Me → Edit Styles at crm/me/styles stores light and dark colors and three fonts on the server and applies them to the partner's clients' interface, which is the one place where La Machine applies a stored identity; it is not available to an ordinary account. There is no saved brand kit that other pages use, and no brand portal or guidelines publisher.

5.7 Keeping the System Alive

A system that is not maintained stops being a system within a year, and it stops quietly. The templates are still there, the guide is still on the shelf, and the documents have begun to drift away from both. Keeping the system alive is a set of small routines rather than a project, and the routines are the same ones that keep any other part of a small business in order: a change log, a yearly review, a rule about retiring what is no longer used, and one person who owns the whole thing.

The change log is the first routine and the cheapest. It is a single page with four columns: the date, what changed, who asked and who approved. Every change at the third level and above is written into it, which in Nordec's case means every template change, every token change and every change to the guide. The log does two things that nothing else does. It answers the question of when something changed, which is the question that arrives when a document from last year is compared with one from this year. And it shows, over twelve months, where the system is under pressure: a template that is changed every month is a template that does not fit the work, and the log is what makes that visible.

The second routine is the yearly review, and it is the same shape as the reviews described in Chapter 10. The brand keeper reads the change log and counts the changes at each level. They open the asset folder and confirm that the masters are current, that the retired section holds what it should and that the licenses are still recorded. They read the guide against the templates and correct the places where the two have drifted apart. They count the exceptions granted during the year, because a rule that is excepted every month is a rule that should be rewritten. And they check the tokens against the outputs, which is the step that catches the most common failure of all: a template that was edited by hand and now holds a value of its own.

The third routine is retiring what is no longer used, and it is the routine that small businesses skip. A template that has been replaced, a color that is no longer in the palette, a lock-up that is no longer used: each of them continues to exist in the folder and in the habits of the people who used it. The remedy is a date. When a template is replaced, the old one is moved to the retired section with the date of replacement, the people who used it are told, and a date is set after which it will not be accepted. Without the date, the old template is used for another two years, and the business has two systems rather than one.

The fourth routine is the one that carries a legal dimension, and it is the reason accessibility rules belong in the tokens rather than in a paragraph of advice. The Web Content Accessibility Guidelines set a success criterion at level AA, numbered 1.4.3, under which text must reach a contrast ratio of at least 4.5 to 1, and large text — 18 points, or 14 points bold — at least 3 to 1; text that is part of a logo or a brand name has no contrast requirement (World Wide Web Consortium, 2024). A second criterion, 1.4.11 at level AA, asks that graphical objects and interface components needed to understand content reach at least 3 to 1 against adjacent colors. A third, 1.4.6 at level AAA, raises the text requirement to 7 to 1, and to 4.5 to 1 for large text. The book states these as the guidelines state them and claims no legal obligation in Canada or Québec.

The reason the rules belong in the tokens is that a token is the only place where a rule can be applied once. If the allowed pairs are recorded beside the color tokens, then every template built from those tokens carries the rule, and a person filling a template cannot produce a document that fails it. If the rule is written only in the guide, it is applied by whoever remembers it, which is the same failure mode as every other rule in this chapter. Nordec's five measured pairs are recorded with the color tokens for exactly this reason, and the pair that fails — Signal Amber text on white at 2.16 to 1 — is recorded as a prohibition rather than left to be discovered.

THE RULE Contrast rules belong in the tokens

The Web Content Accessibility Guidelines 2.2 set out success criterion 1.4.3 at level AA: text must reach a contrast ratio of at least 4.5 to 1 against its background, and large text — 18 points, or 14 points bold — at least 3 to 1, while text that is part of a logo or a brand name has no contrast requirement (World Wide Web Consortium, 2024). Success criterion 1.4.11 at level AA asks for at least 3 to 1 for graphical objects and interface components needed to understand content, and success criterion 1.4.6 at level AAA raises the text requirement to 7 to 1, and to 4.5 to 1 for large text. The rule for a design system is that these ratios are recorded beside the color tokens, as a list of pairs that are allowed and pairs that are not, rather than in a paragraph of the guide. A pair recorded with a token travels into every template built from that token; a pair written in a paragraph travels only as far as the reader's memory. The book states the guidelines as published and claims no legal obligation in Canada or Québec.

5.8 The Case's Guide and Templates

Nordec's system did not begin as a system. It began as a delivery. An independent design studio, on a fixed fee of $7,800, produced the logo, the palette, the typography, eleven templates and a guide, and presented three routes before the business chose one. The chosen route kept the letter N and drew a horizon line through it, the building seen a year ahead, and dropped the generic wrench that had been in the mark since 2022. The studio's delivery was correct and complete, and it was still a set of files. What turned it into a system was the work that followed, done by the administrator over several weeks with the guide open on one side of the desk.

The first step was the tokens. The administrator listed every color, size and spacing value that appeared in the studio's files and in the documents the business already used, which produced a longer list than the identity contained, because it included the five blues and the four typefaces that were still in circulation. The near-duplicates were removed, the remaining values were named by their role, and the twenty-one tokens were written down. The second step was the templates: the eleven files were saved in a folder the administrator controlled, with the components already placed and the fields left empty, and the older documents were moved to a retired section rather than deleted. The third step was the guide, which was reorganised around the questions people ask rather than around the order of the design work, and a one-page summary was produced from it.

The first template the technicians used was the visit sheet, and it is worth describing because it shows what a template does that an instruction cannot. The sheet had existed for four years as a form the technicians filled in their own way: some wrote the readings in the margin, some wrote the time of arrival at the top, some wrote nothing about the parts they had used. The templated version placed the fields in a fixed order — the building, the date, the time of arrival, the work done, the parts used, what comes next — with the record codes set in the monospaced face so that they could be read back over the telephone. The technicians did not need to be trained in the new sheet. They needed to be shown it once, and the sheet itself carried the rest.

The one-page summary is the part of the system that travels furthest. It is taped inside each of the two vans, beside the door, where a technician can see it while standing on the step. It carries the logo lock-up, the five colors with their codes, the two typefaces, the tagline « Votre bâtiment, un an d'avance. » and the name of the brand keeper. Its purpose is not to make the technicians into designers. It is to answer the three questions that arrive when somebody is holding a marker, a garment or a sign and needs an answer in ten seconds: which blue, which typeface, who do I ask. The summary is the reason the system reaches the vans at all, because a twenty-four-page guide does not fit in a van and a one-page summary does.

Six months after the rollout, the touchpoint inventory was taken again. Thirty-three of the thirty-six touchpoints were consistent with one another, against thirteen before. The three that were not were known and named: a sign at the depot that had been ordered before the guide was finished, a supplier's garment that carried the manufacturer's own mark under the manufacturer's rules, and a photograph on the website that had been taken before the new vans were wrapped. None of the three was a surprise, and each had a date by which it would be corrected. That is what a system looks like in a small business: not perfection, but a short list of known exceptions with dates attached, and one person who holds the list.

CASE IN POINT Nordec Entretien inc.: the summary taped inside the van

Nordec Entretien inc. is a fictional business created for this encyclopedia; its owner, its employees, its customers, its competitors, its figures and the events described are invented, and any resemblance to a real business is coincidental. The one-page summary was produced in an afternoon from the twenty-four-page guide, and it was the part of the system that the technicians used first. It was taped inside each of the two vans, beside the door, at a height where it can be read while standing on the step. Within a month, the questions that had previously arrived by telephone — which blue for a sign, which typeface for a label, who to ask about a garment — were answered from the sheet. The full guide was opened perhaps once a month, by the administrator, for the questions the summary does not cover. The lesson is not that the summary is better than the guide. It is that the two documents have different jobs, and a system that produces only the longer one will find that its rules are applied from memory, which is how the five blues appeared in the first place.

Costly Mistakes

• Delivering files and calling it a system. A set of logos, colors and layouts is the result of decisions, not the decisions themselves, and without tokens, templates and a guide the business re-takes every decision in every document.
• Naming tokens after their look instead of their role. A token named after its color has to be renamed the day the color changes, which is the day the system was supposed to make cheap.
• Writing a guide for the designer rather than the reader. A guide that presents the design instead of answering the questions people ask is admired once and then closed, and the questions are answered from memory.
• Leaving the guide out of date. The first person who finds a template in the guide that no longer matches the template in the folder stops trusting the whole document, and the drift that follows is faster than the drift the guide was built to stop.
• Letting anyone write in the asset folder. A folder that several people can edit fills with exports saved over masters, files named final and final-latest, and versions that cannot be told apart.
• Deleting retired files instead of moving them. Old documents exist in customers' hands, on buildings and on vans, and a business that cannot identify the file a document came from cannot answer the simplest question a printer can ask.
• Requiring the owner's approval for every document. A governance rule that is too heavy is not followed, and the documents are produced without approval, which is the outcome the rule was written to prevent.

Key Takeaways

• A brand identity arrives as a set of files, and a system is the set of decisions behind those files written down in a form that people who are not designers can apply; governance is the fifth phase of the identity process, described as the managing of assets (Wheeler & Meyerson, 2024).
• Design tokens are named decisions about color, type, spacing and logo size, named by their role rather than their look, and Nordec's system holds twenty-one of them in four families; the vendor-neutral format reached its first stable version, 2025.10, on October 28, 2025, as a Community Group report and not a W3C Standard (Design Tokens Community Group, 2025).
• Components and templates are built from small reusable pieces upwards, in the five levels of atoms, molecules, organisms, templates and pages (Frost, 2016), and Nordec's nine components carry eleven templates and thirty-six touchpoints.
• The guide records the system for its readers: Nordec's runs to twenty-four pages in nine parts, with a one-page summary that is taped inside each van and answers the questions that arrive in ten seconds.
• Governance is a short list of levels — filling a template, assembling a new document, changing a template, changing an identifier, changing the platform — and the brand keeper holds the first three while Sophie approves the promise and the identifiers.
• The asset folder holds one folder per identifier, masters separated from exports, a retired section and file names with version and date, with licenses recorded beside the files (SIL International, 2007); the contrast rules live in the tokens, where they reach every template (World Wide Web Consortium, 2024).

Your Action Plan

1. List every value in the identity and name each one by its role. Write down every color, type size, spacing and logo size in use, remove the near-duplicates, and give each remaining value a name that describes what it is for rather than what it looks like. Deliverable: a token list of twenty or so named values. Verification: no template holds a color code or a size of its own, and every token name would survive a change of its value.
2. Point the templates at the tokens and separate the masters from the exports. Save one editable file per template with the components already placed, keep the masters apart from the exports, and move the older documents to a retired section rather than deleting them. Deliverable: a folder with one file per template and a retired section. Verification: the newest version of each piece is the last file in its group, and its name carries the version and the date.
3. Write the guide around the questions people ask, and produce the one-page summary. Organize the guide by the reader's question rather than by the design process, and produce a single sheet with the lock-up, the colors, the typefaces, the tagline and the brand keeper's name. Deliverable: a full guide and a one-page summary. Verification: a person who did not take part in the design can answer the three most common questions from the summary alone.
4. Write the levels of change and name the brand keeper. Set out the five levels, from filling a template to changing the promise, say who approves each one, and name the person who holds the templates and the asset folder. Deliverable: a one-page governance list and a named brand keeper. Verification: every change made in the last six months can be placed at a level and matched to an approver.
5. Open the change log and book the yearly review. Start a single page with the date, what changed, who asked and who approved, and place the yearly review on the calendar with its tasks: read the log, check the folder, compare the guide with the templates and count the exceptions. Deliverable: a change log with its first entries and a dated review. Verification: the review has a date within twelve months and the log has an entry for every change at the third level and above.

In La Machine

The application copy read for this book shows a platform that holds parts of a design system and not the system itself (La Machine, 2026). Business → Branding at business/branding offers a logo drop zone, a brand name, a tagline and a website address, a primary and a secondary color with ten preset swatches, a choice among five fonts, an email footer and three social links, and it computes the contrast of the primary color with a pass or fail badge at the 4.5 to 1 threshold, with a live preview of a client invitation card and a client email. Its values are not saved: the save button shows a confirmation and no other page uses what was entered, so the page is a place to try a palette and see its contrast rather than a brand register. Business → My Website at business/website, with its builder and preview routes, keeps a Theme & Branding panel with a gallery of twenty-eight themes, a brand name, a tagline, a logo, a favicon, a banner image, five color overrides and a heading and a body font chosen from a catalog of two hundred, held in the browser. Assets → Visual Studio at business/social/visual-studio is an editor for social, video, print and web formats whose designs are saved and whose exports go to the Gallery, and none of those pages reads a stored brand palette or logo.

What the platform does not have belongs in the same breath, because it is the reason this chapter's system lives outside it. There is no saved brand kit that other pages use, no brand portal or guidelines publisher, no digital asset management with versions, usage rights or expiry dates, and no brand consistency audit. The one place where La Machine applies a stored identity is Me → Edit Styles at crm/me/styles, which stores light and dark colors and three fonts on the server for each white-label site and applies them to the partner's clients' interface; it is available to the platform's OEM partners only and not to an ordinary account. For everyone else, the tokens, the templates, the asset folder and the change log are kept in the applications the business already uses, and the guide is a document rather than a page. The website update that carries the identity onto the screen is the work of Book C2 (Websites, Landing Pages and Conversion), and the recurring checks that keep the system current are the work of Chapter 10.

Self-Check

1. Why is a set of delivered files not a system, and what are the four layers of the system shown in Figure 5.1?
2. What is a design token, why is it named by its role rather than its look, and what did the design tokens specification reach on October 28, 2025?
3. What are the five levels of atomic design, and how do they map onto a small business's components and templates?
4. What does a brand guide contain, and why does a business need both a full guide and a one-page summary?
5. What are the five levels of change in a brand, who approves each one, and where do the contrast rules live?

Answers. 1. A set of files is the visible result of decisions and cannot explain them, so the decisions are re-taken in every document; the system has four layers, from the twenty-one design tokens at the bottom, through the components built from them and the eleven templates, to the thirty-six touchpoints at the top, with the twenty-four-page guide and the brand keeper governing all four. 2. A design token is a named decision about a visual value, written once and referred to by name everywhere it is used; it is named by its role so that the name survives a change of value, and on October 28, 2025 the design tokens specification reached its first stable version, numbered 2025.10, published by the Design Tokens Community Group of the World Wide Web Consortium as a Community Group report and not a W3C Standard (Design Tokens Community Group, 2025). 3. The five levels are atoms, molecules, organisms, templates and pages (Frost, 2016); in a small business the atoms are the color values, the type sizes and the letter N, a molecule is the logo lock-up, an organism is the header of the monthly report, a template is the report with its fields empty and a page is that report filled in for one building in one month. 4. A guide records the platform, the voice, the logo system, color and contrast, typography, imagery and icons, the templates, the vehicles, uniforms and signs, and the approvals and contacts; the full guide answers the harder questions and the one-page summary answers the three questions that arrive in ten seconds, which is why the summary is the document that gets used most. 5. The five levels are filling a template, assembling a new document from existing components, changing a template, changing an identifier such as a token or a lock-up, and changing the platform such as the promise or the tagline; the brand keeper holds the first three, Sophie approves the identifiers with the studio and the platform with the work of Chapter 2, and the contrast rules live beside the color tokens, where they reach every template built from them (World Wide Web Consortium, 2024).

Chapter References

Charter of the French language, CQLR c C-11 (Québec). LégisQuébec, updated to June 10, 2026 — as read for Book C2 (Websites, Landing Pages and Conversion); its rules on trademarks, signs, products and business names are stated in this book through the explanations of the Office québécois de la langue française and of the Registraire des entreprises.

Design Tokens Community Group. (2025, October 28). Design tokens specification reaches first stable version. World Wide Web Consortium. Retrieved September 25, 2026, from https://www.w3.org/community/design-tokens/2025/10/28/design-tokens-specification-reaches-first-stable-version/

Frost, B. (2016). Atomic design. Brad Frost. https://atomicdesign.bradfrost.com/

La Machine. (2026). Documentation — route reference inventory. Internal product documentation, read September 25, 2026, together with the code of the pages this book describes: business/branding, business/website, business/social/visual-studio and crm/me/styles.

SIL International. (2007). SIL Open Font License (Version 1.1, February 26, 2007). Retrieved September 25, 2026, from https://openfontlicense.org/open-font-license-official-text/

Wheeler, A., & Meyerson, R. (2024). Designing brand identity: A comprehensive guide to the world of brands and branding (6th ed.). Wiley.

World Wide Web Consortium. (2024). Web Content Accessibility Guidelines (WCAG) 2.2. W3C Recommendation, 12 December 2024 (first published as a Recommendation on 5 October 2023). Retrieved September 25, 2026, from https://www.w3.org/TR/WCAG22/

Chapter 6 — Brand Architecture

As soon as a business sells more than one thing, it must decide how its names relate — one master brand, sub-brands, endorsed brands or separate brands — and that decision, brand architecture, fixes how far each reputation can carry the others.

What You'll Be Able to Do

• Explain what brand architecture decides, using the idea of an organizing structure that gives every brand in a portfolio a role and a stated relationship to the others.
• Place any offer on the brand relationship spectrum, from a single master brand through sub-brands and endorsed brands to a house of brands.
• Choose between a branded house, a house of brands and the positions in between by weighing what each one suits, what it costs and what it risks.
• Name a new product, tier or service under the master brand, in French first in Québec, following the rules that apply to business names and to trademarks on products.
• Use another company's mark — a certification, a partner logo or a co-brand — only with permission and under the owner's own rules.
• Document the architecture on a one-page map with rules for new names, and decide what to do with a name that arrives through an acquisition.

Why It Matters

A business that sells one thing under one name has no architecture problem, and that is why the problem is so often discovered late. The moment a business adds a tier, an entry offer, a service line, a second territory or an acquired firm, a question arrives that nobody asked out loud: does this new thing carry the same name as everything else, does it carry a name of its own, or does it carry a name that leans on the first one? The answer can be given deliberately, in an afternoon, with a page of writing to show for it. It can also be given by default, one purchase order at a time, and the default answer is almost always the expensive one, because a business that never decides ends up with several names, several visual identities and one reputation that is asked to carry all of them.

The business followed through this book reached the question with three tiers, an entry offer, a guarantee and a single name. Nordec Entretien inc. sells Essential, Preventive and Portfolio maintenance, a fixed-price assessment called « Bilan Nordec », and a 90-minute guarantee attached to its contract work. Every one of those offers is sold under the word Nordec, and that was a decision rather than an accident. The alternative was on the table: give the assessment its own name and its own identity, so that it could stand apart from the maintenance contracts and be sold to buildings that had not signed anything. The case rejected that option, and the reason is the reason this chapter exists. A separate name would have had to earn its own recognition among the roughly 1,400 buildings in the two territories, with no budget to earn it, while the Nordec name was already opening doors: of the fifteen proposals the business decided in the six months before the refresh, three began with a Building Check letter signed by Nordec.

The cost of getting this wrong runs in both directions, and a small business can fail either way. A business that splits too early spends a small budget building recognition twice, and neither name ever becomes strong enough to be remembered without help. A business that stretches one name over everything it can sell ends up with a promise so broad that it stops meaning anything, and with a failure in one line of work landing on the reputation of all the others. Between those two failures sits a decision that is made with a small number of questions, and the rest of this chapter is those questions, the positions they lead to, and the way to write the answer down so that the next offer, the next hire and the next acquisition all land in the right place.

In This Chapter

Section 6.1 explains what brand architecture decides and why a brand portfolio is managed as a system rather than as a set of separate names. Section 6.2 sets out the brand relationship spectrum, the positions that run from a single master brand to a house of brands, and the benefits a coherent architecture brings. Section 6.3 examines the branded house, the house of brands and the middle positions, with their costs and their risks. Section 6.4 covers the naming of products, tiers and services under a master brand, including the French-first rules that apply in Québec. Section 6.5 deals with endorsements, partner and certification marks, and co-branding. Section 6.6 covers what happens to a name that arrives through growth or acquisition. Section 6.7 gives the criteria and the one-page map that turn the decision into a rule. Section 6.8 applies all of it to the case.

6.1 What Brand Architecture Decides

Brand architecture is the organizing structure of a brand portfolio, and the phrase is worth taking apart because both halves do work. It is a structure, which means it is written down and it holds across offers, years and people. It organizes a portfolio, which means it assumes that a business may own more than one brand and that those brands stand in some relation to one another. The relation is the substance of the thing. Aaker and Joachimsthaler describe brand architecture as an organizing structure of the brand portfolio that specifies brand roles and the relationships between brands, and they treat the choice of that structure as one of the central challenges a business faces as it grows (Aaker & Joachimsthaler, 2000). A business with one name and one offer has an architecture; it is simply the shortest one possible, and it will be tested the first time a second offer appears.

The reason architecture deserves a chapter of its own is that it is not the same question as positioning, and it is not the same question as the offer. Book A3 (Strategy and Competitive Positioning) chose where Nordec competes: two territories, mid-size buildings, owner-operators and small portfolios, with a set of deliberate refusals. Book A6 (Designing a Value-Added Offer) designed what Nordec sells: three tiers, two add-ons, a guarantee, an exit in the first ninety days and a fixed-price assessment. Neither of those books asked how the names of those things relate to one another, and neither of them had to. That question belongs here, and it arrives after the offer exists, because architecture is a decision about names and reputations rather than about features and prices. Book A7 (Pricing Strategy and Revenue Models) owns what each tier costs, and this chapter does not restate a single price.

The second half of the idea comes from the same body of work. Managing brands as a system, rather than one at a time, is one of the concerns that Aaker sets out in the book that distinguishes brand identity from brand position: identity is what the strategist aspires the brand to be, and position is the part of that identity which is actively communicated (D. A. Aaker, 1996). The distinction matters to architecture because a portfolio has one identity at the top and several positions underneath it. Nordec's identity — calm, precise, plain-spoken, neighborly, and the promise that the business knows the building and tells the owner ahead — is the same whether a customer is reading a Preventive contract or a Bilan Nordec report. What changes is the part of that identity each offer puts forward. The Preventive tier leads with the 90-minute response; the assessment leads with the capital planning note. One identity, several positions, one name.

The same book offers four perspectives from which a brand can be considered: as a product, as an organization, as a person and as a symbol (D. A. Aaker, 1996). Those four perspectives are useful here because they explain why a single name can carry several offers without becoming incoherent. A customer who hires Nordec is buying a service, which is the product perspective; a company with seven people and a depot on Montréal's north shore, which is the organization perspective; a way of behaving that customers described in interviews as on time, straight and friendly, which is the person perspective; and a letter N with a horizon line drawn through it, which is the symbol perspective. An architecture decision is, in practice, a decision about which of those four perspectives a new offer will lean on. A new offer that leans on the organization and the person can usually share the master brand. A new offer that leans on none of them is a candidate for a name of its own.

Architecture answers five practical questions, and a business that can answer all five has an architecture whether or not it has written one down. The first is how many names the business will maintain, because every name costs money to make known and attention to keep consistent. The second is what each name covers, which is the question of scope: does the name cover a tier, a service, a territory, a customer type or a whole company? The third is which name leads when two of them appear together, because a document that shows two logos of equal size has answered that question by accident. The fourth is what happens when a new offer arrives, which is the question that turns architecture from a description into a rule. The fifth is what happens when a business is acquired or acquires, which is the question that most often forces a small business to think about architecture for the first time.

DEFINITION Master brand, sub-brand, endorsed brand and house of brands

The master brand is the name that leads: it is the name a customer uses to ask for the business, and the name under which the promise is made. A sub-brand is a name that belongs to an offer and is connected to the master brand, so that the offer has some standing of its own while the master brand still carries it. An endorsed brand is a name that leads on its own, with the master brand appearing behind it as a guarantee — the shape a business uses when it wants a separate name but not a separate reputation. A house of brands is a portfolio in which several brands lead in their own markets with little or no visible connection to one another. A branded house is the opposite end: one master brand, with descriptive names underneath it for each offer. The brand relationship spectrum is the range of positions between those two ends, and it is the tool this chapter uses to place a business on it (Aaker & Joachimsthaler, 2000).

6.2 The Brand Relationship Spectrum

The brand relationship spectrum is the tool that Aaker and Joachimsthaler set out for the architecture challenge, and its value is that it replaces a binary argument with a range (Aaker & Joachimsthaler, 2000). The binary argument is the one a small business usually has: either everything carries the company name, or the new thing gets a name of its own. The spectrum says that those are two ends of a line with positions in between, and that most businesses belong somewhere in the middle once they have more than one offer. The positions run from a single master brand, through sub-brands, through endorsed brands, to a house of brands in which the individual names carry everything and the parent is invisible. The article's own vocabulary for the positions in between is detailed, and this book does not reproduce it; what matters for a small business is the direction of travel and the questions that decide how far along the line to go.

The direction of travel is easy to state. Moving toward the branded-house end means putting more weight on one name: recognition accumulates in one place, every offer benefits from every other offer's reputation, and the cost of making the name known is paid once. Moving toward the house-of-brands end means putting more weight on separate names: each offer can speak to its own customer in its own voice, a failure in one line does not land on the others, and a business can sell or retire one brand without disturbing the rest. Neither direction is better in the abstract. The question is which of those two sets of consequences a particular business, at a particular size, with a particular budget, can afford.

The article names what a coherent architecture brings: impact, clarity, synergy and leverage. It also names what an incoherent one produces instead: weakness, confusion, waste and missed opportunities (Aaker & Joachimsthaler, 2000). Those eight words are worth reading as a checklist rather than as a slogan, because each of them describes something a small business can observe in its own materials. Impact is whether the name registers at all when a customer sees it. Clarity is whether a customer can tell what the business sells and what each offer is. Synergy is whether the offers make each other stronger — whether a satisfied Preventive customer is more likely to buy the assessment, and whether an assessment customer is more likely to sign a contract. Leverage is whether the reputation built in one place can be spent in another. The four failures are the same four read backwards, and the case followed through this book had all of them in a mild form before the refresh: three versions of the logo in circulation, five different blues, four typefaces, and only thirteen of thirty-six touchpoints consistent with one another.

The spectrum is drawn in Figure 6.1. The figure shows a horizontal line with four positions marked along it, from left to right: a branded house, sub-brands, endorsed brands and a house of brands. Under each position, one line says how the names relate — one name over everything, a master name with offer names attached to it, offer names leading with the master name behind them, and separate names with no visible parent. Nordec is marked at the branded-house end of the line, with a note that its tiers and its assessment carry descriptive names under the single word Nordec. At the other end, the regional integrator is shown as an example of a group that runs several service lines under separate names, which is a shape that suits a large firm buying and selling businesses and does not suit a business with one reputation to build.

Figure 6.1 — The brand relationship spectrum. A small business with one reputation to build belongs at the branded-house end until a real reason moves it.

Reading the spectrum correctly requires one caution, which is that the positions describe how names relate and not how good a business is. A house of brands is not more sophisticated than a branded house, and a branded house is not more disciplined than a house of brands. Each position is a way of spending a finite budget of attention, and the budget is what decides. A business with one marketing person, one website and one van cannot make four names known in a market of 1,400 potential customers. A business with several hundred million dollars of revenue and a portfolio of companies bought over decades can, and does, and the shape of its portfolio reflects that. The mistake a small business makes is copying the visible shape of a large company's portfolio without the budget that makes the shape work.

The spectrum also has a temporal dimension that is easy to miss. A business can move along it in either direction, and moves are usually triggered by events rather than by strategy reviews: an acquisition, a new service line that does not fit the promise, a partnership that requires a separate identity, or a decision to sell part of the business. Because moves are triggered by events, the useful preparation is not a five-year plan but a rule: a short statement of where the business sits on the spectrum today, and what would have to be true for it to move. Nordec's rule, written on the architecture map described in section 6.7, is that a new offer carries a descriptive name under the master brand unless it serves a different customer with a different promise, and that the question is decided by the owner rather than by whoever is closest to the new offer.

WHAT CHANGED From the brand system to the spectrum

The idea that a business manages a portfolio of brands rather than a single brand was set out in Aaker's 1996 book, which treats managing brands as a system as one of its concerns and distinguishes brand identity from brand position (D. A. Aaker, 1996). Four years later, the same author and Joachimsthaler published the article that gave the idea its working tool: the brand relationship spectrum, with brand architecture defined as an organizing structure of the brand portfolio that specifies brand roles and the relationships between brands, and with sub-brands and endorsed brands placed at the center of the architecture challenge (Aaker & Joachimsthaler, 2000). The change is one of usable detail rather than of principle. The 1996 book says that brands in a portfolio have to be managed together; the 2000 article says what the possible relationships are, what each one costs and what each one buys, which is what a business needs in order to choose. For a small business, the practical consequence is that the question "should this have its own name?" now has a place to be answered, and the place is a line with four positions on it rather than a yes or a no.

6.3 Branded House, House of Brands and the Middle

The branded house is the position in which one master brand covers everything, and the offers underneath it carry descriptive names rather than names of their own. It suits a business whose offers serve the same customers with the same promise, and it suits a business whose budget for making a name known is small. Its cost is that everything the business does lands on the same reputation: a badly handled job in one service line is a badly handled job for the whole company, and an offer that does not fit the promise drags the promise out of shape. Its risk is the opposite of its strength. Because the master brand carries everything, a business that keeps adding offers under it eventually reaches a point where the name means so many things that it stops meaning any of them, and the promise that made the name worth having is no longer visible in it.

The house of brands is the position in which several names lead in their own markets with little or no visible connection between them. It suits a business that serves genuinely different customers with genuinely different promises, and it suits a business that grows by buying other businesses and intends to keep them. Its cost is that every name must be made known separately, which means a separate budget, a separate identity, separate templates and separate attention. Its risk is fragmentation: a business with four names and one marketing person usually has four weak names, and the reputation that each one has earned is not available to the others. The middle positions exist precisely because most businesses want some of each. A sub-brand gives an offer a name of its own while keeping the master brand visible and in charge. An endorsed brand lets an offer lead in its own market while the master brand stands behind it as a guarantee, which is the shape that makes a transition possible when a business buys a firm whose name its customers know.

For a small business, the branded-house end is usually the right place, and the reason is arithmetic rather than taste. Almost every business that will read this chapter is small: a business with a handful of employees, one website, one or two vehicles and one marketing budget. A business of that size has one budget for making a name known and a small number of people who can keep a visual identity consistent. Splitting a name in two splits that budget in two, and the second name starts with no recognition at all. The case followed through this book made exactly that calculation: a separate brand for the assessment would have had to earn its own recognition among the roughly 1,400 buildings in the two territories, with no budget to earn it, while the master brand was already opening doors.

The middle positions are not a compromise for businesses that cannot decide. They are the correct answer in specific circumstances, and the circumstances can be named. A sub-brand is right when an offer serves the same customers as the master brand but needs to be talked about as a distinct thing — a tier, a service line, a program. An endorsed brand is right when a name arrives from outside, through an acquisition or a partnership, and its customers would be alarmed by its disappearance. A house of brands is right when the offers genuinely serve different markets, and when the business can fund each name separately. What is not right is to arrive at a house of brands by accident, one new name at a time, each one added because the person closest to the offer wanted a name of their own.

The following table sets out the four positions on the spectrum with the way names relate in each, what each position suits, what it costs and what it risks. Read it as a comparison rather than as a ranking: the four rows describe four ways of spending the same budget of attention, and the right row for a business is the one whose costs it can carry and whose risks it can survive.

Table 6.1 — Four positions on the spectrum

KEY CONCEPT One reputation or several

The whole of brand architecture reduces to a single question: how many reputations is this business able to build? A reputation is not a name and not a logo; it is what a customer remembers about a business and expects from it the next time. Building one takes years of consistent behavior and consistent presentation, and it is paid for with attention, money and repetition. A business that builds one reputation can spend every repetition on the same memory, and each offer it sells adds to that memory rather than starting a new one. A business that builds several must divide the same repetitions between them, and the division is rarely even, because the newest name always needs the most. That is why the branded-house end of the spectrum is the default for a small business and why moving along the spectrum is a decision that has to be justified by something more than the wish for a fresh name. The question is not which position sounds more sophisticated. The question is how many reputations the business can actually afford to build, and the honest answer for most small businesses is one.

WATCH OUT A new brand for every new offer

The trap is quiet and it is almost always well intentioned. A business adds a service line, and the person who will run it wants a name, a logo and a color of its own, because the new line is different and deserves to be seen as different. The request is granted, because refusing it feels like refusing the new line. A year later the same thing happens with a second offer, and then with a partnership, and the business finds itself with four names, four sets of documents and one marketing budget that was never increased. Nothing about this sequence looks like a mistake at any single step, and that is what makes it dangerous. The remedy is a rule that exists before the request arrives: a new offer carries a descriptive name under the master brand unless it serves a different customer with a different promise, and any exception is decided by the owner and written on the architecture map with its date and its reason. A rule of that kind costs nothing to write and saves a business from discovering, three years later, that it has been paying to make four names known and has made none of them known well.

6.4 Naming Products, Tiers and Services

Once the position on the spectrum is settled, the naming work underneath it is mostly descriptive, and descriptive is a compliment rather than a criticism. A descriptive name says what the offer is, in ordinary words, so that a customer who has never heard of the business can understand the offer from the name alone. Nordec's three tiers are named this way: Essentiel, Préventif and Portefeuille in French, Essential, Preventive and Portfolio in English. Each name tells a building owner what kind of arrangement it is, and none of them requires a story to explain. The entry offer is named the same way: « Bilan Nordec » in French and "Nordec Building Check" in English, a name that says what the customer receives and carries the master brand inside it. The guarantee is not named at all; it is described as the 90-minute guarantee, which is what it is.

The alternative to a descriptive name is a suggestive one, and the research on suggestiveness is worth knowing before a business reaches for it. Keller, Heckler and Houston found that a brand name which suggests a benefit improves recall of advertised claims consistent with that benefit, but lowers recall of claims about other benefits (Keller et al., 1998). The finding is narrow and its consequence is practical. A name that suggests speed will help a customer remember that the business is fast, and it will make it harder for the same customer to remember that the business also keeps a building health record and writes a capital planning note. For a business with one benefit to sell, suggestiveness is an asset. For a business with several benefits, and most service businesses have several, a descriptive name leaves the room open. Nordec's promise — that the business knows the building and tells the owner ahead — has two halves, and a name that suggested only one of them would have tilted the whole identity toward that half.

Descriptive naming also has a practical advantage that has nothing to do with memory: it makes the naming decision repeatable. A business that names its offers descriptively has a rule it can apply to the next offer without a workshop, and the rule produces names that sit together without competing. A business that names its offers suggestively has to invent a story each time, and the stories accumulate into a portfolio of names that no longer sound like they belong to the same company. The case followed through this book generated twelve candidates for the entry offer, of which seven passed a screen for language and meaning — they had to work in French and in English — and four passed a knockout search of the trademark databases. The choice was descriptive, and the four that survived the search were the ones that said plainly what the offer was.

The French-first rule in Québec is not a stylistic preference, and it shapes the naming work from the beginning rather than at the end. The Registraire des entreprises explains that a business name's generic term must be French, and that it is mandatory when the specific element is in another language; where the generic term is the same in both languages, it comes first, so that "TRANSPORT FASTER" is correct and "FASTER TRANSPORT" is not (Registraire des entreprises, 2025). The same guide notes that a business may declare versions of its name in other languages, and that a trademark in another language may be declared as another name without a French generic term only if it is a Canadian trademark registered with the Canadian Intellectual Property Office, with no French version, and shown with its symbol. The guide also states plainly that the Registraire is not the authority on signs, which is the Office québécois de la langue française, and that the guide is information rather than a legal interpretation (Registraire des entreprises, 2025).

The rules on products run alongside the rules on names, and they apply to the words inside a trademark rather than to the trademark itself. Since June 1, 2025, the generic or descriptive words inside a trademark on a product — its nature, its ingredients, its color, its scent — must appear in French, while the product and company names may stay in another language; until June 1, 2027, some products made before June 1, 2025 may still be sold (Office québécois de la langue française, 2025b). For a service business the practical reach of this rule is narrow, because a service business has few products, but the principle behind it is the one that governs the whole naming exercise in Québec: the descriptive layer of a name is the layer that has to be in French, and the distinctive layer is the layer that may be in another language. Nordec's tiers follow that principle exactly. The descriptive word is French first — Essentiel, Préventif, Portefeuille — and the English version exists for documents that are read in English.

The following table sets out the names Nordec uses under its single master brand, with the French and English forms, the type of name each one is and the rule that governs it. Read it as the naming layer of the architecture: every row is an offer, every offer carries a descriptive name, and the master brand appears in the rows where the offer needs to be identified as Nordec's rather than as a generic category.

Table 6.2 — Nordec's names under one brand

QUÉBEC & CANADA NOTE French names for tiers and services

In Québec, the naming layer of an architecture is governed before it is designed. The Registraire des entreprises states that a business name's generic term must be French and that it is mandatory when the specific element is in another language; where the generic term is the same in both languages it comes first, so that "TRANSPORT FASTER" is correct and "FASTER TRANSPORT" is not; a business may declare versions of its name in other languages; and a trademark in another language may be declared as another name without a French generic term only if it is a Canadian trademark registered with the Canadian Intellectual Property Office, with no French version, shown with its symbol (Registraire des entreprises, 2025). On products, since June 1, 2025, the generic or descriptive words inside a trademark must appear in French, while product and company names may stay in another language, with some products made before that date sellable until June 1, 2027 (Office québécois de la langue française, 2025b). The Registraire is not the authority on signs; the Office québécois de la langue française is. This book explains these rules and does not replace the advice of a lawyer or a trademark agent.

6.5 Endorsements, Partners and Co-Branding

An endorsement is a name standing behind another name, and it is the position on the spectrum that most small businesses meet without planning for it. The meeting usually happens in one of three ways. A business buys another firm and wants to keep the acquired name for a while. A business joins a program run by a manufacturer or a franchisor and is given a mark it may display. A business works with a partner on a joint offer and the two names appear together on the same document. In each case the business is being asked to place a second name next to its own, and the question is which one leads, which one follows, and under what rules either may be used.

The certification mark is the most common of the three and the easiest to get wrong. A manufacturer's certified-installer mark is not the business's asset; it belongs to the manufacturer, and the business displays it under the manufacturer's own usage rules. Those rules typically say where the mark may appear, at what size relative to the business's own logo, in what colors, with what wording beside it, and for how long it may be displayed after the certification lapses. A business that treats a certification mark as part of its own identity will eventually break one of those rules, and the consequence is not a design problem but a permission problem. The case followed through this book handles this by allowing a manufacturer's certified-installer mark on the website only, and only under the manufacturer's own usage rules, which keeps the mark out of the templates, the vans and the printed documents where its placement would be governed by nobody.

Partner marks and co-branding raise the same question with a different owner. When two businesses present an offer together, both names appear, and the document has to say which one leads. The usual answer is that the name of the business the customer is buying from leads, and the partner's name appears in a supporting position with a line explaining the relationship. The answer is not a matter of taste; it is a matter of what the customer will remember and who will be called when something goes wrong. A document that shows two logos at the same size has told the customer that either business may be called, which is rarely what either business wants. The same logic applies when a customer's name appears on a case study or a reference: the customer's mark is used with the customer's permission, in the form the customer supplies, and the permission is recorded rather than assumed.

The legal layer under all of this is short and worth stating plainly. The owner of a registered trademark controls its use for the goods and services of the registration, which means another company's logo is used only with that company's permission and under its rules. The Trademarks Act sets out the framework for registration in Canada, including that a person may apply to register a trademark for goods or services if they use it or propose to use it and are entitled to use it in Canada, that the goods or services are stated in ordinary commercial terms, that they are grouped by class of the Nice Classification with each group preceded by its class number, and that a registration lasts ten years from the date of registration and is renewable for further periods of ten years (Trademarks Act, ss. 30 and 46). A trademark agent assesses registrability and the risk of confusion, and that assessment is not something a business can perform for itself by reading a database.

The practical consequence for architecture is that a business should keep a short file of every mark it displays that it does not own. The file holds the name of the owner, the permission or the program under which the mark may be used, the usage rules, the date the permission began and the date it ends or is reviewed. In the case followed through this book, that file is part of the trademark file that the administrator keeps, and it is read at the same time as the rest of the brand materials. The file is short because the business displays few marks it does not own, and it is worth keeping because the alternative is discovering, at the moment a certification lapses, that the mark has been printed on two vans and eleven templates.

THE RULE Another company's mark is used on its terms

The owner of a registered trademark controls its use for the goods and services of the registration, and the Trademarks Act sets out the framework: a person may apply to register a trademark for goods or services if they use it or propose to use it and are entitled to use it in Canada, the goods or services are stated in ordinary commercial terms and grouped by class of the Nice Classification with each group preceded by its class number, and a registration lasts ten years from the date of registration and is renewable for further periods of ten years (Trademarks Act, ss. 30 and 46). The rule that follows for a small business is simple to state and easy to break: a partner's, a manufacturer's or a customer's mark is displayed only with that owner's permission and under that owner's own usage rules, at the size, in the colors and in the position those rules specify, and for as long as the permission lasts. A trademark agent assesses registrability and the risk of confusion, and this book explains the law rather than replacing the advice of a lawyer or a trademark agent.

6.6 Architecture After Growth or Acquisition

Growth changes an architecture in two ways, and only one of them is usually planned for. The first is internal: the business adds an offer, a tier or a service line, and the question is whether the new thing carries the master brand. The second is external: the business acquires another firm, or is acquired, and the question is what happens to the name that arrives. The internal case is answered by the rule described in section 6.7. The external case is answered by a choice among three options, and the choice is worth making before the acquisition rather than during it, because during an acquisition every party has an opinion and none of them has time.

The first option is to absorb the name. The acquired firm's name disappears, its customers are told that they are now served by the acquiring business, and everything is presented under the master brand from a stated date. Absorption is the cheapest option to run and the fastest to complete, and it is the right answer when the acquired firm's name carries little recognition, when its customers buy the same kind of work the acquiring business already sells, and when the acquiring business can serve them without changing anything. Its cost is paid at the moment of the change, because a customer who has dealt with the same firm for years and receives a letter from a name they have never heard may take the opportunity to look around. The remedy is a transition period with a letter, a named contact and no change in the service, which is the same remedy the case followed through this book uses for its own refresh.

The second option is to endorse the name for a transition. The acquired name continues to lead for a stated period, with a line underneath it saying that the firm is now part of, or a company of, the acquiring business. This is the endorsed-brand position on the spectrum, used as a bridge rather than as a destination. It costs more than absorption because two identities are maintained for the length of the transition, and it buys the one thing absorption cannot buy: time for the acquired firm's customers to connect a name they know to a name they do not. The transition needs an end date, written down, because an endorsement that is never withdrawn becomes a permanent second identity and the business pays for two names indefinitely.

The third option is to keep the name. The acquired firm continues under its own name with no visible connection to the acquiring business, which is the house-of-brands position. It is the right answer when the acquired firm serves genuinely different customers with a genuinely different promise, when its name is strong enough in its own market to be worth maintaining, and when the acquiring business is prepared to fund a separate identity, a separate website and separate attention. It is the wrong answer when it is chosen because nobody wanted to have the conversation, which is how many small acquisitions end up with two names, two sets of documents and one person trying to keep both consistent.

The regional integrator described in this book's case material runs several service lines under separate names, and the shape suits it. A large facility-services firm that buys and sells businesses needs names that can be detached from the group and sold, and a house of brands gives it that. Nordec has one reputation to build and no portfolio to trade, so the same shape would cost it recognition it cannot afford to lose. The contrast is worth keeping in view, because it shows that the same architecture question has different answers at different sizes, and that the answer follows from what the business does with its names rather than from what sounds impressive.

If Nordec ever acquired a small firm of trades in a third territory, the three options would be on the table in the order above. Absorbing the name would be the cheapest and would put the acquired customers under the promise Nordec already makes. Endorsing it for a transition — "a Nordec company" — would keep the acquired name visible while the connection was established, at the cost of maintaining two identities for a stated period. Keeping the name would preserve whatever recognition it had in its own territory, at the cost of funding a second identity in a market where Nordec has none. The book does not say that Nordec will acquire anyone; it says only that the decision has three shapes and that the shape should be chosen before the name arrives.

IN LA MACHINE One account, one identity

The application copy read for this book shows a platform built around one identity per account, which is the branded-house shape applied to software (La Machine, 2026). Business → My Website (business/website, with business/website/builder and business/website/preview) carries a Theme & Branding panel with a gallery of 28 themes, a brand name, a tagline, a logo, a favicon and a banner image, five color overrides and a heading font and a body font chosen from a catalog of 200; the configuration is kept in the browser, and it describes one site identity rather than a portfolio of them. The one place where the platform applies a stored identity to more than one brand is Me → Edit Styles (crm/me/styles), available to the platform's OEM partners only, which stores light and dark colors and three fonts for each white-label site on the server and applies them to that partner's clients' interface. That is an architecture in which one platform serves several brands, each with its own colors and fonts, and it is not available to an ordinary account. What the platform does not have is a brand architecture tool: there is no place to record which name leads, which name endorses and which name stands alone.

6.7 Deciding and Documenting the Architecture

The decision procedure for a new offer is short enough to be performed in an afternoon and specific enough to produce the same answer from two different people. It has five steps, and the order matters because each step narrows the next. The first step is to name the customer the offer serves, in the terms the business already uses: an owner-operator of a building of a given size, a property manager running several buildings, a manufacturer, a general contractor. The second step is to check whether the promise is the same. Nordec's promise is that the business knows the building and tells the owner ahead; an offer that keeps that promise belongs under the master brand, and an offer that requires a different promise is a candidate for a name of its own. The third step is to ask what would happen to the master brand if the offer failed. An offer that could damage the promise — a low-cost service with no response commitment, a line of work the business cannot staff consistently — is one to keep at a distance. The fourth step is to choose a descriptive name under the master brand unless the first three steps say otherwise, which they usually will not. The fifth step is to record the decision on the architecture map with its date and its reason, so that the next person to ask the question starts from the answer rather than from the beginning.

The architecture map is one page, and its shortness is the point. It carries four things. The first is the position on the spectrum, stated in one sentence: Nordec is a branded house, with one master brand and descriptive names for its tiers and its entry offer. The second is the list of names the business uses, with the master brand marked and each offer name placed under it. The third is the rule for new names, stated as a rule rather than as an example: a new offer carries a descriptive name under the master brand unless it serves a different customer with a different promise, and any exception is decided by the owner. The fourth is the list of marks the business displays that it does not own, with the owner of each and the rules under which it may be used. One page is enough for all four, and one page is what a person will actually read.

The map is not a strategy document and it does not replace one. Book A3 (Strategy and Competitive Positioning) holds the positioning, Book A6 (Designing a Value-Added Offer) holds the offer, and this chapter holds the names and their relationships. What the map does is make the architecture visible at the moment a decision is being made, which is the only moment at which it can be applied. A business that keeps its architecture in a strategy document that is read once a year will make its naming decisions from the strategy document's absence, which is how a portfolio acquires a name that nobody can explain.

The naming rules that sit under the map are three or four sentences long and they cover the cases that actually arrive. The first rule says what a new offer is called: a descriptive name in French first, with an English version, carrying the master brand where the offer needs to be identified as the business's. The second rule says who decides: the owner, on the recommendation of whoever runs the offer, with the decision recorded on the map. The third rule says what happens when a name arrives from outside, through an acquisition or a partnership: the three options of section 6.6 are considered in order, and the choice is recorded with the date on which any transition ends. The fourth rule says what happens when a name is retired: the name stops appearing on new materials from a stated date, and the materials that carry it are replaced in the order set out in Chapter 8.

The last piece of the documentation is the trademark file, and it belongs with the map because the two answer different halves of the same question. The map says which names the business uses and how they relate; the file says which of them are registered, in which classes, with which dates, and which marks belonging to other people the business is permitted to display. In the case followed through this book, the file holds the application for the word mark NORDEC in Nice class 37 for building maintenance and repair services, filed through a trademark agent, and the record of the two registered marks found close to "Nordec" in unrelated classes of goods during the knockout search. The file is read at the same time as the map, and both are kept by the administrator.

METHOD Deciding where a new offer sits

First, name the customer the offer serves, using the terms the business already uses for its customers, and be specific about the kind of building, the kind of buyer and the territory. Second, check whether the promise is the same: if the offer keeps the promise the master brand already makes, it belongs under the master brand, and if it requires a different promise, it is a candidate for a name of its own. Third, check whether the offer could hurt the master brand if it failed, because an offer the business cannot staff consistently, or one that trades the promise for a lower price, is one to keep at a distance from the name. Fourth, choose a descriptive name under the master brand unless the first three answers say otherwise, in French first with an English version, and check the name against the trademark databases before it is printed anywhere. Fifth, record the decision on the architecture map with its date and its reason, so that the next person to ask the question starts from the answer rather than from the beginning.

WHO DOES THIS The owner decides the architecture

In a small business, the architecture decision belongs to the owner, and it belongs there for a reason that is not about authority. The decision trades one kind of cost against another — the cost of building a second name against the risk of stretching the first one — and the person who holds the budget and the promise is the only person who can weigh those two things against each other. In the case followed through this book, Sophie Nadeau decides the architecture, and the administrator keeps the map, the naming rules and the trademark file. The administrator's role is not to decide but to make the decision visible: they hold the templates and the asset folder, they check every new material against the map, and they raise the question when an offer appears that the map does not cover. The design studio that produced the visual identity works inside the architecture rather than on it, and the trademark agent advises on registrability and on the risk of confusion. Nobody else in a business of seven people needs to hold the architecture, and everybody needs to know where it is written down.

TOOLS OF THE TRADE What an architecture decision needs

Four things carry an architecture decision, and none of them is a product. The first is a one-page architecture map: the position on the spectrum in one sentence, the list of names with the master brand marked, the rule for new names, and the marks the business displays that it does not own. The second is a list of offers and their customers, kept in the same place as the offer design, so that the first step of the decision procedure can be performed without a meeting. The third is a set of naming rules, three or four sentences long, covering what a new offer is called, who decides, what happens when a name arrives from outside and what happens when a name is retired. The fourth is a trademark file: the applications and registrations the business holds, with their classes and dates, and the permissions under which it displays marks belonging to other people. A document, a list, a rule and a file are enough, and a business that keeps all four can answer the architecture question in an afternoon.

6.8 The Case's Architecture

Nordec Entretien inc. is a branded house, and the decision was made deliberately rather than inherited. One master brand, the word Nordec, covers the business, its three tiers and its entry offer. The tiers carry descriptive names underneath it — Essentiel, Préventif and Portefeuille in French, Essential, Preventive and Portfolio in English — and the entry offer carries a descriptive name with the master brand inside it, « Bilan Nordec » in French and "Nordec Building Check" in English. The guarantee is described rather than named, so that it cannot drift from what is promised. The legal name stays Nordec Entretien inc., and the name Nordec is kept, which means the architecture decision did not require a renaming exercise and did not disturb the recognition the business had already earned in its two territories.

Two alternatives were considered and rejected, and the reasons are worth stating because they are the reasons a small business usually faces. The first was a separate brand for the entry offer, so that the assessment could stand apart from the maintenance contracts and be sold to buildings that had signed nothing. It was rejected because a separate name would have had to earn its own recognition among the roughly 1,400 buildings in the two territories, with no budget to earn it, while the master brand was already opening doors: three of the fifteen proposals the business decided in the six months before the refresh began with a Building Check letter signed by Nordec. The second was a separate brand for the Portfolio tier, on the argument that property managers are a different kind of customer from owner-operators. It was rejected for the same reason in a different form: the promise is identical for both, the work is identical, and the only thing that differs is the number of buildings the customer runs, which is a difference the tier name already carries.

The third element of the architecture is the manufacturer's certified-installer mark, which Nordec is permitted to display under the manufacturer's own usage rules. The mark is allowed on the website only, and it is not placed in the templates, on the vans or in the printed documents, because those materials are governed by Nordec's own identity rather than by the manufacturer's rules. The mark belongs to the manufacturer, and Nordec displays it under permission rather than owning it. The architecture map records the mark, its owner and the rules under which it may be used, so that a person preparing a new document can see at a glance that the mark does not belong there.

The architecture also has a rule for the future, and the rule is the part that makes the decision useful rather than merely descriptive. A new offer carries a descriptive name under the master brand unless it serves a different customer with a different promise, and any exception is decided by Sophie Nadeau and recorded on the map with its date and its reason. If Nordec ever acquired a small firm of trades in a third territory, the three options of section 6.6 would be considered in order: absorb the name, endorse it for a transition as "a Nordec company", or keep it. The book does not say that Nordec will acquire anyone. It says that the decision has three shapes, that the shape should be chosen before the name arrives, and that the choice belongs on the map.

The contrast with the regional integrator is the clearest way to see what the architecture is doing. The integrator runs several service lines under separate names, which suits a large firm that buys and sells businesses and needs names it can detach and trade. Nordec has one reputation to build and no portfolio to trade, so the same shape would divide a small budget between names and leave none of them strong. The integrator's trucks, uniforms and reports look the same everywhere; Nordec's did not, before the refresh, and the architecture is what makes them look the same now. One name, one promise, several offers underneath it, and a rule for what happens next.

DO THE MATH What a separate brand would have cost Nordec

Illustrative figures; the case is fictional. Step 1 — Of the fifteen proposals Nordec decided in the six months before the refresh, three began with a Building Check letter signed by Nordec, which is 3 ÷ 15 = 20% of proposals that the master brand already opened. Step 2 — A separate brand for the assessment would have started with no recognition at all among the roughly 1,400 buildings in the two territories, and the case records no budget for building recognition in a new name. Step 3 — A second identity would have required a second design brief, a second set of templates and a second trademark application, each of which carries its own cost and its own filing. Step 4 — None of that is priced in the case, because the option was rejected before a quote was asked; the figures that exist are the ones the business already had. Step 5 — The decision therefore rests on the 20% of proposals that the Nordec name already opened, weighed against the cost of a name that would have had to earn the same recognition from nothing.

CASE IN POINT Nordec Entretien inc.: why the Building Check stayed a Nordec product

Nordec Entretien inc. is a fictional business created for this encyclopedia; its owner, its employees, its customers, its competitors, its figures and the events described are invented, and any resemblance to a real business is coincidental. The assessment was the offer most likely to leave the master brand, because it is sold to buildings that have signed nothing and it is priced as a fixed assessment rather than as a contract. The argument for a separate name was that the assessment could then be marketed as a product in its own right, to a wider set of buildings, without dragging the maintenance business into a sales conversation. The argument against it was arithmetic. A separate name would have had to earn its own recognition among the roughly 1,400 buildings in the two territories, with no budget to earn it, and the master brand was already doing that work: three of the fifteen proposals decided in the six months before the refresh began with a Building Check letter signed by Nordec. The assessment stayed a Nordec product, with a descriptive name that carries the master brand inside it, and the eleven assessments sold in those six months, four of which converted into contracts, were sold under a name the customer already knew.

Costly Mistakes

• Giving every new offer a name of its own. A business that grants a fresh name to each new service line ends up with several weak names and one marketing budget that was never increased, and none of the names becomes strong enough to be remembered without help.
• Copying the portfolio shape of a large company. A house of brands suits a firm that buys and sells businesses and can fund each name separately, and a small business that copies the shape without the budget pays for several identities and builds none of them.
• Naming an offer suggestively when the business sells several benefits. A name that suggests one benefit helps customers remember claims about that benefit and makes it harder for them to remember the others, which is a poor fit for a service business with more than one thing to say.
• Leaving the French layer of a name to the end. In Québec the descriptive layer of a business name and of a trademark on a product is governed by rules that apply before the name is designed, and a name that has to be rebuilt after it is printed costs more than one that was built in the right order.
• Displaying another company's mark without its rules. A certification or partner mark belongs to its owner, is displayed under that owner's usage rules, at the size and in the position those rules specify, and for as long as the permission lasts.
• Endorsing an acquired name without an end date. An endorsement is a bridge, and a bridge with no end date becomes a permanent second identity that the business maintains and pays for indefinitely.
• Keeping the architecture in a strategy document nobody opens. The architecture has to be visible at the moment a naming decision is being made, which means one page, in the place where the templates and the asset folder are kept.

Key Takeaways

• Brand architecture is the organizing structure of a brand portfolio that specifies brand roles and the relationships between brands, and it is a decision a business makes either deliberately or by default (Aaker & Joachimsthaler, 2000).
• The brand relationship spectrum runs from a branded house, through sub-brands and endorsed brands, to a house of brands, and a coherent architecture brings impact, clarity, synergy and leverage instead of weakness, confusion, waste and missed opportunities (Aaker & Joachimsthaler, 2000).
• A small business usually belongs at the branded-house end, because it has one budget for making a name known and splitting that budget between names leaves none of them strong.
• Descriptive names under the master brand keep the room open for every benefit the business sells, while a suggestive name improves recall of the claims consistent with the benefit it suggests and lowers recall of the others (Keller et al., 1998).
• In Québec the descriptive layer of a name is French first: a business name's generic term must be French and is mandatory when the specific element is in another language, and since June 1, 2025 the generic or descriptive words inside a trademark on a product must appear in French (Registraire des entreprises, 2025; Office québécois de la langue française, 2025b).
• A name that arrives through an acquisition has three shapes — absorb it, endorse it for a transition with an end date, or keep it — and the shape should be chosen before the name arrives rather than during the acquisition.

Your Action Plan

1. Write the position on the spectrum in one sentence. State whether the business is a branded house, a sub-brand architecture, an endorsed architecture or a house of brands, and name the offers that sit under it. Deliverable: one sentence of position, with the list of names under it. Verification: every name the business currently uses appears in the list, and the master brand is marked.
2. Draw the one-page architecture map. Put the position, the list of names, the rule for new names and the marks the business displays that it does not own on a single page, and keep it where the templates and the asset folder are kept. Deliverable: a one-page map. Verification: a person preparing a new document can tell from the map whether a new name is allowed and which mark may appear.
3. Write the naming rules in three or four sentences. Cover what a new offer is called, who decides, what happens when a name arrives from outside and what happens when a name is retired. Deliverable: a written naming rule. Verification: the rule answers the last three naming questions the business actually faced.
4. Apply the five-step procedure to the next offer that arrives. Name the customer, check the promise, check what a failure would do to the master brand, choose a descriptive name under the master brand unless the answers say otherwise, and record the decision with its date. Deliverable: one dated architecture decision. Verification: the decision names the customer, the promise and the reason, and it is on the map.
5. Open the trademark file and record every mark the business does not own. List the applications and registrations the business holds with their classes and dates, and every partner, manufacturer or customer mark it displays with the owner, the permission and the usage rules. Deliverable: one trademark file. Verification: every mark that appears on a van, a template or the website is either owned by the business or recorded with its owner and its permission.

In La Machine

Two routes in the application copy read for this book bear on architecture, and neither of them is an architecture tool (La Machine, 2026). Business → My Website (business/website, with business/website/builder and business/website/preview) carries a Theme & Branding panel with a gallery of 28 themes, a brand name, a tagline, a logo, a favicon and a banner image, five color overrides and a heading font and a body font chosen from a catalog of 200, and the configuration is kept in the browser. That panel describes one site identity, which is the branded-house shape: one name, one set of colors, one pair of fonts, applied to one site. The one place where the platform applies a stored identity to more than one brand is Me → Edit Styles (crm/me/styles), available to the platform's OEM partners only, which stores light and dark colors and three fonts for each white-label site on the server and applies them to that partner's clients' interface. That is an architecture in which one platform serves several brands, each with its own colors and fonts, and it is not available to an ordinary account.

What the platform does not do belongs in the same breath. La Machine has no brand architecture tool: there is no place to record which name leads, which name endorses and which name stands alone, and no place to keep the naming rules or the architecture map. It has no saved brand kit that other pages use, no brand portal or guidelines publisher, and no digital asset management with versions, usage rights or expiry dates, so the file of marks the business displays under permission lives outside it. The platform's own modules cover several families a small business otherwise buys separately, and Book N1 maps them; before counting a family as covered, check that the page does what the business needs. The architecture map this chapter describes is a document the business keeps, and the platform is where some of the names it governs are applied.

Self-Check

1. What does brand architecture decide, and why is it a different question from positioning and from the design of the offer?
2. What are the four positions on the brand relationship spectrum, and what does each one suit?
3. Why does a small business usually belong at the branded-house end of the spectrum, and what does it give up by staying there?
4. What does the research on suggestive names say, and what does that mean for a service business that sells several benefits?
5. What are the three options for a name that arrives through an acquisition, and what does each one cost?

Answers. 1. Brand architecture is the organizing structure of a brand portfolio that specifies brand roles and the relationships between brands (Aaker & Joachimsthaler, 2000); it is different from positioning, which Book A3 (Strategy and Competitive Positioning) owns, and from the design of the offer, which Book A6 (Designing a Value-Added Offer) owns, because it is a decision about names and reputations rather than about where the business competes or what it sells. 2. The four positions are the branded house, in which one master brand covers everything with descriptive names underneath it; sub-brands, in which a master brand carries offer names attached to it; endorsed brands, in which an offer name leads with the master brand behind it as a guarantee; and the house of brands, in which separate names lead with no visible parent; the branded house suits offers that serve the same customers with the same promise and a small budget for making a name known, and the house of brands suits genuinely different customers with genuinely different promises, funded separately. 3. A small business usually belongs at the branded-house end because it has one budget for making a name known, and splitting that budget between names leaves none of them strong; what it gives up is the ability to let one offer fail without the failure landing on the master brand, and the ability to sell or retire one name without disturbing the rest. 4. A name that suggests a benefit improves recall of advertised claims consistent with that benefit but lowers recall of claims about other benefits (Keller et al., 1998), which means a service business with several benefits to sell is better served by a descriptive name that leaves the room open for all of them. 5. The three options are to absorb the name, which is cheapest to run and fastest to complete but is paid for at the moment of the change; to endorse it for a transition, which keeps the acquired name visible while the connection is established at the cost of maintaining two identities for a stated period; and to keep it, which preserves whatever recognition it has in its own market at the cost of funding a separate identity.

Chapter References

Aaker, D. A. (1996). Building strong brands. Free Press.

Aaker, D. A., & Joachimsthaler, E. (2000). The brand relationship spectrum: The key to the brand architecture challenge. California Management Review, 42(4), 8–23. https://doi.org/10.2307/41166051

Keller, K. L., Heckler, S. E., & Houston, M. J. (1998). The effects of brand name suggestiveness on advertising recall. Journal of Marketing, 62(1), 48–57. https://doi.org/10.1177/002224299806200105

La Machine. (2026). Documentation — route reference inventory. Internal product documentation, read September 25, 2026, together with the code of the pages this book describes: business/website, business/website/builder, business/website/preview and crm/me/styles.

Office québécois de la langue française. (2025b). Marque de commerce sur les produits (updated August 27, 2025). Retrieved September 25, 2026, from https://www.oqlf.gouv.qc.ca/francisation/entreprises/marque-commerce-produits.html

Registraire des entreprises. (2025). Les noms d'entreprises au Québec. Gouvernement du Québec. Retrieved September 25, 2026, from https://cdn-contenu.quebec.ca/cdn-contenu/adm/min/emploi-solidarite-sociale/registraire_entreprises/IN-531_REQ_noms_ENTR_QC.pdf

Trademarks Act, R.S.C. 1985, c. T-13, ss. 30 and 46 (Canada). Justice Laws Website, current to September 3, 2026 (last amended April 1, 2025) — read September 25, 2026.

Chapter 7 — Brand Experience Across Touchpoints

Customers do not meet a brand in its guidelines but at its touchpoints — a technician at the door, a report in an inbox, a van in the street, a page on a screen — and the brand is whatever those encounters add up to.

What You'll Be Able to Do

• List every place a customer, a prospect or a passer-by meets the business, in four groups — people, physical, documents and digital — so that the brand is mapped rather than assumed.
• Estimate how often each touchpoint is seen in a year, and rank the touchpoints by frequency and by how much they matter at the moment of a decision.
• Roll out an identity in the order that produces the most exposure for the least money, starting with the documents and the digital touchpoints a customer reads most often.
• Brief the people who carry the brand — technicians, the dispatcher, the administrator — so that the promise is said and kept on the call and at the building.
• Design the physical touchpoints with the rules that apply to a sign visible from outside in Québec, and know which parts of that work a lawyer or the language authority must confirm.
• Audit consistency against six criteria, score it on a sample, and hold the score as a number that a small business can move.

Why It Matters

A brand is built in the gap between what a business promises and what a customer encounters, and that gap is closed or opened at touchpoints. The promise itself is made in the brand platform, designed in the visual identity and written down in the guide, and none of those documents is ever read by a customer. What a customer reads is a monthly report that arrived two days late, a technician who could not explain why the part was ordered, an invoice with an amount nobody mentioned, a van that drove past while they were waiting. Each of those is a touchpoint, and each of them either confirms the promise or quietly withdraws it. The chapter before this one built the system that holds the identity together; this chapter is where that system meets a real person.

The stakes are larger than they look because a small business has so few touchpoints compared with a large one, and each of them therefore carries a larger share of the brand. A national firm can afford for one technician to have a bad day, because the customer has a hundred other impressions to weigh against it. A business of seven people cannot, because the customer may never meet another. In the business followed through this book, the same seven people appear at the door of fifty-two buildings, and the same two vans pass through two territories every working day. There is no advertising budget to correct an impression and no second team to send. The touchpoints are the whole of the brand, which makes mapping them the most practical branding work a small business can do.

The measurement in the case shows how far a business can travel without any of this being decided. An inventory of the touchpoints found thirty-six in all, and only thirteen of them — 36.1% — were consistent with one another. Three versions of the logo were in circulation, five different blues, four typefaces, and the two new tiers and the entry offer had no names in French, no templates and no identity at all. None of that was a failure of effort. It was the ordinary result of a business growing for four years without a place where the touchpoints were listed together, and it is the condition this chapter exists to correct.

In This Chapter

Section 7.1 sets out the brand as the sum of experiences at every touchpoint and explains why the behavior of employees weighs more on brand equity than any change in aesthetics. Section 7.2 maps Nordec's thirty-six touchpoints into four groups, estimates their yearly exposures and ranks them. Section 7.3 looks at the people who carry the brand — the technicians, the dispatcher, the administrator and Sophie — and at what a brief for each of them contains. Section 7.4 covers the vehicles, the uniforms, the premises and the printed stock. Section 7.5 covers the digital touchpoints, from the website to the business profile and the email signature. Section 7.6 treats the documents as brand touchpoints and shows why a monthly report is read more often than any advertisement. Section 7.7 sets out the consistency audit and its six criteria. Section 7.8 follows the rollout as the case performed it, with its order, its budget and its first customer comments.

7.1 The Brand Is What People Experience

A brand is often described as if it lived in a document, and it does not. The document states an intention, and the intention is realized only when somebody encounters the business. The encounter may last ninety seconds, as it does when a technician signs in at a reception desk, or it may last several years, as it does when a property manager reads twelve monthly reports and a capital planning note for the same building. What the customer holds afterwards is not the intention but the sum of the encounters, and that sum is what the word brand means in practice. Two businesses may hold identical promises in their guides and deliver entirely different brands, because one keeps the promise at the door and the other does not.

This is why the brand experience crosses functions rather than sitting inside marketing. The promise is made by whoever runs the business, the visual identity is designed by a studio, the guide is written by whoever holds the system, and then the promise is delivered by operations, by service, by the people who answer the telephone and by the people who send the invoice. In a business of seven people, all of those functions sit with the same small group, and the boundaries between them are the place where a touchpoint usually fails. The monthly report is not late because the report is badly designed; it is late because the person who produces it was on the road, and nobody had agreed which of two jobs came first. A brand experience is therefore as much a question of operations as of design, and a business that fixes only the design has fixed the part that was not broken.

The clearest statement of the relative weight of the parts comes from the research on corporate rebranding. In an analysis of 166 rebranded companies and two detailed cases, the authors found that structural change — above all mergers and acquisitions — is the most common trigger of a rebrand, and that a change in marketing aesthetics affects brand equity less than factors such as employees' behavior; rebranding, they conclude, has to be managed as a whole, with the support of the stakeholders (Muzellec & Lambkin, 2006). Read against a small business, the finding is plain. New colors and a new logo are a smaller influence on what customers think than how the people of the business behave on an ordinary Tuesday. This is not an argument against the visual identity; it is an argument about where the visual identity ranks in the work, which is behind the behavior it is meant to support.

The framework that describes where the encounters lead is the customer-based brand equity model. Brand building runs through four steps — identity, meaning, responses and relationships — and through six building blocks, of which salience sits at the base and resonance at the top; resonance itself has four kinds, behavioral loyalty, attitudinal attachment, a sense of community and active engagement, and the model is offered as a yardstick for brand tracking rather than as a validated instrument (Keller, 2001). A business of any size can use that shape. The name and the symbol create salience, the touchpoints give the brand its meaning and provoke responses, and the accumulation of touchpoints over years is what produces resonance, if anything does. There is no shortcut to the top of that model, and there is no way to reach it while the touchpoints contradict one another.

The case shows what an unmapped brand looks like from the inside. Sophie Nadeau founded Nordec Entretien inc. in 2022 and worked fourteen years as a building systems technician before that. The first logo was drawn by a friend on a template site, a blue wrench inside a circle with "NORDEC" in capitals, and it served the business well enough for four years. The brand platform of Chapter 2 gives Nordec a promise in French and in English — « Votre bâtiment, un an d'avance. », and "Your building, a year ahead." — and four personality traits: calm, precise, plain-spoken, neighborly. The visual identity of Chapter 4 gives it a letter N with a horizon drawn through it, a palette led by Nordec Blue #1F3A5F and Signal Amber #E8A33D, and two open-license typefaces. The design system of Chapter 5 gives it twenty-one design tokens, eleven templates and a guide. All of that work is what this chapter now has to place, because a system that lives in a folder has not yet met a customer.

DEFINITION Touchpoint, brand experience, exposure and consistency

A touchpoint is any place where a person meets the business: a technician at a door, a report in an inbox, a van in traffic, a page on a screen, a voice on the telephone. The brand experience is the sum of those meetings over time, as the customer holds it, and it is the brand as the customer knows it rather than as the business intends it. An exposure is one occasion on which a touchpoint is seen, heard or read, and exposures are counted a year so that touchpoints can be compared with one another rather than argued about. Consistency is the degree to which the touchpoints agree — the same logo, the same colors, the same typefaces, the same voice, the same promise kept — and it is measured, in this chapter's method, as the share of touchpoints that pass a stated set of criteria. None of these four words requires a large business. They require a list, a count and a decision about which touchpoints get the money first.

7.2 Mapping Touchpoints

Mapping begins with a list, and the list is longer than any business expects. The method that works is to walk through a year of the business from the customer's side and write down every point at which the customer sees, hears or reads something. It helps to start from a real customer rather than from a diagram: a property manager who has one building under contract and one not, who reads a monthly report, calls the dispatcher twice, signs a proposal, receives an invoice and passes a van on the highway, has already produced seven touchpoints on their own. The same exercise repeated for four or five customer types produces a list that is accurate enough to work from, and the list is then grouped, because thirty-six separate lines cannot be ranked and four groups can.

The four groups are people, physical, documents and digital, and they are chosen because each group is managed by a different part of a small business and fixed in a different way. People are managed by briefing and by hiring, physical things by a purchase, documents by a template, digital by a page. Grouping the touchpoints therefore does something a flat list cannot: it makes visible which part of the business has to act. Nordec's thirty-six touchpoints divide as four people, seven kinds of physical touchpoint, eleven documents and fourteen digital ones, and the division is not even. The documents are eleven lines because a service business writes to its customers constantly, and the digital group is fourteen because a modern small business accumulates pages, profiles and signatures without ever deciding to.

Counting exposures is the second step, and it is what turns a list into a ranking. An exposure is one occasion on which the touchpoint is encountered, and for a business with recurring visits the count is straightforward arithmetic. Nordec performs 624 preventive visits a year, which is fifty-two contract buildings multiplied by twelve months, and produces 576 monthly reports a year, which is forty-eight buildings on the Preventive and Portfolio tiers multiplied by twelve. It handles about 276 urgent calls a year. Two vans are on the road every working day and seven people wear the uniform. Those numbers do most of the thinking, because they show immediately that a visit sheet and a monthly report are seen well over a thousand times a year between them, while a depot sign is seen by whoever passes the depot. A business that spends its first money on the sign and its last on the templates has spent it in the reverse of the order the exposures recommend.

The third step is the rating of importance, which is not the same thing as frequency and is not a substitute for it. Some touchpoints are seen rarely and decide everything, because they occur at a moment of truth: the proposal that precedes a signature, the invoice that arrives after a promise about cost, the telephone call that comes after a failure. A touchpoint of that kind carries a weight out of proportion to its count, and the ranking has to allow for it. The method taught in this chapter multiplies, in effect, the exposures by the importance: the touchpoints that are both frequent and decisive come first, the frequent but ordinary ones come next, the rare and decisive ones are given special care, and the rare and undecided ones wait. For Nordec, the documents and the people come first, the digital group second, and the physical group third, even though the vans are the most visible thing the business owns.

Figure 7.1 draws the map that results, and the map is the deliverable that section 7.8 rolls out from. It shows the four groups, the touchpoints inside each, and the yearly exposures of the three most frequent ones, which are the 624 preventive visits, the 576 monthly reports and the roughly 276 urgent calls.

Figure 7.1 — Where Nordec's customers meet the brand. The documents and the people are seen more often than the logo, so they carry most of the brand.

KEY CONCEPT The most frequent touchpoints carry the brand

The touchpoint seen 1,200 times a year carries more of the brand than the one seen twice, and the arithmetic is not subtle. A customer who receives twelve monthly reports from Nordec and reads each of them has had twelve encounters with the brand before the business has done anything that could be called advertising, and those twelve encounters either repeat the promise or contradict it. The key concept of this chapter is that the ranking by exposure, adjusted for the moment of truth, is the correct order of investment, and that it is almost never the order a business would choose by instinct. The instinct is to buy the visible thing first — the sign, the wrap, the new website — because those are the touchpoints the business itself sees. The exposures point the other way, to the documents and the people, and those are cheaper, which makes the correction painless. Money spent on the frequent touchpoints is not money taken from the visible ones; it is money that makes the visible ones worth looking at.

Table 7.1 sets out the four groups, with examples from Nordec, an estimate of the exposures a year, the number of touchpoints that were consistent before the refresh, and the priority the ranking produced. It is read as a plan of work rather than as an inventory: the priority column is the order in which the identity is applied, and the number of consistent touchpoints is the score the audit will move.

Table 7.1 — Nordec's touchpoints by group

METHOD Ranking touchpoints for the rollout

First, list the touchpoints in the four groups — people, physical, documents and digital — by walking through a year from a real customer's side, and keep the list long rather than tidy. Second, estimate the exposures a year for each touchpoint, using the business's own volume: visits multiplied by buildings, reports multiplied by buildings, vans multiplied by working days, people multiplied by garments. Third, rate the importance of each touchpoint at a moment of truth — the proposal, the invoice, the call after a failure, the first visit — and mark the ones that decide a relationship. Fourth, rank by exposures and importance together, placing the frequent and decisive touchpoints first, the frequent and ordinary ones next, and the rare and decisive ones under special care. Fifth, roll out the top of the list first, within the budget, and leave the rare and undecided touchpoints for the next year. The ranking is a decision, not a discovery; its value lies in making the order explicit.

7.3 People as the Brand

In a service business, most of the brand is delivered by people, and the touchpoints that carry it are the ones a business rarely writes down. Nordec has four kinds of person-touchpoint: the technicians at the building, the dispatcher on the telephone, the administrator by email, and Sophie at renewal meetings. Between them they account for the majority of the customer's impressions of the business, and before the refresh none of the four was consistent with the others, which is a polite way of saying that customers could not have described the business in the same words after meeting two of them.

The technician at the building is the touchpoint with the highest weight, because the technician is the reason the customer is a customer. The brand platform gives Nordec three values, each attached to a decision it changes, and two of the three are delivered at the door. Punctual means the ninety-minute commitment comes before a more profitable job, so the technician who is dispatched to an urgent call is the living form of the first value. Thorough means the building health record is updated on site, before leaving, and the technician who opens the record on the doorstep and writes in it before driving away is the living form of the third. Straight means the two-hour call before exceeding the included work and no invoice the owner has not heard about, and it is delivered partly by the technician and partly by whoever prepares the invoice. A brief for a technician therefore does not consist of values as words; it consists of the three decisions those values change, stated as things to do and to refuse.

The dispatcher carries the promise in the first ninety seconds of a call, and the promise is unusually literal for Nordec. "We know your building, and we tell you ahead" is a sentence a dispatcher either delivers or does not, and it is delivered by what the dispatcher knows and when the customer hears it. A dispatcher who can say what happened at the last visit, what part is on order and when the technician will arrive has kept the promise three times before the technician has left the depot. A dispatcher who has to ask which building is calling has withdrawn it, once, in the ear of the person who decides whether the contract continues. The brief for the dispatcher is consequently about access to information as much as about tone: the building health record has to be openable during a call, and the ninety-minute clock has to be visible, because a promise nobody can check is a promise nobody believes.

The administrator carries the brand in every email sent on behalf of the business, and the volume is large enough to matter. The monthly reports, the capital planning notes, the proposals and the invoices all leave through the administrator's hands or through templates the administrator maintains, which makes that role the single largest distributor of the brand in the business. It is also why the administrator becomes the brand keeper in the system designed in Chapter 5, holding the templates and the asset folder and checking every new material. The brief for the administrator is not about writing style alone. It is about the four principles of the voice — say the time, say the cost, say what comes next, no jargon without its plain word — and about the glossary of twelve words the business uses and avoids, because a service vocabulary leaks into customer emails unless somebody is responsible for keeping it out.

Sophie's own touchpoint is the renewal meeting and the relationship a property manager has with the owner of the business. In a small firm, the founder is a touchpoint in a stronger sense than any employee, because customers infer the character of the business from the character of the person who owns it, and they do so whether or not the owner intends it. The brief for that touchpoint is not a script; it is the same promise, delivered in a room where the customer is deciding whether to renew, and it is supported by the one thing that makes a founder credible in that setting, which is the record of what was promised and what was done.

Uniforms and identification badges belong to this group as much as to the physical one, because they are the visible sign that the person at the door is a person the business has sent. In the case, the uniforms are seven people multiplied by five garments, and the badges carry the technician's name and photograph with the Nordec mark. The two together do a small amount of work that no document can do: they remove the first ten seconds of doubt at the door, which is the ten seconds in which a tenant or a building manager decides whether to let somebody into a mechanical room. Chapter 10 sets out how the briefs are used in onboarding, so that a new technician receives them before the first building rather than after it.

WHO DOES THIS Everyone who meets a customer

In a small business there is no brand department, and the work of carrying the brand falls to whoever meets the customer. The technicians carry it at the building, on roughly 624 visits and 276 urgent calls a year. The dispatcher carries it on the telephone, at the moment a customer is most anxious. The administrator carries it in every document that leaves the business, which is the largest volume of brand impressions in the firm. Sophie carries it in the renewal meeting and in every relationship a property manager has with the business as a whole. Nobody in that list has a marketing title, and each of them is doing brand work on every occasion they meet somebody. What the brand system can do for them is narrow and practical: give them the words, give them the templates, give them access to the building record, and tell them which decisions the values change. What it cannot do is be delivered by somebody else.

7.4 Vehicles, Uniforms, Premises and Print

The physical touchpoints are the ones a small business thinks of first and should fund third, and the reason is not that they are unimportant but that they are visible mainly to the business itself. A van is seen by thousands of people who will never buy maintenance, and by a smaller number who might; the technicians and the customers see it every day. The physical group is still worth doing well, because it is the group that carries the name and the symbol into places the business cannot go, and because it is the only group whose state is obvious to the people who work there. A business with three versions of its logo on its vans is a business whose own staff can see that nobody is minding the brand.

The vehicle wrap is the largest single item in the physical group, and its design is governed by distance rather than by detail. A wrapped van is read at two ranges: from a few metres, when it is parked at a building, and from tens of metres, when it passes on the highway. Only two pieces of information survive the second range, and they are the name and the telephone number; everything else is decoration. The symbol can survive if it is large and simple, which is one reason the horizon line drawn through the letter N of the chosen route works at a distance: it is a single gesture that reads at a hundred metres and rewards a second look at five. The case places the mark on the roof as well as on the sides, because a van on a city street is seen from above by the occupants of higher buildings more often than a designer expects.

Distinctive brand assets are the concept that explains why this repetition is worth the money. Logos, colors, fonts, characters, taglines and jingles form a researched field, and the work examines which branding elements succeed over the long term and which fail (Romaniuk, 2018). The lesson this book draws for a small business, which is not a finding of that work, is that an asset can only become distinctive if it is used the same way every time. A small business cannot buy recognition through repetition at the scale of a national brand, but it can do the same thing in miniature by using the same few assets everywhere: the letter N, the blue, the amber stripe and the tagline. The failure mode is the opposite and is common: a business changes the look of each new material, so that no single asset ever becomes familiar.

In Québec, the physical group carries a legal requirement that has to be designed into the vehicle and into the sign rather than added afterwards. Rules that came into force on June 1, 2025, from the amended Regulation respecting the language of commerce and business, require that on a sign visible from outside a premises, a recognized trademark wholly or partly in another language, and a business name with a non-French specific element, be accompanied by French terms — a generic term, a description or a slogan — so that French is markedly predominant, which the Office québécois de la langue française defines as a much greater visual impact, at least twice the space, permanent, and at least as visible and legible; if a French version of the trademark is registered, it must be displayed (Office québécois de la langue française, 2025a). Nordec designs its van and its depot sign with the French generic term « Entretien » and the French tagline « Votre bâtiment, un an d'avance. » in the larger space, and has the layout checked before anything is produced. The book explains the rules; it does not state that any particular layout satisfies them, and a business should have its signs reviewed by the language authority or by a lawyer.

The uniforms are a smaller line than the wraps and are read at a range the wraps never reach. A uniform is seen by the customer at the door, which is the closest a stranger comes to the business, and it carries three things: that the person is expected, that the person belongs to a business with rules, and what that business is called. The choice of garment matters less than the state of it, and the case provides five garments per person across seven people so that a clean set is always available. The depot sign is the last of the physical touchpoints to be produced, for a reason set out in section 7.8, and the printed stock — business cards, door hangers, the identification badges and the parts labels left on equipment — is replaced as it is used rather than thrown away, which keeps the cost in one year rather than in one month.

QUÉBEC & CANADA NOTE French on the van and the depot sign

On signs visible from outside a premises in Québec, the rules in force since June 1, 2025 require that a recognized trademark wholly or partly in another language, and a business name whose specific element is not French, be accompanied by French terms so that French is markedly predominant — at least twice the space, permanent, at least as visible and legible; where a French version of the trademark is registered, it must be displayed (Office québécois de la langue française, 2025a). The Charter of the French language requires French in commerce and business in Québec, including commercial documents, signs and product inscriptions, and this book states its rules only through the explanations of the Office québécois de la langue française. Two practical consequences follow for a van wrap and a depot sign: the French generic term and the French tagline are part of the design from the first sketch rather than added in review, and the layout is checked by the authority or by a lawyer before production. This book explains the rules and is not legal advice.

WHAT CHANGED Signs and products in Québec since June 1, 2025

Two sets of rules took effect in Québec on June 1, 2025, and they touch different touchpoints. For a sign visible from outside a premises, a recognized trademark in another language and a business name with a non-French specific element must be accompanied by French terms that make French markedly predominant, with a registered French version of the trademark displayed where one exists (Office québécois de la langue française, 2025a). For products, the generic or descriptive words inside a trademark — those describing nature, ingredients, color or scent — must appear in French, while product and company names may stay in another language, and until June 1, 2027 some products made before June 1, 2025 may still be sold (Office québécois de la langue française, 2025b). For a service business the sign rule is the one that binds, because the van and the depot front are signs seen from outside; the product rule matters to a business that packages anything, and it matters to the parts labels Nordec leaves on equipment, which are reviewed with the rest of the printed matter.

7.5 Digital Touchpoints

The digital group is the one that grows without being decided, and Nordec's fourteen touchpoints arrived over four years without anyone ever sitting down to create them. The website came first, then a business profile with its reviews, then email signatures that each person set up differently, then a handful of social pages created for a campaign that ended, then a telephone greeting and an after-hours voicemail recorded by whoever was available, then a set of email templates, then the directory listings that a business is added to without asking. None of these is expensive and none of them is coordinated, and together they are the touchpoints a prospect meets before the first conversation, which makes the group more important than its cost suggests.

The website is the touchpoint a prospect usually meets first and the one least under the control of the branding work, because a website is a technical object as well as a brand object. The identity a visitor meets is carried by the theme, the logo, the palette, the headings and the words, and each of those has to be applied to whatever system the site runs on rather than drawn fresh. Book C2 (Websites, Landing Pages and Conversion) treats the design of the page and its conversion behavior; this chapter treats only the question of whether the site carries the same brand as the report a customer received last week. The test is simple and can be applied in a minute: open the website beside a monthly report and ask whether a stranger would say they came from the same business.

The business profile and its reviews are a digital touchpoint with a property that no other touchpoint has, which is that its content is written by customers. Nordec held twenty-three online reviews with an average of 4.7 before the refresh, and thirty-one with an average of 4.8 six months after it. Reviews are not a brand asset a business controls; they are a brand asset it earns, and they are read by prospects at exactly the moment of decision. What the business controls is how it responds, how promptly it responds, and whether it asks. The case's asking is done at the end of a successful visit, which is why the count moved with the number of visits rather than with any campaign. The profile that displays those reviews is also a touchpoint in its own right, and it carries the logo, the description, the hours and the telephone number that a customer checks before calling.

Email signatures are the least glamorous of the digital touchpoints and one of the most widely seen, because every message that leaves the business carries one. At 576 monthly reports a year plus proposals, invoices and ordinary correspondence, the signature is the touchpoint with the highest exposure in the digital group, and it was also the one with the widest variation before the refresh, since each of the seven people had set up their own. The fix is a template held by the administrator and applied by everyone, and the design question is only what it must contain: the name, the role, the business name with its mark, the telephone number, the website address, and the tagline in the language of the message. Nothing more survives being read. Social pages belong to the same group and are treated in Book N6 (Social Media, Content and the Media Library); for the purposes of the brand they are touchpoints whose profile picture and description must match the rest, whatever the frequency of posting.

Color on screens brings a rule that a small business can check rather than guess. The Web Content Accessibility Guidelines set success criterion 1.4.3 at level AA, requiring text and images of text to have a contrast ratio of at least 4.5:1 against the background, and at least 3:1 for large text, defined as 18 points or 14 points bold; criterion 1.4.6 at level AAA raises those to 7:1 and 4.5:1; and criterion 1.4.11 at level AA requires graphical objects and interface components needed to understand content to have at least 3:1 against adjacent colors (World Wide Web Consortium, 2024). One provision matters particularly to a brand: text that is part of a logo or brand name has no contrast requirement, which is why a mark can be drawn in colors no body text could use. Every other use of color — a heading, a link, a button, a table — is bound by the ratios, and Nordec's palette was checked before the website was touched, with white on Nordec Blue #1F3A5F at 11.48:1, Nordec Blue on Mist #F4F6F8 at 10.60:1, Slate #5B6770 on white at 5.80:1, Nordec Blue on Signal Amber #E8A33D at 5.32:1 and Signal Amber on white at 2.16:1, which is why amber never carries text on white and serves instead for accents, the van stripe and safety details.

THE RULE Contrast on screens

The Web Content Accessibility Guidelines 2.2 state, at success criterion 1.4.3 (level AA), that text and images of text must have a contrast ratio of at least 4.5:1 against the background, and at least 3:1 for large text, which is text of at least 18 points or 14 points bold; criterion 1.4.6 (level AAA) raises those thresholds to 7:1 and 4.5:1; and criterion 1.4.11 (level AA) requires graphical objects and interface components needed to understand content to reach at least 3:1 against adjacent colors (World Wide Web Consortium, 2024). One provision is written for brands: text that is part of a logo or a brand name has no contrast requirement, so a mark may use colors that body text may not. The rule's value to a small business is that it makes a color decision checkable. A palette either produces the ratios on the backgrounds where it is used or it does not, and the check costs a minute per pair rather than a design round.

IN LA MACHINE The digital touchpoints the platform carries

Four pages of the application copy read for this book touch the digital group directly (La Machine, 2026). Business → My Website at business/website, with its builder and preview, carries the site's identity through a gallery of 28 themes, a brand name, a tagline, a logo, a favicon and a banner image, five color overrides and a heading and body font chosen from a catalog of 200, with the configuration kept in the browser. Assets → Visual Studio at business/social/visual-studio is an editor for social, video, print and web formats — a business card, a poster, banners for social profiles — whose designs are saved and whose exports go to the Gallery. Marketing → Templates at business/marketing/templates stores message templates by channel with variables and a preview, though it holds no brand colors, logo or footer. Social → Accounts at business/social/accounts lists connected accounts on nine platforms, while profile pictures and bios are not edited from the platform at all. What none of them does is audit consistency across touchpoints.

7.6 Documents as Brand Touchpoints

The documents are the touchpoints this chapter asks a small business to take most seriously, because they are the most frequent, the least expensive to fix and the most often neglected. A customer of a maintenance contract reads a visit sheet after every visit, a monthly report every month and a capital planning note once a year, and none of those is read as marketing. They are read as work, which is precisely why they carry the brand so effectively: a document that is clear, on time and consistent with what the customer was told is a demonstration of the promise rather than a statement of it. A business that sends an accurate monthly report in the same layout every month has said "we tell you ahead" twelve times without using the words.

The visit sheet is the smallest document and the one produced most often, at 624 a year. It records what was checked, what was found and what will be done next, and it is signed or acknowledged at the building. Its brand work is done by three things: that it exists at all, that it is legible, and that it is the same sheet every time. A visit sheet that is a different scrap of paper each visit tells the customer that the visits are different too, which is the opposite of what a preventive maintenance contract is sold as. The template fixes the layout; the technician fixes the content; the record in the building health record fixes the continuity between one visit and the next, which is the part the customer actually notices.

The one-page monthly report is the touchpoint with the highest combination of frequency and importance, and the case's whole offer rests on it. It is produced 576 times a year, read in an inbox, and it exists to close the gap between what was done and what the owner knows was done. The brand platform's English promise, "we tell you ahead", is delivered in this document by the capital planning note that accompanies it and by the one line that says what comes next. A monthly report that summarizes the past without saying anything about the future is a receipt; the report that carries the promise says what is coming and when, and it is one paragraph longer. The template makes that paragraph mandatory rather than optional, which is a design decision with a brand consequence: the document cannot be produced in a hurry without the forward-looking line.

The capital planning note is the document customers named as one of the strongest features of the offer, and in the twenty value interviews of Book A6, twelve of twenty people named it among the reasons they stayed. It is written once a year for each contract building and lists the equipment approaching the end of its life with a replacement window. Its brand work is unusually direct, because it is the documented form of the promise: the business is telling the owner about a cost before the cost arrives. A note of that kind has to look like something the owner can file and act on, which means a stable layout, a consistent heading, a table of equipment and dates, and a date of issue. The template is consequently one of the eleven in the system, and it is treated as a brand asset rather than as an attachment.

Proposals and invoices are the two documents where the brand meets the customer's money, and they are the touchpoints at which the value Straight is either kept or broken. The rule in the brand platform is that there is no invoice the owner has not heard about, and the two-hour call before exceeding the included work is the mechanism that makes the rule true. A proposal therefore has to state the tiers, the add-ons, the ninety-minute guarantee and its $150 credit, and the first-ninety-days exit, in the same wording the customer will hear on the telephone, because a proposal that uses different words from the contract is the beginning of a dispute. Book E5 (Proposals, Quotes, Business Cases and Contracts) treats the construction of the proposal and its commercial logic; this chapter asks only that the proposal and the invoice look and read like the rest of the business, since a customer who receives a well-set report every month and an invoice that looks like an unrelated firm has noticed something, even if they cannot name it.

The remaining documents are the ones that mark a relationship rather than a transaction: the Bilan Nordec report that follows the fixed-price assessment, the contract, the welcome letter that a new customer receives in the first week, the guarantee credit notice that is issued when the ninety-minute promise is missed, and the letterhead on which ordinary correspondence is written. The guarantee credit notice deserves a sentence of its own, because it is the document a business sends when it has failed. A notice that arrives promptly, in the same layout as everything else, saying what happened and what is being credited, turns a miss into evidence that the promise is real. A notice that never arrives, or that arrives as a deduction nobody explained, turns the same miss into evidence of the opposite. The document is the promise, in the only situation where the promise is at risk.

WATCH OUT A beautiful van and a careless visit sheet

The trap is the order of the work, and it is almost universal. A business commissions a van wrap first, because the van is the visible symbol of having done something, and leaves the visit sheet as a scrap of paper filled in at the wheel. The customer then meets the two touchpoints in the wrong order: a professionally wrapped vehicle at the curb, and a document that looks as if the business had never thought about it. The next month the same customer receives the same sheet, while the van arrives again looking equally good, and the contrast stops being neutral and starts to be read as a statement about what the business cares about. The remedy is not to skip the wrap; it is to reverse the order of production. Templates cost a fraction of a wrap, they are seen far more often, and they change what a customer thinks about the work itself rather than about the business's taste in colors.

7.7 Consistency Audits

An audit is how a business knows whether the touchpoints agree, and it is a smaller exercise than the word suggests. It requires a list of the touchpoints, a set of criteria, a way of scoring each touchpoint against those criteria, and a sample that a person can examine in an hour or two. The output is a single number — the share of sampled touchpoints that pass — and that number is comparable from one audit to the next, which is what makes it useful. In the case, the first audit sampled the thirty-six touchpoints and found thirteen consistent, or 36.1%, and the second, six months after the rollout, found thirty-three of thirty-six. The number is not a measure of quality; it is a measure of agreement, and a business can be consistent and still be wrong about its brand. Its value is that it moves when the work moves, which makes it one of the few brand figures a small business can hold itself to.

The criteria are chosen so that each one can be checked by looking at the touchpoint rather than by interpreting it. Six criteria cover the whole identity: the logo version, the colors, the typefaces, the voice, French first where required, and the promise respected. Each criterion is stated as a pass rule so that two people auditing the same touchpoint would reach the same answer, and the pass rules are deliberately plain: one version of the mark, no more than the named colors in their stated roles, two typeface families, the four principles of the voice, the French generic term and tagline where the law requires them, and the promise kept in what the touchpoint actually does. The sixth criterion is the one that separates a design audit from a brand audit, because it cannot be checked by looking at a color. It is checked by reading what the document says and comparing it with what was promised, and it is the criterion that a careless monthly report fails while passing all five of the others.

The scoring is binary and that is the point. A touchpoint either passes a criterion or it does not, and a touchpoint that fails one criterion fails the audit for that touchpoint, even if the other five are impeccable. Binary scoring removes the temptation to average a business's way to a comfortable number, and it produces an audit result that a person can act on, because the failures name the work rather than describing a general condition. Where a criterion is not applicable — a document in English only, produced for a customer outside Québec, does not fail the French criterion — the criterion is marked as not applicable rather than as passed, so that the denominator stays honest.

The sample is what makes the audit affordable, and it should be chosen rather than taken. Sampling every touchpoint every time is unnecessary, because a business with thirty-six touchpoints is auditing categories rather than objects. The method that works is to audit all touchpoints the first time, since the first audit is also the mapping exercise, and then to sample a fixed set thereafter: the four most frequent documents, two of the physical touchpoints, two of the digital ones, and every new template or material produced since the last audit. New materials are included because they are the place inconsistency enters; a business that has been consistent for a year and adds one template drawn by a supplier without the guide can be inconsistent again within a week. The administrator's role as brand keeper exists for exactly this reason, and the audit is what confirms that the keeper's check is being performed.

The audit ends with a short list of failures, ranked by the exposure of the touchpoint they belong to, and a date for the next audit. The ranking matters because a failure at a touchpoint seen once a year is a smaller problem than a failure at one seen every month, and a business with three failures should fix the frequent one first. Holding the audit to twice a year is enough for a small business: the exposure volumes it deals with change slowly, and an audit performed more often than the work changes produces the same number twice and teaches nobody anything. The next section follows one business through a rollout built on exactly this list.

TOOLS OF THE TRADE What a touchpoint audit needs

Five functions carry a touchpoint audit, and none of them requires a named product. The first is a touchpoint list, held in one editable place, with a column for the group, the exposures a year and the person who owns the touchpoint. The second is a scoring sheet that puts the six criteria against the sampled touchpoints and records a pass, a fail or a not applicable for each, so that the share of passing touchpoints can be computed. The third is a photograph of each touchpoint in use — the wrapped van in a real street, the visit sheet as it comes back from a building, the report as it appears in an inbox — because a touchpoint is audited as it is met, not as it was designed. The fourth is the guide's one-page summary, which is what an auditor holds beside a touchpoint to check a color, a typeface or a line of the voice. The fifth is a monthly report example, kept from the last audit, so that the sixth criterion can be checked against what the customer actually received rather than against the template.

Table 7.2 sets out the six criteria of the consistency audit, what each one checks, the pass rule that makes the check repeatable, and the state of that criterion in Nordec before the refresh. It is read as the scoring sheet of the audit: each row is one criterion applied to every sampled touchpoint, and the last column is the evidence that the audit has something to move.

Table 7.2 — Nordec's consistency audit criteria

7.8 The Case's Touchpoint Rollout

The rollout in the case followed the ranking, not the instinct, and the order is the most transferable thing in this chapter. It ran in three movements. From day one to day sixty, the documents and the digital touchpoints were applied: the eleven templates, the website, the email signatures, the business profile and the telephone greeting. Through months three and four, the uniforms and the two vans were produced, with the vans wrapped one at a time so that a vehicle was always on the road. In month five, the depot sign was made. The printed stock was replaced as it was used rather than discarded, which spread its cost across the year and kept the old and the new from ever appearing together.

Documents first, because they carry the most exposures and cost the least to change. Once the templates exist, the 624 visit sheets and the 576 monthly reports of the following year are produced from them at no extra cost, which is the whole argument of this chapter expressed as an arithmetic. The templates were designed as part of the studio's fixed fee, so the marginal cost of the most frequent touchpoints in the business was zero, while the rarest and most visible touchpoint, the depot sign, was left for month five. The order also had a second effect that was not planned: because the documents changed before the vans, the first customers to notice the refresh saw it in a report they were already reading, which meant the change arrived as a small improvement to something familiar rather than as an announcement of a new look.

Digital second, because the website and the signatures are read by prospects before the first conversation. The website was updated as part of the work described in Book C2, with the palette applied through the site's own theme rather than redrawn, and the five color overrides set to the new values. The signatures were applied in one sitting by the administrator, whose role as brand keeper made the change uniform for the first time in four years. The business profile was updated with the same mark, description and hours, and the telephone greeting and after-hours voicemail were re-recorded with the French first and the same three sentences, because the greeting is the first thing a new customer hears and it had previously been recorded by whoever was free that day.

The vans came third and were wrapped one at a time, which is a scheduling decision with a brand consequence. Wrapping both vehicles in the same week would have taken two vans off the road at once, in a business whose ninety-minute commitment depends on them, and it would also have produced a single day on which the entire brand changed in public. Doing them one at a time kept a vehicle available and produced a sequence of small, unremarkable changes that customers noticed individually and mentioned conversationally, which is a gentler way for a business to change its look than a single launch. The uniforms were ordered at the same time as the first wrap, five garments for each of the seven people, so that the person at the door and the van at the curb changed within the same few weeks.

The depot sign came last, in month five. It is the least frequently seen touchpoint in the physical group and it is the one most bound by the language rules, so it was designed with the French generic term « Entretien » and the French tagline in the larger space and reviewed before production. Putting it last also meant that the business had already lived with its new identity for several months before committing a permanent structure to it, which is a cheap way to check that a design works before it is cast in a fixed form. The printed stock followed the same logic: business cards, door hangers and identification badges were produced when the old ones ran out, and the parts labels left on equipment were updated with the first print run after the refresh.

The budget is set out line by line in the callout below, and it totals $24,120, which is 5.4% of the $447,012 run-rate, spent once. To that figure the business added sixty-four hours of the administrator's time at $36 an hour, which is $2,304, for the work of applying the templates, the signatures and the profile. The total is the price of moving thirty-six touchpoints from thirteen consistent to thirty-three consistent, and it is worth comparing with the alternative, which is spending the same money on a van wrap and leaving the documents where they were.

The first customer comments arrived within a month and they were about the documents, which is the outcome the ranking predicted. Two property managers mentioned the monthly report had become easier to read; one asked whether the capital planning note would now arrive in the same format each year; nobody mentioned the colors. The first comment about a van came in month four, from a customer who saw it parked at another building in the same portfolio and said the business looked bigger than they had thought. Chapter 8 places this rollout inside the wider question of evolution and revolution, and Chapter 9 measures whether any of it moved the numbers, with the caution that the case does not attribute its growth to the brand because the offer, the entry assessment and referrals moved at the same time.

DO THE MATH Where Nordec's rollout money went

Illustrative figures; the case is fictional. Step 1 — Identity and rights: the design studio $7,800 for the logo, palette, typography, eleven templates and the guide, plus the trademark agent $1,200, which is $9,000. Step 2 — Physical production: two vehicle wraps at $4,100 each is $8,200, uniforms of five garments for seven people at $42 is $1,470, the depot sign is $1,900 and the printed items are $650, which is $12,220. Step 3 — Digital: the website update of Book C2 is $2,900. Step 4 — The rollout totals $9,000 + $12,220 + $2,900 = $24,120, which is 5.4% of the $447,012 run-rate, spent once, plus 64 hours of the administrator's time at $36, or $2,304. Step 5 — The documents, seen about 1,200 times a year between 624 visit sheets and 576 monthly reports, cost nothing extra once the templates exist, which is why they were funded first and why the physical group was funded third.

CASE IN POINT Nordec Entretien inc.: one van at a time

Nordec Entretien inc. is a fictional business created for this encyclopedia; its owner, its employees, its customers, its competitors, its figures and the events described are invented, and any resemblance to a real business is coincidental. The decision to wrap one van at a time was made for operational reasons and turned out to be a brand decision. Nordec runs two vehicles for 624 preventive visits and about 276 urgent calls a year, and the ninety-minute commitment does not pause while a van is in a shop. Wrapping both in the same week would have left the business with one vehicle for the days the work took, which the lead technician judged unacceptable at the start of a heating season. Wrapping them four weeks apart meant that a vehicle was always in service, and it also meant that customers met the new look gradually, at one building after another, rather than receiving a letter announcing that everything had changed. The dispatcher reported the practical result: for about a month the business had one van that looked like the future and one that looked like the past, and the customers who saw both never mentioned it once.

Costly Mistakes

• Buying the visible touchpoint first. A business that wraps a van before it fixes a visit sheet has spent its money on the touchpoint its own staff see most and its customers see least, and has left the more frequent touchpoint saying the opposite.
• Treating the brand experience as a marketing task. The promise is delivered at the door, on the telephone and in the invoice, which means the people who carry it sit in operations, in dispatch and in administration rather than in a marketing function.
• Letting the digital touchpoints grow undecided. A website, a business profile, seven signatures, a handful of social pages and a voicemail accumulate over years, and a prospect meets them all before the first conversation.
• Auditing the design and not the promise. A monthly report that uses the right colors and arrives without the forward-looking line passes five criteria and fails the one that matters to the customer.
• Designing a sign or a van before the language rules are applied. In Québec, French must be markedly predominant on a sign visible from outside, and that requirement belongs in the first sketch rather than in a review after production.
• Rolling out everything in one week. A single launch day takes vehicles off the road, forces printed stock into the bin and gives customers one change to react to instead of a sequence of small ones.
• Keeping the old materials in circulation. An old letterhead used for a single urgent letter, or a spare box of door hangers used to save money, reintroduces the version of the brand the business has just retired.

Key Takeaways

• A brand is the sum of what customers experience at its touchpoints, and a change in aesthetics weighs less on brand equity than the behavior of employees (Muzellec & Lambkin, 2006).
• Resonance sits at the top of the customer-based brand equity model built on four steps and six building blocks, and it is produced by the accumulation of touchpoints rather than by any single one (Keller, 2001).
• Mapping a brand means listing its touchpoints in four groups — people, physical, documents and digital — counting the exposures a year, rating the importance at a moment of truth, and ranking before spending.
• Nordec's thirty-six touchpoints divide as four people, seven physical kinds, eleven documents and fourteen digital ones; the documents are seen about 1,200 times a year between 624 visit sheets and 576 monthly reports, which is more often than any advertisement.
• Consistency is audited against six criteria — logo version, colors, typefaces, voice, French first and the promise respected — scored as pass or fail, and held as the share of touchpoints that pass; Nordec moved from 13 of 36 to 33 of 36.
• In Québec, a sign visible from outside carrying a trademark in another language must be accompanied by French terms that make French markedly predominant under rules in force since June 1, 2025 (Office québécois de la langue française, 2025a), and on screens, text outside a logo needs a contrast ratio of at least 4.5:1, or 3:1 for large text (World Wide Web Consortium, 2024).

Your Action Plan

1. List every touchpoint in four groups. Walk through a year from a real customer's side and write down every place the business is seen, heard or read, grouping the results as people, physical, documents and digital. Deliverable: one touchpoint list in four groups. Verification: every touchpoint has a named person who owns it, and the total is checked against the business's own volume of visits, reports and calls.
2. Count the exposures and rank the list. Estimate how often each touchpoint is met in a year, mark the ones that fall at a moment of truth, and rank by exposure and importance together. Deliverable: a ranked list with an exposure figure for each touchpoint. Verification: the three most frequent touchpoints are identified and the ranking differs visibly from the order in which the business would have spent by instinct.
3. Apply the identity to the top of the list first. Start with the documents and the digital touchpoints, using the templates and the guide, and confirm each one against the guide's one-page summary before it goes into circulation. Deliverable: the most frequent touchpoints produced from the templates. Verification: a visit sheet and a monthly report taken from live use carry the same mark, colors, typefaces and forward-looking line.
4. Brief the people who carry the brand. Write a short brief for the technicians, the dispatcher and the administrator, stating the decisions the values change and the four principles of the voice, and give it to each of them before their next building or call. Deliverable: three written briefs, delivered. Verification: each person can state the promise and one decision their value changes without reading the brief.
5. Audit consistency and set the next date. Score a fixed sample against the six criteria, record the share that passes, list the failures by the exposure of the touchpoint they belong to, and place the next audit on the calendar. Deliverable: one dated audit with a score and a ranked list of failures. Verification: the score is computed from the sample rather than estimated, and every failure has an owner and a date.

In La Machine

Four pages in the application copy read for this book carry part of the digital group (La Machine, 2026). Business → My Website, at business/website with its builder and preview, holds a Theme & Branding panel with a gallery of 28 themes, a brand name, a tagline, a logo, a favicon and a banner image, five color overrides and a heading and a body font chosen from a catalog of 200, and the configuration is kept in the browser. Assets → Visual Studio at business/social/visual-studio is an editor for social, video, print and web formats whose designs are saved and whose exports go to the Gallery. Marketing → Templates at business/marketing/templates stores message templates by channel with variables and a preview. Social → Accounts at business/social/accounts lists connected accounts on nine platforms, while profile pictures and bios are edited outside the platform.

What the platform does not do is as important as what it does. It has no saved brand kit that other pages use, so the colors entered on the website page do not travel to the visual editor or to the message templates; it has no brand portal or guidelines publisher, no digital asset management with versions, usage rights or expiry dates, and no brand consistency audit. Nothing in it checks that the touchpoints agree, which means the audit of section 7.7 is performed by a person with the guide's one-page summary and a photograph of each touchpoint in use. Chapter 5 covers the system and the guide those checks are held against, Chapter 8 places the rollout inside the wider question of evolution, and Chapter 9 sets out how the results are measured.

Self-Check

1. Why does a brand belong to the touchpoints rather than to the guide, and what does the research on corporate rebranding say about the relative weight of aesthetics and employee behavior?
2. What are the four groups of touchpoints, and why are they grouped rather than left as one list?
3. How is a touchpoint ranked, and why does the ranking usually differ from the order in which a small business spends its money?
4. What are the six criteria of the consistency audit, and which one cannot be checked by looking at a color?
5. In what order did Nordec roll out its identity, and why was the depot sign produced last?

Answers. 1. A brand is the sum of what customers experience at every touchpoint, because the promise in the guide is realized only when somebody meets the business; the analysis of 166 rebranded companies found that a change in marketing aesthetics affects brand equity less than factors such as employees' behavior, and that rebranding must be managed as a whole with the support of stakeholders (Muzellec & Lambkin, 2006). 2. The four groups are people, physical, documents and digital; they are grouped because each group is managed by a different part of the business and fixed in a different way, so the grouping makes visible which part of the business has to act. 3. A touchpoint is ranked by estimating its exposures a year and rating its importance at a moment of truth, the two combined placing the frequent and decisive touchpoints first; the ranking usually differs from instinct because a business spends first on the touchpoint it sees itself — the van, the sign, the website — while the exposures point to the documents and the people. 4. The six criteria are the logo version, the colors, the typefaces, the voice, French first where required, and the promise respected; the sixth cannot be checked by looking at a color, because it is checked by reading what the touchpoint says and comparing it with what was promised. 5. Documents and digital first, in days one to sixty; uniforms and the two vans in months three and four, wrapped one at a time so that a vehicle was always on the road; and the depot sign in month five, because it is the least frequently seen physical touchpoint, it is bound by the language rules, and it was worth checking that the identity worked in daily use before committing a permanent structure to it.

Chapter References

Charter of the French language, CQLR c C-11 (Québec). LégisQuébec, updated to June 10, 2026 — as read for Book C2 (Websites, Landing Pages and Conversion); its rules on trademarks, signs, products and business names are stated in this book through the explanations of the Office québécois de la langue française and of the Registraire des entreprises.

Keller, K. L. (2001). Building customer-based brand equity: A blueprint for creating strong brands (Report No. 01-107). Marketing Science Institute.

La Machine. (2026). Documentation — route reference inventory. Internal product documentation, read September 25, 2026, together with the code of the pages this book describes: business/website, business/social/visual-studio, business/marketing/templates and business/social/accounts.

Muzellec, L., & Lambkin, M. (2006). Corporate rebranding: Destroying, transferring or creating brand equity? European Journal of Marketing, 40(7/8), 803–824. https://doi.org/10.1108/03090560610670007

Office québécois de la langue française. (2025a). Affichage des marques de commerce et des noms d'entreprises (updated August 27, 2025). Retrieved September 25, 2026, from https://www.oqlf.gouv.qc.ca/francisation/entreprises/affichage-marques-noms.html

Office québécois de la langue française. (2025b). Marque de commerce sur les produits (updated August 27, 2025). Retrieved September 25, 2026, from https://www.oqlf.gouv.qc.ca/francisation/entreprises/marque-commerce-produits.html

Romaniuk, J. (2018). Building distinctive brand assets. Oxford University Press.

World Wide Web Consortium. (2024). Web Content Accessibility Guidelines (WCAG) 2.2. W3C Recommendation, 12 December 2024 (first published as a Recommendation on 5 October 2023). Retrieved September 25, 2026, from https://www.w3.org/TR/WCAG22/

Chapter 8 — Rebranding and Brand Evolution

A brand changes when the business has changed, not when its owner is bored of the logo, and the discipline of a rebrand is to keep what people already recognize, drop what was accidental, and move every touchpoint in an order that never leaves customers confused.

What You'll Be Able to Do

• Distinguish a refresh, an evolution and a revolution, and choose the smallest degree of change that does the work the business actually needs.
• Recognize the good reasons to rebrand — a merger or an acquisition, a new strategy or offer, a name that cannot be protected, a legal constraint, an identity that no longer works on screens — and the bad ones.
• Inventory your distinctive brand assets before touching anything, so that the elements people already recognize are named, protected and kept.
• Test a candidate identity with six to ten long-standing customers, reading the objections of the most committed separately from the enthusiasm of everybody else.
• Plan the transition in phases, with an order of touchpoints, a stock count, a date after which nothing old goes out, and an announcement that reaches customers before it reaches the public.
• Complete the legal and administrative steps a change of name or logo requires, from a trademark search to the assignment of rights in the new files and the update of a business register.

Why It Matters

A brand is the part of a business that customers carry in their heads, and that is exactly why changing it is delicate. The name, the color, the letterform and the tagline are not decorations on a van; they are the handles by which a customer finds the business again among a hundred others. When those handles change, the customer has to do work, and a customer who is asked to do work at the wrong moment simply stops. The whole craft of rebranding consists in deciding which handles must change, which must stay, and in what order the change reaches the people who pay.

The temptation to change everything at once is strong, and it is usually a mistake. A business that has grown, added services, hired people and moved into new territory often feels that its identity has fallen behind, and it has. But the gap between what a business is and what its identity says is rarely closed by replacing every element. It is closed by keeping the elements that carry recognition and replacing the ones that carry nothing — the accidental wrench, the fourth shade of blue, the typeface nobody chose. The research on distinctive brand assets addresses those handles: logos, colors, fonts, characters, taglines and jingles are a studied field, and the work examines which branding elements succeed and which fail over the long term (Romaniuk, 2018). Throwing away an asset that works is not a fresh start; it is a donation to competitors.

There is also a cost to getting it wrong that a small business cannot absorb. A rebrand that confuses customers, that arrives before the service is ready, or that is reversed under pressure spends money twice and spends trust once. The public record contains both kinds of outcome, and the useful lesson is not that change is dangerous but that the sequence matters: test with the people who know you best, keep what they name, announce to customers before the public, and let the promise stay still while the surface moves. This chapter sets out that sequence for a business of any size, and follows one fictional business through a refresh that kept its name, its blue family and its letter N, and dropped everything that had been accidental.

In This Chapter

Section 8.1 sets out why and when to rebrand, separating the structural reasons from the cosmetic ones. Section 8.2 places a refresh, an evolution and a revolution on a single scale of degrees of change. Section 8.3 turns to what must be kept, through an inventory of distinctive brand assets. Section 8.4 reads two public reversals for their lesson about committed customers. Section 8.5 plans the transition, phase by phase, with an order of touchpoints and a date after which nothing old goes out. Section 8.6 covers the announcement, to customers first and to the public last. Section 8.7 lists the legal and administrative steps a change of name or logo requires. Section 8.8 follows the case's own refresh from the decision to the last van.

8.1 Why and When to Rebrand

The first question in any rebrand is not what the new identity should look like. It is whether the business has actually changed, and in what way. A brand is a promise the business makes and repeats, and a rebrand is justified when the promise has moved — when the offer, the market, the ownership or the legal position of the business is no longer the one the identity was built for. Everything else is taste, and taste is a poor reason to spend money and a worse reason to ask customers to relearn a name.

The strongest reason is structural change, and the research on corporate rebranding is clear about which structural change dominates. A study of 166 rebranded companies, with two detailed cases alongside, found that structural change — above all mergers and acquisitions — is the most common trigger for a rebrand (Muzellec & Lambkin, 2006). The reason is easy to see. When two businesses become one, the customers of both hold two sets of handles, two names, two colors and two histories, and something has to give. The study also found that a change in marketing aesthetics affects brand equity less than factors such as employees' behavior, and that a rebranding must be managed as a whole, with the support of stakeholders (Muzellec & Lambkin, 2006). That second finding is the one small businesses most often ignore: the logo is the visible part of the change and the least consequential part of it.

A second good reason is a change of strategy or offer. When a business stops doing one thing and starts doing another — when it adds a service line, moves upmarket, or narrows to a specialty — the identity that described the old business begins to mislead. In the case followed through this book, the business added three named service tiers and a fixed-price assessment, and the identity it had carried since its founding had no names for any of them. The gap was not aesthetic; it was operational. A customer could not ask for a tier by name because the tier had no name in the customer's language.

A third good reason is a name that cannot be protected. A name that is too close to an existing registered mark, or too descriptive to distinguish anything, is a liability that grows with the business. A trademark agent assesses registrability and the risk of confusion, and the answer sometimes forces a change that no amount of design can avoid. A fourth is a legal constraint: a rule that requires French to be markedly predominant on a sign visible from outside a premises, or that requires the generic words inside a product mark to appear in French, can make an existing sign or package non-compliant, and compliance is not optional. Section 8.7 sets out those rules as the Office québécois de la langue française explains them.

A fifth good reason is narrower and increasingly common: an identity that no longer works on screens. A logo drawn for a sign, with fine lines and a wide lock-up, can become illegible at the size of a favicon or a profile picture, and a palette chosen for print can fail a contrast threshold on a white page. The Web Content Accessibility Guidelines set the thresholds that matter here: text at least 4.5:1 against its background at level AA, large text at least 3:1, and graphical objects needed to understand content at least 3:1 against adjacent colors (World Wide Web Consortium, 2024). An identity that cannot meet those numbers on the surfaces where the business now lives has a technical reason to change, and a technical reason is a good one.

The bad reasons are worth naming because they arrive dressed as good ones. Boredom is the most common: the owner has seen the logo every day for four years and can no longer see it at all. A competitor's redesign is the second: the regional integrator repaints its fleet and the smaller business feels shabby by comparison, although nothing about its own promise has moved. A bad quarter is the third: a slow season produces the feeling that the brand is the problem, when the problem is usually the offer, the price or the follow-up. And a new owner's taste is the fourth: a change of ownership is a real structural event, but the taste of the new owner is not a strategy, and customers do not owe the new owner a relearning.

DEFINITION Rebrand, refresh, brand evolution and brand migration

A rebrand is a deliberate change to the elements that identify a business — its name, its logo, its colors, its typography, its tagline or its architecture — made because the business itself has changed. A refresh is the smallest form of rebrand: the same name and the same core assets, with details corrected, simplified or brought up to date. A brand evolution is a larger change that keeps the assets people already recognize and replaces the ones that carry nothing, so that the brand looks and sounds like the same business at a later stage of its life. A revolution is a change of name, or of the whole identity, that asks customers to learn something new; it is justified by a merger, an acquisition, a legal constraint or a name that cannot be protected, and it is the most expensive and the riskiest of the four. Brand migration is the operational work of moving every touchpoint from the old identity to the new one — the templates, the vehicles, the uniforms, the signs, the accounts and the records — in an order that never leaves a customer facing a business they do not recognize.

WATCH OUT Rebranding to fix a service problem

A rebrand is the most expensive way to avoid fixing an operational problem, and it is the one businesses reach for first. When response times slip, when invoices arrive with surprises in them, when the phone is answered badly, the identity becomes the visible thing to blame, because it can be changed in a month while the service takes a year. The research on rebranding points the other way: a change in marketing aesthetics affects brand equity less than factors such as employees' behavior (Muzellec & Lambkin, 2006). A new logo on a late technician is a new logo on a late technician. Before approving any change to the identity, a business should be able to name the operational change that accompanies it, and to say how it will be measured. If the answer is that nothing operational changes, the honest conclusion is that the business wants a refresh of its surfaces and should say so, rather than promising customers a new business and delivering a new color.

The timing question has a practical answer. A rebrand should not be launched in the middle of a busy season, in the week a major contract is being renewed, or immediately before a price change, because each of those moments already asks customers to pay attention. It should be launched when there is time to run the old and the new side by side, when the people who answer the phone have been told what to say, and when the stock of printed items is low enough that the change costs little. The best moment is usually the one the business can plan three months ahead, and the worst is the one it cannot.

8.2 Refresh, Evolution or Revolution

Once a business has decided that something must change, the next decision is how much. The scale runs from a correction of details to a new name, and the discipline is to choose the smallest degree that does the work. A larger change than necessary costs money, spends recognition and asks customers to relearn something they had already learned for no reason they can see. A smaller change than necessary leaves the business with the same problem and a bill.

The smallest degree is the refresh. A refresh keeps the name, the core color and the central mark, and corrects what has drifted: the five shades of blue that accumulated over four years, the four typefaces that arrived with four different projects, the logo files that exist in three versions and get used at random. A refresh is cheap, it is fast, and it is almost invisible to customers, which is its point. It is the right answer when the business has not changed but its materials have decayed, and it is the answer that most small businesses actually need when they believe they need a rebrand.

The middle degree is the evolution. An evolution keeps the assets that carry recognition — usually the name, the color family and one distinctive shape — and replaces the rest. It is the degree that requires the most judgment, because every element must be sorted into one of three piles: keep, drop or add. The case followed through this book chose this degree. It kept the name Nordec, the blue family and the letter N; it dropped the generic wrench, four of the five blues and the four typefaces, replacing the typefaces with two; and it added a horizon line drawn through the letter N, a second color called Signal Amber, a tagline in French first, and French names for the service tiers. The result is recognizably the same business, which is the whole objective.

The largest degree is the revolution, and it is a change of name or of the entire identity. A revolution is justified by a merger or an acquisition, by a name that cannot be protected, by a legal constraint, or by a business that has genuinely become a different business. It is the most expensive degree because every touchpoint must change, every record must be updated, and a new trademark application is usually required. It is also the degree with the longest shadow: customers who knew the old name will use it for years, and the business must decide in advance how it will answer the phone when they do.

The following table sets out the three degrees side by side, with what changes, what stays, the trigger that usually justifies each one and the risk each one carries. Read it as a decision aid rather than a ladder: the degrees are not stages a business passes through, and a business may spend a decade at the middle one without ever needing the largest.

Table 8.1 — Degrees of change

The figure below places the case's own decision on that scale. Figure 8.1 shows three columns — what Nordec keeps, what it drops and what it adds — and, below them, the three phases of the rollout that carries the change to every touchpoint.

Figure 8.1 — Nordec's refresh: keep, drop, add. The assets people already recognize stay; what was accidental goes.

The choice of degree is made easier by one test, which is to ask what a customer would have to relearn. A refresh asks the customer to relearn nothing. An evolution asks the customer to notice a change and to keep the same mental handle, which is a small and usually welcome effort. A revolution asks the customer to replace one handle with another, and that is a real cost paid by somebody who did not ask for it. The test is not a rule, but it sorts most proposals quickly: if the answer is that customers must relearn the name, the business had better have a merger, a legal constraint or an unprotectable name behind it.

KEY CONCEPT Change what was accidental, keep what is recognized

The central idea of this chapter is a sorting rule. Every element of an identity is either an asset that people already recognize or an accident that arrived without a decision, and the two must be treated in opposite ways. The name, the color family and one distinctive shape are usually assets: they are what a customer uses to find the business again, and they should be kept unless a strong reason says otherwise. The generic symbol, the fourth shade of blue, the typeface that came with a template and the logo version that only one person uses are usually accidents: they carry no recognition, they cost consistency, and they should go. The rule is easy to state and hard to apply, because accidents feel like assets to the people who have seen them every day for four years. That is why the sorting is done with customers rather than with the owner, and why the test in section 8.4 exists.

WHAT CHANGED A logo reversed in a week, October 2010

In early October 2010, the retailer Gap unveiled a new logo, replacing the white-on-blue box it had used for more than twenty years. The reaction was immediate and largely hostile, and it came mainly through social media. The company's North American president first defended the new logo. On October 11, 2010, about a week after the unveiling, the company announced that it would return to the old logo (NBC News, 2010). The episode is worth reading carefully, because it is often told as a story about public opinion being fickle. It is more usefully read as a story about sequence. The change was made to an asset that a large and committed group of customers used to recognize the business, it was made without a visible reason, and it was announced before it had been tested with the people most attached to what was being replaced. The reversal cost a week of attention and the credibility of the process, and no figure for its cost was published.

The degree of change also determines how long the transition takes and how much of it can be done quietly. A refresh can be completed in a month, mostly by replacing files. An evolution takes a quarter or two, because vehicles, uniforms and signs have their own schedules and their own budgets. A revolution takes a year or more, because records, contracts, accounts and legal filings all move with the name. A business that has chosen the degree has therefore also chosen its calendar, and the calendar is what section 8.5 turns into a plan.

8.3 What to Keep: Distinctive Assets

Before a single element is redesigned, the business should know what it already owns. The inventory of distinctive brand assets is the most useful hour in a rebrand, and it is the hour most often skipped, because the people doing the work are impatient to get to the design. An asset inventory is not a list of everything the business has ever used. It is a short list of the elements that a customer could use to recognize the business without being told its name, and each one is assessed for how widely it is recognized and how much of that recognition the business can claim.

The concept behind the list is distinctive brand assets, and the field is a researched one: logos, colors, fonts, characters, taglines and jingles have been studied over the long term, and the work examines which branding elements succeed and which fail (Romaniuk, 2018). The practical consequence for a small business is that recognition is built by repetition of the same few elements, not by novelty. An element used consistently for four years is worth more than a better element introduced last month, because the value is in the memory, not in the drawing. This is why the inventory comes before the design brief and not after it.

The inventory is built by asking four questions of each candidate element. First, is it distinctive — could a customer pick it out of a line-up of competitors' materials? Second, is it recognized — do customers actually associate it with the business, or is it only familiar to the people inside? Third, is it ownable — can the business use it without confusion with another business, and can it be protected? Fourth, is it durable — will it still work in five years, at the sizes and on the surfaces the business will be using? An element that fails the second question is an internal favorite rather than an asset, and an element that fails the fourth is a temporary solution that has been promoted to a permanent one.

The governance phase of the identity process is where this work belongs in a formal project. The five-phase identity process runs through research, strategy, design, launch and governance, and governance is a phase of the work rather than an afterthought, with checklists, tools and case studies attached to it (Wheeler & Meyerson, 2024). An asset inventory is the first artefact of that phase, and it is what makes the difference between a rebrand that keeps the business's recognition and one that trades it for a designer's preference. It is also what makes the next rebrand cheaper, because the business that knows which elements carry recognition can protect them without redoing the analysis.

The following table sets out the inventory for the case followed through this book, taken before any design work began. It lists each candidate asset, what it is today, who recognizes it, the verdict reached and the reason. Read it as a worked example of the four questions rather than as a template to copy: the verdicts depend on what customers actually recognize, and that differs from business to business.

Table 8.2 — Nordec's distinctive assets before the refresh

The inventory has a second use that is easy to miss. It is the document that makes the case for the change to the people inside the business, including the ones who liked the wrench. A verdict written down with a reason is easier to accept than a decision announced, and the reason for dropping the wrench — that it is a category symbol used by every trade in the region, and cannot be protected — is a reason a technician can repeat to a customer. The inventory also protects the business from its own future: when somebody proposes a new color in two years, the list says which one is the asset and why.

METHOD Testing a redesign with your most committed customers

First, choose six to ten long-standing customers — the ones who have renewed most often, who refer others, and who would notice if the business changed. Second, show the current identity and the candidate routes side by side, in the same room or on the same call, without explaining which one the business prefers. Third, ask what they recognize and what they would miss, in that order, and write down their words rather than your interpretation of them. Fourth, note the objections from the most committed customers separately from the reactions of the rest, because the research in section 8.4 shows that the two groups respond differently to the same change. Fifth, keep the assets they name unless a strong reason says otherwise, and record the reason when you do not. Deliverable: a written note of what each customer recognized, what they would miss and which route they preferred. Pitfall: showing the routes to the people who will implement them rather than to the people who use them, which produces approval for the wrong design.

The inventory also settles an argument that otherwise runs for weeks. When the business can see, in one table, that the name is recognized by referral customers and that the wrench is recognized by nobody, the debate about the wrench ends. What remains is the harder question of what to add, and that question is answered by the strategy rather than by the design: the new element should express something the business actually does and the customer actually values. In the case, the horizon line drawn through the letter N expresses the promise of telling the owner ahead, and Signal Amber expresses the safety and warning side of the trade. Neither is decoration; both carry a meaning the business can explain in a sentence.

8.4 Lessons from Public Reversals

Two public episodes are worth reading closely, because together they explain why the order of a rebrand matters more than the quality of its design. The first is a field experiment on logo redesigns and the customers who are most attached to a brand. The second is the reversal of a retailer's new logo in October 2010. Neither is a story about bad design. Both are stories about who is asked first.

The experiment tested what happens when a logo is redesigned from angular to more rounded. It involved 632 participants and two sportswear brands' logos, and the result was a split that any small business should know before it changes anything. When the logo was redesigned from angular to more rounded, strongly committed consumers judged it more negatively and their attitude to the brand fell, while weakly committed consumers responded positively (Walsh et al., 2010). The study measured attitudes, not sales, and it does not say that every rounded logo harms every brand. What it does say is that the people most attached to a brand are the most sensitive to a change in its appearance, and that their reaction runs in the opposite direction from everybody else's. A business that tests a redesign with a general audience will hear the positive half of that split and conclude that the change is welcome.

The second episode is the retailer's reversal. In early October 2010, the retailer Gap unveiled a new logo, replacing the white-on-blue box it had used for more than twenty years. The reaction was immediate and largely hostile, and it came mainly through social media. The company's North American president first defended the new logo, and on October 11, 2010, about a week after the unveiling, the company announced that it would return to the old logo (msnbc.com staff and news service reports, 2010). No cost figure was published, and the episode should not be told as a story about the power of online complaint. The useful reading is narrower and more practical: the change was made to an asset that a large and committed group of customers used to recognize the business, it was made without a visible reason, and it was announced before it had been tested with the people most attached to what was being replaced.

Put the two together and the lesson is a sequence rather than a warning. Test with the committed customers first, because they are the ones whose reaction the general audience will not show you. Give the change a reason the customer can repeat, because a change with a reason is a decision and a change without one is a whim. And keep the asset that carries recognition unless there is a strong argument against it, because the committed customers are attached to it for the same reason the business should be: it is how they find you.

The sequence has a second half, which concerns what to do with the answers. A business that tests with six long-standing customers will usually receive a mixture of reactions, and the temptation is to count them. Counting is the wrong method, because the committed customers are not a sample of the market; they are the people whose recognition the brand depends on. The right method is to read their objections one by one and ask whether each objection names an asset or a preference. An objection that names an asset — "I would not know it was you without the blue" — is a finding. An objection that names a preference — "I liked the old one better" — is a reaction, and reactions are expected whenever anything changes.

CASE IN POINT Nordec Entretien inc.: six customers and three routes

Nordec Entretien inc. is a fictional business created for this encyclopedia; its owner, its employees, its customers, its competitors, its figures and the events described are invented, and any resemblance to a real business is coincidental. The independent design studio presented three routes, and Sophie Nadeau showed all three to six long-standing customers before choosing. Four of the six preferred the route that kept the letter N, and none of the six asked to keep the wrench. The result is worth reading against the experiment in this section. The customers who were most attached to Nordec were not the ones who resisted the change; they were the ones who named the element that had to survive. The wrench, which the business had assumed was its most recognizable symbol because it was the most literal, was recognized by nobody as belonging to Nordec in particular. The letter N, which the business had treated as a detail, was the element four of the six named. The test cost two hours and settled a question that the workshops had argued about for a week.

The same reasoning applies to the timing of the test. It should happen before the routes are finalized, not after, because a test that can only approve or reject a finished design is not a test; it is a presentation. The routes should be shown side by side, in the same session, so that the customer is comparing rather than reacting. And the questions should be about recognition and loss rather than about preference, because preference is easy to give and recognition is what the business needs. A customer who says "I would miss the blue" has told the business something it can act on. A customer who says "the second one looks nicer" has told it almost nothing.

The final lesson from both episodes concerns the announcement, which is the subject of section 8.6. A change that has been tested with committed customers, that keeps the assets those customers named, and that has a reason the customer can repeat, arrives with its own defense. A change that has none of those three things arrives with nothing, and the only remaining argument is the taste of the person who approved it. That is the position the retailer was in for a week in October 2010, and it is a position no small business can afford to occupy.

8.5 Planning the Transition

A rebrand is not finished when the new logo is approved. It is finished when the last old touchpoint has been replaced, and the work between those two moments is a migration: an ordered list of everything that carries the identity, with a date, an owner and a condition for completion attached to each item. The migration is where a rebrand succeeds or fails in practice, because it is the part that touches customers directly and the part that is easiest to leave half done.

The first task is the touchpoint list. Every place the identity appears is written down, from the website and the email signature to the vehicle wraps, the uniforms, the depot sign, the invoices, the reports and the profile pictures on social accounts. In the case followed through this book, the inventory of touchpoints counted 36, of which only 13 were consistent with one another before the refresh. The list is the same document that Chapter 7 uses to describe the brand across every touchpoint, and it is the document the migration is planned from. A touchpoint that is not on the list will be discovered by a customer.

The second task is the order. The general principle is to move the surfaces that are cheapest, most frequent and most easily corrected first, and the surfaces that are expensive, infrequent and permanent last. Documents and digital materials come first, because they are produced in-house, they are seen constantly, and a mistake costs nothing to fix. Uniforms and vehicles come next, because they are seen by the public every working day but they are bought in batches and cannot be changed overnight. The depot sign comes last, because it is the most expensive single item and the least frequently seen by the customers who matter. In the case, the order was exactly that: documents and digital first, then vans and uniforms, then the depot sign.

The third task is running the old and the new side by side. A business that switches everything on one day creates a period in which its materials contradict one another, and a customer who sees two different identities in the same week concludes that something is wrong. The remedy is a transition window with a stated end: the new materials are used for everything produced from a given date, the old materials are used until the stock runs out, and a date is fixed after which nothing old goes out, even if stock remains. The date is what makes the window a plan rather than a drift, and the stock count is what makes the date realistic.

The fourth task is the stock count of printed items. Printed materials are bought in batches and they do not change when the design changes. A business that has 2,000 old letterheads in a cupboard has a choice between using them and writing them off, and the choice is made better with a number in front of it. The count is taken before the order is placed for the new items, so that the quantity ordered reflects what is actually needed, and the old stock is either used within the window or recycled on the date the window closes.

The fifth task is the vehicles, and they are handled one at a time. A vehicle wrap is the most visible and the most expensive per unit of the touchpoints a small business owns, and wrapping the whole fleet on one day takes every vehicle off the road at once. Doing them one at a time spreads the cost, keeps the fleet working, and produces a period in which the old and the new vans are both on the road, which is normal and harmless as long as the window has an end. In the case, two vehicles were wrapped at $4,100 each, and the second was scheduled after the first had been on the road for a month.

The following table sets out the case's transition plan in four phases, with the timing of each, the touchpoints it covers, the person who owns it and the condition that marks it done. Read it as a worked plan rather than a schedule to copy: the phases are the same in most businesses, and the timings depend on how much stock is in the cupboard and how many vehicles are on the road.

Table 8.3 — Nordec's transition plan

TOOLS OF THE TRADE What a transition needs

Five ordinary things carry a migration, and none of them is a specialist product. The first is a touchpoint list with a date per item: one row for each place the identity appears, with the phase it belongs to, the person who owns it and the date it must be done. The second is a stock count of printed items, taken before the new order is placed, so that the quantity ordered matches what will actually be used and the old stock has a closing date. The third is a template for the announcement letter, one page, in the languages the customers read, with fixed sections for what changes, what does not change, when the vehicles change and how to reach the business. The fourth is a checklist of accounts to update, covering the domain, the email addresses, the social handles, the profiles on directories, the bank and supplier records and the trademark filings. The fifth is a single folder for the new assets, with the logo files, the color values, the typefaces and the templates in one place, so that nobody has to ask which version is current. No brand names belong on this list; the requirement is the function.

IN LA MACHINE What to update in the platform after a refresh

A refresh reaches the platform in a few specific places, and it is worth knowing which ones (La Machine, 2026). Business → My Website (business/website) has a Theme & Branding panel with a gallery of 28 themes, a brand name, a tagline, a logo, a favicon and a banner image, five color overrides and a heading font and a body font chosen from a catalog of 200; the configuration is kept in the browser, so the new logo, the new colors and the two typefaces are entered there. Assets → Gallery (business/marketing/gallery) is a library of royalty-free stock images that records, for each image, its license, the license's address and the attribution text, plus a provenance form for the user's own images, which is where new photography is filed. Marketing → Templates (business/marketing/templates) stores message templates by channel with variables and a preview; it holds no brand colors, logo or footer, so the templates are updated by hand. Social profile pictures and bios are not edited from La Machine; they are changed on each platform. And Business → Branding (business/branding) is a place to try a palette and see its contrast, not a brand register: its values are not saved, and the "Save branding" button only shows a confirmation, so it is not the record of the change.

The migration has one more requirement that is easy to forget, and it is the reason the phase list has a closing date. Some touchpoints are not owned by the business at all: a directory listing, a supplier's catalog, a partner's website, a customer's own document. Those are updated by asking, and asking takes time, which is why the request goes out in the first phase rather than the last. A business that waits until everything else is done before writing to the directories will find that the old logo is still on somebody else's page a year later, and that the page is one a prospect reads.

8.6 Announcing the Change

The announcement is where a rebrand meets its audience, and the order of the audience is the whole of the craft. Customers come first, because they are the people who have to recognize the business tomorrow morning. Suppliers and partners come second, because they need the new name and the new files for their own records. The public comes last, and for most small businesses the public announcement is not an event at all; it is simply the moment when the new materials are visible to anybody who looks. A press release is not required, and a business that stages a public reveal before its own customers have been told has the order backwards.

What to say is simpler than it looks, and it has four parts. The first is what changes: the look, the colors, the lettering, the names of the service tiers. The second is what does not change: the team, the contract, the guarantee, the phone number, the way the work is done. The third is when the visible items change, particularly the vehicles, because a customer who sees a new van and has not been told will wonder whether the business has been sold. The fourth is the promise, stated once and in the same words the business uses everywhere else. A letter with those four parts fits on one page, and one page is the right length.

The reason the second part matters most is the finding that a change in marketing aesthetics affects brand equity less than factors such as employees' behavior (Muzellec & Lambkin, 2006). The announcement is therefore not really about the aesthetics at all. It is an opportunity to restate the promise and to show that the behavior behind it has not moved. A letter that says "we have a new look, and the technician who answers at two in the morning is the same person" tells the customer what the change means for them, which is nothing. A letter that lists the new colors tells them what it means for the business, which is not their concern.

The channel matters less than the order. For a business with 52 contract customers, a letter or an email to each of them is the whole of the customer announcement, and it should arrive before the first new van is on the road. For a business with thousands of customers, the announcement goes into the channels the customers already read, in the same place they read everything else. In both cases the principle is the same: the announcement travels on a channel the customer already uses, in the language the customer already reads, and it does not ask the customer to do anything.

The internal announcement comes before the external one, and it is the one most often skipped. The people who answer the phone, drive the vans and write the reports are the ones who will be asked about the change, and they need the same four parts of the message, plus one more: what to say when a customer asks why. A one-page note and fifteen minutes at a team meeting are enough, and the note should include the reason for the change in the words the business wants repeated. A technician who has been told the reason will give it; a technician who has not will invent one.

WHO DOES THIS The owner decides, the brand keeper runs the migration

In a small business, the rebrand divides into two roles that should not be held by the same person for the whole project. The owner decides: whether to change at all, which degree of change to choose, which route to take, what the promise says and when the announcement goes out. Those are judgment calls that carry the business's strategy, and they cannot be delegated to a designer or an administrator. The brand keeper runs the migration: they hold the touchpoint list, the stock count, the template folder and the checklist of accounts, they check every new material against the guide, and they report what is done and what is not. In the case followed through this book, the administrator is the brand keeper, and Sophie approves any change to the promise or to the identifiers. The split matters because the two roles have different failure modes. An owner who runs the migration personally will leave it half done during a busy month; a brand keeper who decides the strategy will choose what is easiest to implement rather than what the business needs.

The announcement also has a closing condition, and it is worth writing down. The announcement is complete when every customer has received it, every person in the business can state the reason for the change in one sentence, and the first new touchpoint has appeared in public. Until those three things are true, the business is in the middle of a change rather than at the end of one, and the middle is where confusion lives.

A final point about the announcement concerns the customers who do not reply. Most of them will not, and silence is not disapproval. The measure of a good announcement is not the number of responses; it is whether the customers who call in the following weeks use the new name, ask about the new tiers by their new names, and continue to describe the business in the same terms as before. That is what recognition looks like from the inside, and it is the subject of Chapter 9.

8.7 Legal and Administrative Steps

A change of identity has a legal and administrative tail, and the length of the tail depends on the degree of change. A refresh that keeps the name and the core mark produces new files and new materials, and almost nothing else. An evolution that keeps the name but changes the logo produces new files, a possible new trademark application for the new mark, and a set of records to update. A revolution that changes the name produces all of that plus a new business name, a new trademark application, new domain names, new email addresses and a long list of accounts to correct. The steps below are the ones that apply in Canada and Québec, and they are described as the sources explain them; this book explains the law and does not replace the advice of a lawyer or a trademark agent.

The first step is the trademark search, and it comes before the design is finalized rather than after. The Canadian Intellectual Property Office maintains the Canadian Trademarks Database, which can be searched free of charge in a basic and an advanced search, and a search is how a business finds out whether a proposed name or mark is already in use (Canadian Intellectual Property Office, 2026b). A search is not a clearance opinion, and a trademark agent assesses registrability and the risk of confusion. In the case followed through this book, a knockout search found two registered marks close to "Nordec" in unrelated classes of goods, and the business filed an application for the word mark NORDEC in Nice class 37, which covers building maintenance and repair services, through a trademark agent.

The second step is the application itself, where the name or the mark has changed. Under the Trademarks Act, a person may apply to register a trademark for goods or services if they use it or propose to use it and are entitled to use it in Canada, the goods or services must be stated in ordinary commercial terms, and they must be grouped by class of the Nice Classification with each group preceded by its class number (Trademarks Act, s. 30). A registration lasts ten years from the date of registration and is renewable for further periods of ten years (Trademarks Act, s. 46). The official fees are published: online filing in 2026 is $491.06 for the first class and $149.04 for each additional class, rising on January 1, 2027 to $499.41 and $151.57, and online renewal in 2026 is $595.06 for the first class and $185.49 for each additional class, rising to $605.18 and $188.64 in 2027; there is no separate registration fee for applications filed after June 17, 2019 (Canadian Intellectual Property Office, 2026a). A business that changes its name or its mark should expect to file again, because a registration covers the mark it was granted for and not a different one.

The third step concerns the rights in the new files. A refresh produces new artwork, new templates and a new guide, and the copyright in those works belongs to their author unless it has been assigned. Under the Copyright Act, the author of a work is its first owner, an assignment or grant of copyright is valid only if it is in writing and signed by the owner of the right, and moral rights — the integrity of the work and the right to be associated with it — cannot be assigned but can be waived, in whole or in part, and an assignment of copyright does not by itself waive them (Copyright Act, ss. 13(4) and 14.1). The practical consequence is that the contract with a design studio should assign the copyright in writing and should include a waiver of moral rights, so that the business can use, adapt and reprint its own identity without asking permission. In the case, the contract with the independent design studio did both.

The fourth step is the business register, and it applies when the legal name changes rather than when the logo changes. In Québec, a business name's generic term must be French, it is mandatory when the specific element is in another language, and where the generic term is the same in both languages it comes first, so that "TRANSPORT FASTER" is written rather than "FASTER TRANSPORT"; a business may declare versions of its name in other languages, and a trademark in another language may be declared as another name without a French generic only if it is a Canadian trademark registered with the office, with no French version, shown with its symbol (Registraire des entreprises, 2025). The guide is information rather than a legal interpretation, and the Registraire is not the authority on signs; the Office québécois de la langue française is.

The fifth step is the French language rules that apply to signs and products, and they are the ones most likely to affect a small business in Québec. On a sign visible from outside a premises, a recognized trademark wholly or partly in another language, and a business name with a non-French specific element, must be accompanied by French terms — a generic term, a description or a slogan — so that French is markedly predominant, which means a much greater visual impact, at least twice the space, permanent, at least as visible and legible; if a French version of the trademark is registered, it must be displayed (Office québécois de la langue française, 2025a). These rules are in force since June 1, 2025. A rebrand that produces a new depot sign is therefore also a compliance exercise, and the sign should be designed with the French requirement in mind rather than corrected afterwards.

The sixth step is the long list of records that carry the name, and it is longer than most businesses expect. The domain names and the accounts that hold them, the email addresses, the social handles and profile names, the listings on directories and maps, the bank and credit accounts, the supplier records, the insurance policies, the contracts with customers and the vehicle registrations all carry the name or the mark. Each one is a small task and none of them is difficult, but together they are the difference between a business that has changed its identity and one that has changed its logo. The checklist in section 8.5 is where they are tracked, and the brand keeper is who tracks them.

THE RULE The rights in the new identity

Under the Copyright Act, the author of a work is its first owner, and an assignment or grant of copyright is valid only if it is in writing and signed by the owner of the right (s. 13(4)). Moral rights — the integrity of the work and the right to be associated with it — cannot be assigned but can be waived, in whole or in part, and an assignment of copyright does not by itself waive them (s. 14.1). The rule matters to a rebrand because a rebrand produces new works: a logo, a set of templates, a guide, a website design. If the contract with the studio is silent, the business may hold a license to use the work and not the work itself, which becomes a problem the first time it wants to adapt the logo, reprint a template with a different supplier, or move the files to a new designer. The remedy is a clause, not a negotiation: the contract assigns the copyright in writing and includes a waiver of moral rights. This book explains the law and does not replace the advice of a lawyer or a trademark agent.

QUÉBEC & CANADA NOTE A new name means a new registration

In Québec, a change of business name is declared to the Registraire des entreprises, whose guide sets the rules for the generic and specific parts of a name: the generic term must be French, it is mandatory when the specific element is in another language, and where the generic is the same in both languages it comes first (Registraire des entreprises, 2025). In Canada, a new trademark needs its own application, because a registration covers the mark it was granted for; an application states the goods or services in ordinary commercial terms and groups them by class of the Nice Classification, and a registration lasts ten years and is renewable (Trademarks Act, ss. 30 and 46). The two procedures are separate, they are handled by different authorities, and a business that changes its name will usually have to do both. The guide published by the Registraire is information rather than a legal interpretation, and the Registraire is not the authority on signs; the Office québécois de la langue française is. This book explains the law and does not replace the advice of a lawyer or a trademark agent.

The order of the legal steps is not arbitrary. The search comes before the design, because a name that cannot be used is a design wasted. The application comes before the announcement where the mark has changed, because a public launch creates use that a later filing has to account for. The assignment comes with the contract, because a studio that has been paid and has moved on is harder to reach than a clause signed at the start. And the records are updated in the first phase of the migration, because a customer who writes to an old address is a customer lost for a reason that had nothing to do with the brand.

8.8 The Case's Refresh

The case followed through this book chose an evolution, and the reasoning is worth following because it shows the sorting rule in action. Six months after the refresh of its offer, Nordec Entretien inc. had 52 buildings under contract, revenue of $37,251 a month and a run-rate of $447,012 a year. Its identity had not moved since 2022: a blue wrench inside a circle with "NORDEC" in capitals, drawn by a friend on a template site. An inventory of its 36 touchpoints found three versions of the logo in circulation, five different blues, four typefaces, and only 13 of 36 touchpoints consistent with one another. The new tiers and the fixed-price assessment had no names in French, no templates and no identity.

The business considered a new name and rejected it, and the figures behind that decision are the clearest argument in the chapter. In the six months since the refresh of the offer, 15 proposals were decided and 8 signed. Asked how they first heard of Nordec, prospects answered: a referral by a customer 6, another building of a current manager 3, the Building Check letter 3, a web search 2, a Nordec van 1. Five of the 15 knew Nordec before the first contact. The name was therefore doing real work: it was carried by referrals, it was on the assessment letter that produced three proposals, and it was recognized by a third of the prospects before they spoke to anybody. A new name would have discarded that recognition and required a new trademark application, at least the $491.06 filing fee of 2026 and the agent's fees again. The business kept the name and changed what was accidental.

What it kept was three things: the name Nordec, the blue family, and the letter N. What it dropped was the generic wrench, four of the five blues, and the four typefaces, which were replaced by two open-license faces — a humanist sans-serif for everything and a monospaced face for building record codes. What it added was a horizon line drawn through the letter N, the building seen a year ahead; a second color, Signal Amber #E8A33D, for accents, the van stripe and safety details; the tagline in French first, « Votre bâtiment, un an d'avance. », with "Your building, a year ahead." in English; and French names for the tiers, Essentiel, Préventif and Portefeuille, with the English names alongside.

The palette was chosen with the contrast thresholds in hand, and the numbers are worth stating because they are the reason amber never carries text on white. White text on Nordec Blue #1F3A5F measures 11.48:1; Nordec Blue text on Mist #F4F6F8 measures 10.60:1; Slate #5B6770 text on white measures 5.80:1; Nordec Blue text on Signal Amber measures 5.32:1; and Signal Amber text on white measures 2.16:1, below the 4.5:1 that level AA requires for text (World Wide Web Consortium, 2024). Amber is therefore an accent color, not a text color, and the guide says so in one line. The design studio worked on a fixed fee of $7,800 for the logo, the palette, the typography, eleven templates and the guide, and presented three routes.

The test with committed customers is described in section 8.4, and its result shaped the choice. Sophie showed the three routes to six long-standing customers; four preferred the route that kept the letter N, and none asked to keep the wrench. The business also wrote to all 52 customers, in French and in English, on one page: what changes is the look; what does not change is the team, the contract and the 90-minute guarantee; the vans change over the following weeks; and the promise is stated once, in the same words used everywhere else. A line in the next monthly report repeated the announcement for anybody who had not read the letter.

The transition followed the order set out in section 8.5. Documents and digital came first, because they are produced in-house and seen constantly: the visit sheet, the one-page monthly report, the capital planning note, the consolidated portfolio report, the proposal, the invoice, the Bilan Nordec report, the letter, the email signature, the slide and the social media post, eleven templates in all, plus the website update and the business profile. Uniforms and vans came next: seven people, five garments each at $42, and two vehicle wraps at $4,100 each, the second scheduled a month after the first. The depot sign came last, at $1,900. The whole rollout was budgeted at $24,120, which is 5.4% of the $447,012 run-rate, one time, plus 64 hours of the administrator's time at $36, or $2,304.

DO THE MATH Why Nordec kept its name

Illustrative figures; the case is fictional. Step 1 — Of the 15 proposals decided in the six months after the offer refresh, 3 came from the Building Check letter, which is signed Nordec: 3 ÷ 15 = 20%. Step 2 — Of the same 15 prospects, 5 knew Nordec before the first contact: 5 ÷ 15 = 33.3%. Step 3 — Six of the 15 came from a referral by a customer, and a referral carries the name: 6 ÷ 15 = 40%. Step 4 — A new name would have discarded that recognition and required a new trademark application, at least the $491.06 online filing fee for the first class in 2026, plus the agent's fees again, which were $1,200 in the case. Step 5 — The refresh kept the name, the blue family and the letter N, and changed what was accidental — the wrench, four of the five blues and the four typefaces — for a design fee of $7,800 and a rollout of $24,120, without asking a single customer to relearn who the business is.

The refresh is an evolution rather than a revolution, and the distinction is not academic. The business kept the assets its customers named, it changed the elements that carried no recognition, and it moved the touchpoints in an order that never left a customer facing a business they did not recognize. Six months later, the touchpoints consistent with one another had risen from 13 of 36 to 33 of 36. The book does not attribute the growth that followed to the brand; the offer, the assessment and referrals moved at the same time. What the refresh did was make the business look like one business instead of five, which is the modest and achievable objective of most rebrands.

Costly Mistakes

• Rebranding to fix a service problem. A new identity on a late technician is a new identity on a late technician, and a change in marketing aesthetics affects brand equity less than factors such as employees' behavior (Muzellec & Lambkin, 2006).
• Choosing a larger degree of change than the business needs. A revolution where an evolution would do asks every customer to relearn a name, requires a new trademark application and updates every record, for a problem that a corrected palette would have solved.
• Skipping the asset inventory. A business that redesigns before it knows what customers recognize is guessing, and the element it discards is often the one people used to find it.
• Testing the routes with the wrong people. Showing a redesign to a general audience produces the positive half of a split that runs the other way among the most committed customers (Walsh et al., 2010).
• Announcing to the public before telling customers. A customer who sees a new van and has not been told will wonder whether the business has been sold, and the answer arrives too late to prevent the question.
• Leaving the migration half done. A touchpoint that is not on the list will be discovered by a customer, and an old logo on somebody else's directory a year later is the visible sign of a change that was never finished.
• Assuming the contract with the studio transferred the rights. An assignment of copyright is valid only if it is in writing and signed by the owner of the right, and moral rights cannot be assigned but can be waived (Copyright Act, ss. 13(4) and 14.1).

Key Takeaways

• A rebrand is justified when the business has changed — through a merger or an acquisition, a new strategy or offer, a name that cannot be protected, a legal constraint, or an identity that no longer works on screens — and not when the owner is bored of the logo.
• Structural change, above all mergers and acquisitions, is the most common trigger in a study of 166 corporate rebrands, and a change in marketing aesthetics affects brand equity less than factors such as employees' behavior (Muzellec & Lambkin, 2006).
• The three degrees of change are the refresh, the evolution and the revolution, and the discipline is to choose the smallest degree that does the work, measured by what a customer would have to relearn.
• Distinctive brand assets are what people already recognize, and the inventory of those assets comes before the design brief; the governance phase of the identity process is where that work belongs (Wheeler & Meyerson, 2024; Romaniuk, 2018).
• Test a candidate identity with six to ten long-standing customers before choosing, because a logo redesigned from angular to more rounded lowered the evaluations of strongly committed consumers while weakly committed ones responded positively (Walsh et al., 2010).
• Plan the migration in phases — documents and digital, then uniforms and vans, then the depot sign, with printed stock closing on a fixed date — announce to customers before the public, and complete the legal steps: a trademark search and application where the mark changes, the assignment of copyright in writing, and the update of the business register.

Your Action Plan

1. Write down the reason for the change in one sentence. State what has changed in the business — the offer, the ownership, the law, the name — and the operational change that accompanies the identity work. Deliverable: one sentence naming the structural reason and the operational change. Verification: the sentence names something other than the look of the logo.
2. Build the inventory of distinctive assets. List every element a customer could use to recognize the business, and assess each one for distinctiveness, recognition, ownability and durability. Deliverable: a table of assets with a keep, drop or add verdict and a reason for each. Verification: every verdict is supported by something a customer said or did, not by an internal preference.
3. Test the candidate routes with six to ten long-standing customers. Show the current identity and the routes side by side, ask what they recognize and what they would miss, and note the objections of the most committed separately. Deliverable: a written note of what each customer recognized and which route they preferred. Verification: the assets the customers named are kept, or the reason for not keeping them is written down.
4. Write the transition plan with a date per item. List every touchpoint, assign it to a phase, name its owner, and fix the date after which nothing old goes out. Deliverable: a phased plan covering documents and digital, uniforms and vans, the depot sign and printed stock. Verification: every touchpoint has an owner and a date, and the stock count is taken before the new order is placed.
5. Send the announcement and complete the legal steps. Write the one-page letter with what changes, what does not, when the vans change and the promise stated once; send it to customers before the first new touchpoint appears in public; and complete the trademark search, the application where the mark changes, the written assignment of copyright and the update of the business register. Deliverable: a sent announcement and a completed legal checklist. Verification: every customer has received the letter, every person in the business can state the reason for the change in one sentence, and the assignment is signed.

In La Machine

A refresh touches the platform in four places, and the application copy read for this book is specific about what each one does (La Machine, 2026). Business → My Website (business/website, with business/website/builder and business/website/preview) has a Theme & Branding panel with a gallery of 28 themes, a brand name, a tagline, a logo, a favicon and a banner image, five color overrides and a heading font and a body font chosen from a catalog of 200; the configuration is kept in the browser, which is where the new logo, the new colors and the two typefaces are entered. Assets → Gallery (business/marketing/gallery) is a library of royalty-free stock images that records, for each image, its license, the license's address and the attribution text, plus a provenance form for the user's own images, which is where new photography is filed with its rights. Marketing → Templates (business/marketing/templates) stores message templates by channel with variables and a preview, and it holds no brand colors, logo or footer, so the templates are updated by hand. Social profile pictures and bios are not edited from La Machine; they are changed on each platform.

What the platform does not do belongs in the same breath, and one page in particular deserves a plain statement. Business → Branding (business/branding) offers a logo drop zone, a brand name, a tagline and a website address, a primary and a secondary color with ten preset swatches, a choice among five fonts, an email footer and three social links; it computes the contrast of the primary color and shows a pass or fail badge at the 4.5:1 threshold, with a live preview of a client invitation card and a client email. Its values are not saved: the "Save branding" button only shows a confirmation, and no other page uses what was entered. It is a place to try a palette and see its contrast, not a brand register, and it is not the record of a change. La Machine has no saved brand kit that other pages use, no brand portal or guidelines publisher, no digital asset management with versions, usage rights or expiry dates, and no brand consistency audit; the touchpoint list, the stock count and the checklist of accounts that a migration needs are kept outside it.

Self-Check

1. What are the good reasons to rebrand, and which structural change is the most common trigger in the study of 166 corporate rebrands?
2. What distinguishes a refresh, an evolution and a revolution, and how is the right degree chosen?
3. What is a distinctive brand asset, and why does the inventory come before the design brief?
4. What did the field experiment with 632 participants find about committed and weakly committed consumers, and what does it mean for the order of a rebrand?
5. What are the legal and administrative steps a change of name or logo requires in Canada and Québec?

Answers. 1. The good reasons are a merger or an acquisition, a new strategy or offer, a name that cannot be protected, a legal constraint, and an identity that no longer works on screens; structural change, above all mergers and acquisitions, is the most common trigger in the study of 166 rebranded companies (Muzellec & Lambkin, 2006). 2. A refresh corrects details and keeps the name and core assets; an evolution keeps the assets people already recognize and replaces the accidental ones; a revolution changes the name or the whole identity; the right degree is the smallest one that does the work, measured by what a customer would have to relearn. 3. A distinctive brand asset is an element a customer could use to recognize the business without being told its name, such as a logo, a color, a font, a character or a tagline, and such assets are a researched field (Romaniuk, 2018); the inventory comes first because the value of an asset is in the memory rather than in the drawing, and a business that redesigns before it knows what customers recognize is guessing. 4. When a logo was redesigned from angular to more rounded, strongly committed consumers judged it more negatively and their attitude to the brand fell, while weakly committed consumers responded positively (Walsh et al., 2010); the consequence is that the committed customers must be tested first, because a general audience shows only the positive half of the split. 5. A trademark search in the Canadian Trademarks Database, a new application where the name or mark changes under the Trademarks Act, the assignment of copyright in writing with a waiver of moral rights under the Copyright Act, the declaration of a new business name to the Registraire des entreprises in Québec, compliance with the French language rules for signs and products, and the update of domains, email addresses, social handles, directories, bank and supplier records.

Chapter References

Copyright Act, R.S.C. 1985, c. C-42, ss. 13 and 14.1 (Canada). Justice Laws Website, current to September 3, 2026 (last amended November 7, 2024) — read September 25, 2026.

Canadian Intellectual Property Office. (2026a). Fees for trademarks (page modified July 31, 2026). Retrieved September 25, 2026, from https://ised-isde.canada.ca/site/canadian-intellectual-property-office/en/trademarks/fees-trademarks

Canadian Intellectual Property Office. (2026b). Trademarks (page modified October 5, 2026). Retrieved October 29, 2026, from https://ised-isde.canada.ca/site/canadian-intellectual-property-office/en/trademarks

La Machine. (2026). Documentation — route reference inventory. Internal product documentation, read September 25, 2026, together with the code of the pages this book describes: business/website, business/marketing/gallery, business/marketing/templates and business/branding.

NBC News. (2010, October 11). Gap scraps new logo after taking flak online [msnbc.com staff and news service reports]. https://www.nbcnews.com/id/wbna39622202

Muzellec, L., & Lambkin, M. (2006). Corporate rebranding: Destroying, transferring or creating brand equity? European Journal of Marketing, 40(7/8), 803–824. https://doi.org/10.1108/03090560610670007

Office québécois de la langue française. (2025a). Affichage des marques de commerce et des noms d'entreprises (updated August 27, 2025). Retrieved September 25, 2026, from https://www.oqlf.gouv.qc.ca/francisation/entreprises/affichage-marques-noms.html

Registraire des entreprises. (2025). Les noms d'entreprises au Québec. Gouvernement du Québec. Retrieved September 25, 2026, from https://cdn-contenu.quebec.ca/cdn-contenu/adm/min/emploi-solidarite-sociale/registraire_entreprises/IN-531_REQ_noms_ENTR_QC.pdf

Romaniuk, J. (2018). Building distinctive brand assets. Oxford University Press.

Trademarks Act, R.S.C. 1985, c. T-13, ss. 30 and 46 (Canada). Justice Laws Website, current to September 3, 2026 (last amended April 1, 2025) — read September 25, 2026.

Walsh, M. F., Winterich, K. P., & Mittal, V. (2010). Do logo redesigns help or hurt your brand? The role of brand commitment. Journal of Product & Brand Management, 19(2), 76–84. https://doi.org/10.1108/10610421011033421

Wheeler, A., & Meyerson, R. (2024). Designing brand identity: A comprehensive guide to the world of brands and branding (6th ed.). Wiley.

World Wide Web Consortium. (2024). Web Content Accessibility Guidelines (WCAG) 2.2. W3C Recommendation, 12 December 2024 (first published as a Recommendation on 5 October 2023). Retrieved September 25, 2026, from https://www.w3.org/TR/WCAG22/

Chapter 9 — Measuring Brand Health

Brand health is measured with a few indicators — awareness, associations, loyalty and advocacy, consistency — tracked the same way over time and read with the humility that small samples demand.

What You'll Be Able to Do

• Define brand health as the state of what people know and feel about a business, and separate the part of it a small business can measure from the part it cannot.
• Distinguish unprompted awareness, prompted awareness and salience, and explain why a business that is known is not always a business that comes to mind when a building needs work.
• Run a sixty-call awareness survey by telephone, with the unprompted question asked before the prompted one and the words people use recorded as they are spoken.
• Compute the margin of error of a proportion measured on a small sample, and read two measurements against each other before claiming that anything has moved.
• Build a brand scorecard of a handful of measures — awareness, associations, loyalty, advocacy and consistency — and repeat it on the same dates with the same questions.
• Explain why monetary brand valuation and the global brand rankings concern large businesses, and why a small business measures its brand for decisions rather than for a number.

Why It Matters

A brand is the one asset of a small business that nobody can see on a balance sheet, and that is exactly why it is so easy to leave unmeasured. The vans are counted, the contracts are counted, the invoices are counted, and the promise that holds them together is discussed only when something goes wrong. The result is a business that can say precisely how many buildings it serves and cannot say whether the owners of the buildings it does not serve have ever heard its name. Those two facts belong to the same business, and only one of them is usually written down.

The measurement matters because the brand is what makes the other numbers move, and because a business that does not measure it will attribute every change to the last thing it did. When a small business refreshes its identity, the months that follow bring new contracts, more calls and a fuller calendar, and the temptation is to credit the design. The honest reading is harder and more useful: the offer changed, the entry product changed, referrals continued, the season turned, and the brand work was one of several things happening at once. A scorecard that was taken before the work began is the only thing that lets a business tell those apart, and it is the only defense against a story that flatters the work and teaches nothing.

There is also a plain commercial reason to measure, and it belongs to the way a small business is bought and sold. An international standard defines a brand as a marketing-related intangible asset and sets requirements for measuring its monetary value (International Organization for Standardization, 2010); a business needs such a figure only at particular moments. Those moments are rare for a business of seven people, and the discipline that produces them is not. The same habit — asking the same questions of the same kind of person at the same interval — is what tells an owner whether the promise on the wall is reaching the street, and it costs a few hours a year rather than a study.

In This Chapter

Section 9.1 defines brand health and sets out what a small business can and cannot measure. Section 9.2 separates unprompted awareness, prompted awareness and salience, and describes how Nordec asked. Section 9.3 covers associations and perception, from open questions to the five dimensions of brand personality. Section 9.4 covers preference, loyalty and advocacy, including the single recommendation question and the reviews customers leave in public. Section 9.5 covers consistency and distinctiveness, and the audit that turns both into a number. Section 9.6 places the two international standards and the global rankings in their proper scale, and warns against very short-term metrics. Section 9.7 explains the margin of error and the honest reading of a small sample. Section 9.8 presents Nordec's brand scorecard before and six months after the rollout.

9.1 What Brand Health Means

Brand health is the state of what people know and feel about a business, and it is a description of memory rather than of opinion. The useful way to think about it comes from the idea of customer-based brand equity, which is the differential effect that brand knowledge has on how customers respond to the marketing of a brand; brand knowledge itself has two components, brand awareness and brand image, and equity is positive when customers respond more favorably to the branded offer than they would to an unbranded one (Keller, 1993). Read plainly, that sentence says three things a small business can act on. People must know the name. What they know must be favorable, strong and unique. And the test of both is the response, not the intention.

The second component is where most small businesses are thinner than they think. Awareness is the easy half, because a van, a sign and a referral can produce it. Image is the hard half, because it is made of the associations that come to mind when the name is heard, and those associations are built by everything the business does rather than by anything it says about itself. A business can be known by every property manager in its territory and still be known as the firm that arrives late, and the second fact will decide the next contract. Measuring brand health therefore means measuring both halves, and measuring them separately, because a business that asks only whether people have heard of it will learn only the half that is already working.

The model that gives the measurement its shape is the one built around four steps of brand building — identity, meaning, responses, relationships — and six building blocks that run from salience at the base through performance, imagery, judgments and feelings to resonance at the top, with resonance itself described in four kinds: behavioral loyalty, attitudinal attachment, a sense of community and active engagement (Keller, 2001). The model is offered as a yardstick for brand tracking rather than as a validated instrument, and that is precisely how a small business should use it. It is a ladder of questions. Is the brand the first thing that comes to mind? Does the offer perform? What images and feelings does the name carry? Do people come back, and do they speak about the business without being asked? A scorecard that answers a few of those questions in the same way each year is worth more than a study that answers all of them once.

What a small business can measure is narrower than what a large one measures, and saying so is part of the discipline. A business of seven people can count the buildings it serves, the contracts that renew, the proposals that are signed, the referrals that arrive and the reviews that are left in public. It can telephone sixty owners of buildings it does not serve and ask three questions. It can read the words its customers use and compare them with the words on its own wall. It can audit its own materials against its own guide. What it cannot do is measure the brand's contribution to revenue in isolation, because the offer, the price, the season and the referrals all move at the same time, and no small sample can separate them. The honest position is that the scorecard describes direction and consistency, and that a single movement in it is a question rather than a conclusion.

The case followed through this book is a good illustration of the difference between a measurement and a proof. Nordec Entretien inc. is a fictional business created for this encyclopedia, and its scorecard was taken twice, before the rollout and six months after it. Prompted awareness among owners of buildings not under contract in its two territories rose from 13 of 60 to 19 of 60. The number is real, the calls were made, and the rise cannot be distinguished from chance at that sample size, which is a fact the business chose to write down beside the figure rather than to leave out. That choice is the whole of this chapter in miniature: a small business measures its brand to make better decisions, and the first better decision is to read its own numbers honestly.

DEFINITION Prompted awareness, unprompted awareness, salience and net promoter score

Unprompted awareness is the share of people who name a business when asked an open question about a category, with no list and no hint; it measures what comes to mind on its own. Prompted awareness is the share who say they have heard of a business once its name is read to them; it measures recognition rather than recall, and it is normally the higher of the two. Salience is the propensity of a brand to be noticed or to come to mind in buying situations, and it reflects the quantity and quality of the associations held in memory rather than the simple fact of knowing the name (Romaniuk & Sharp, 2004). The net promoter score is the common convention built on one question — how likely a customer is to recommend the business, on a scale from 0 to 10 — where answers of 9 or 10 are counted as promoters, 7 or 8 as passives and 0 to 6 as detractors, and the score is the percentage of promoters minus the percentage of detractors.

KEY CONCEPT Measure the same way every time

A brand scorecard is worth having only if its measures are comparable across time, and comparability is produced by repetition rather than by sophistication. The same questions, in the same order, asked of the same kind of person, drawn from the same kind of list, at the same interval, by the same method. Change one of those and the movement in the result belongs to the change rather than to the brand. A business that asks a prompted question one year and an unprompted one the next has learned nothing about awareness, because the two questions measure different things and the second is always smaller. A business that telephones sixty owners in one year and surveys two hundred online the next has learned nothing either, because the two populations answer differently and the response rate differs again. The practice is unglamorous and it is the whole of the discipline: write the questions down, write the list criteria down, write the date down, and do not improve the instrument between two readings. Improvement is welcome, and it belongs after the second reading, not between the two.

9.2 Awareness and Salience

Awareness is the first thing a small business measures and the easiest to measure badly. The bad version is a single question — have you heard of us? — asked of people who already know the answer, and it produces a number that rises every year and means nothing. The good version separates two questions that are genuinely different, asks the harder one first, and asks both of people who have no reason to be kind. The harder question is the unprompted one, and it is the one that tells a business whether its name is present in a market or merely familiar to the people it already serves.

The distinction between the two questions is not a technicality. Prompted awareness measures recognition: a person hears a name and confirms that it is familiar, which is a low bar that a van, a sign and a single conversation can clear. Unprompted awareness measures recall: a person is asked which companies they know that do a certain kind of work, and the name either arrives or it does not. The gap between the two figures is one of the most useful things a small business can learn about itself, because a wide gap says that the name is stored somewhere but is not connected to the moment of need. In the case followed through this book, the gap is wide in both readings: 13 of 60 and then 19 of 60 on the prompted question, against 3 and then 5 on the unprompted one. The business is recognized far more often than it is remembered, and that is a finding a scorecard can act on.

Salience is the concept that explains the gap, and it is worth stating precisely because it is often confused with awareness. Brand salience is the propensity of a brand to be noticed or to come to mind in buying situations, and it reflects the quantity and quality of the associations a person holds in memory; it is distinct from awareness and distinct from attitude (Romaniuk & Sharp, 2004). The distinction matters to a small business because the two call for different work. A business with low awareness needs to be seen: more vans, more signs, more presence in the places where its buyers look. A business with high prompted awareness and low salience needs to be linked: the name has to arrive attached to the situation, which is the work of consistent cues, repeated language and a promise that names the moment it answers. Nordec's promise — "We know your building, and we tell you ahead." — is a salience instrument as much as a promise, because it attaches the name to a situation an owner recognizes.

The way Nordec asked is worth setting out because it is reproducible by any small business with a telephone and a list. The administrator drew a list of owners of buildings not under contract in the two territories, in the size band the business serves, and called sixty of them. The first question was unprompted: "Which companies do you know that maintain commercial buildings on the north shore?" The second was prompted: "Have you heard of Nordec Entretien?" The third was open: "What comes to mind when you think of Nordec?" The order is deliberate, because asking the prompted question first would plant the name and destroy the value of the unprompted one. The calls were made by the administrator rather than by Sophie, because an owner who is known in the territory changes the answer simply by being on the line.

Two rules make the exercise honest. The first is that the list is drawn by a criterion rather than by convenience: every owner in the band, in the two territories, not under contract, taken in a fixed order, so that the sixty are not the sixty who are easiest to reach. The second is that the answers are recorded as they are spoken, including the ones that are unflattering, because the words are the material of section 9.3. A survey that records only the answers that confirm the brand is not a measurement; it is a press release with a sample size attached.

METHOD Running a sixty-call awareness survey

First, draw a list of prospects at random from a defined population — for Nordec, owners of buildings of 2,000 to 9,000 square metres, in the two territories, not under contract — and write the selection criteria down so that the next survey draws from the same population in the same way. Second, ask the unprompted question before the prompted one, always, because the prompted question plants the name and makes the unprompted answer worthless if it comes first. Third, record the words people use in answer to the open question, verbatim, including the ones the business would rather not hear, because those words are the raw material of the association work. Fourth, repeat the same questions, with the same list criteria and the same procedure, a year later, and resist the urge to improve the questionnaire in between. Fifth, compute the margins of error before reading any change, and state the intervals beside the figures so that the reader sees what the sample can and cannot support.

QUÉBEC & CANADA NOTE Calling businesses for a survey

A survey of businesses is not the same exercise as a survey of consumers, and the difference is worth stating plainly in a Québec and Canadian context. Nordec calls businesses, not individuals at home, and it asks no sales question: the call is a measurement, and the person on the other end is being asked for their time rather than offered a service. A survey call must not become a sales pitch, and the answers must be kept for the purpose the survey states and no other. How contact information is collected, kept and used is the subject of Book H2 (Privacy and Personal Information), and the requirements that govern telemarketing and commercial electronic messages are the subject of Book H3 (Anti-Spam and Telemarketing Compliance); this chapter states only the two habits that keep a survey a survey. The book explains these matters and does not replace the advice of a lawyer or a privacy professional.

9.3 Associations and Perception

The second half of brand knowledge is image, and image is made of associations: the words, pictures and feelings that arrive when a name is heard. A small business cannot choose its associations directly, because they are built by experience, but it can find out what they are and it can decide whether they match the platform it wrote in Chapter 2. The instrument is an open question, and the discipline is to record the answers without translating them into the business's own vocabulary.

The case's twenty value interviews, described in Book A6 (Designing a Value-Added Offer), produced the words customers used about Nordec: "on time" or "fast" from 15 of the 20, "knows my building" from 11, "tells me ahead" from 9, "straight with me" from 7 and "friendly" from 5. Set beside the platform of Chapter 2, those words are close to the promise and closer to the values than the business had assumed. "Tells me ahead" is the promise in the customer's own words. "On time" is the value Punctual, spoken by the people it was written for. "Knows my building" is the building health record doing its work in a conversation nobody designed. The comparison is not a score; it is a check that the words on the wall and the words in the market are the same words, and it is the cheapest brand research a small business will ever do, because the interviews had already been conducted for another purpose.

The same open question, asked of people who are not customers, produces a different and equally useful picture. An owner who has never bought from Nordec and has heard the name may answer with a category word ("maintenance"), a place word ("the vans"), or nothing at all. A high proportion of category words is not a failure; it means the name is filed under the right heading. A high proportion of nothing is a salience problem rather than an awareness problem, and it points to the work described in section 9.2. The rule for reading the answers is to count them without arguing with them, and to resist the temptation to explain an unflattering word as a misunderstanding. The customer's association is the brand, whatever the guide says.

Personality is the part of perception that can be measured against a published instrument rather than against intuition. The dimensions of brand personality — sincerity, excitement, competence, sophistication and ruggedness — were established with a measurement scale reported as reliable, valid and generalizable (J. L. Aaker, 1997). The scale was developed in English and this book does not claim that it applies unchanged in Québec, so a small business should use the five dimensions as a vocabulary rather than as a test with a pass mark. The vocabulary is still useful. Nordec's four traits — calm, precise, plain-spoken, neighborly — sit close to sincerity and competence, and a survey that asks customers to place the business on those dimensions will show whether the traits the business chose are the traits the market perceives. Where the two diverge, the divergence is information: a business that believes it is precise and is described as rushed has a delivery problem, not a design problem.

The comparison between the platform and the market is the reason this section sits between awareness and loyalty. Awareness without the right associations produces recognition that does not help, and associations that contradict the promise produce a customer who arrives expecting one thing and receives another. The check is annual, it costs a few hours, and it produces the only evidence a small business will ever have that its identity work is landing. In the case, the check produced a short list of words to keep using and a short list to stop using, and the glossary of twelve words in Chapter 3 was revised once as a result.

WHO DOES THIS The brand keeper measures, the owner reads

In the case followed through this book, the administrator is the brand keeper, and the measurement work sits with them for the same reason the templates do: they already hold the customer list, the review pages and the monthly report, and they are the person who notices when a word changes. The brand keeper draws the survey list, makes the calls, records the answers verbatim, keeps the scorecard in one place and updates it on the same dates each year. The owner does not make the calls, because an owner on the line changes the answer, and does not compute the margins of error, because that is arithmetic rather than judgment. What the owner does is read the scorecard once a year, in the same sitting as the annual brand review of Chapter 10, and decide what the business will do about what it shows. The split matters because it protects both halves of the exercise: the person collecting the answers has no stake in what they say, and the person reading them has the authority to act.

9.4 Preference, Loyalty and Advocacy

The measures that matter most to a small business are the ones that show up in behavior, and behavior is easier to count than attitude. A contract that renews is a preference expressed in money. A customer who names the business to another owner is advocacy expressed in a referral, and referrals are the channel that brought six of the fifteen proposals Nordec decided in the six months after its refresh. Neither of those events needs a survey, and both belong on the scorecard, because they are the outcomes the softer measures are supposed to predict.

The single recommendation question is the best-known attempt to turn advocacy into one number. The article that proposed it offers a single question — how likely a customer is to recommend the company, on a scale from 0 to 10 — as a loyalty measure that, in the author's claim, is linked to growth (Reichheld, 2003). The claim is the author's, and this book presents it as such. What a small business should take from the article is the discipline of asking one question consistently rather than the promise of a growth formula. The cut-offs that turn the answers into a score are a common convention rather than part of the article: answers of 9 or 10 are counted as promoters, 7 or 8 as passives, and 0 to 6 as detractors, and the score is the percentage of promoters minus the percentage of detractors. Stated as a convention, it is a usable measure. Stated as a law of growth, it would be a claim the source does not support.

The case's reading is a good example of both the value and the limit of the instrument. The survey was answered by 41 of 52 customers, which is a high response for a business that asked by telephone and email. Of those 41, 27 were promoters, 10 were passives and 4 were detractors, which gives a score of 56.1. The number is worth having, and it is worth exactly what one measurement is worth: nothing can be said about whether it rose or fell, because there is no earlier value to compare it with. The four detractors are the more useful part of the result, because four customers out of forty-one is a small enough group to telephone individually and a large enough group to matter in a business of 52 contracts. A scorecard that produces a number and no names has done half the work.

Public reviews are the third behavioral measure, and they have the advantage of being visible to everyone including the next prospect. Nordec's count moved from 23 reviews with an average of 4.7 to 31 reviews with an average of 4.8 over the six months. The count is the measure that matters more than the average, because an average built on 23 reviews moves slowly and an average built on 31 is not meaningfully different from it. What the count shows is that more customers are willing to speak in public, which is a form of advocacy that costs the business nothing and reaches people the business cannot call. The average is worth watching for a fall rather than for a rise: a drop from 4.7 to 4.2 would be a signal, and a rise from 4.7 to 4.8 is noise.

Renewals and referrals complete the picture, and they are the measures a business should trust most because they cost the customer something. A renewal costs the customer the decision to continue paying. A referral costs the customer a small piece of their own reputation with another owner. Both are harder to produce than a favorable answer on a survey, and both are counted from records the business already keeps. The scorecard of section 9.8 therefore places them beside the survey measures rather than beneath them, because a business whose recommendation score is high and whose renewals are falling has learned something the survey could not tell it.

TOOLS OF THE TRADE What a brand scorecard needs

Five functions carry the whole measurement, and none of them is a specialist product. The first is a defined list of the people to be surveyed, drawn in a fixed way from a population the business can describe, so that the next survey draws from the same population. The second is a form or a script that holds the questions in their fixed order, with the unprompted question first, and a place to record answers verbatim rather than in categories. The third is a spreadsheet or a table that holds the scorecard itself, with one row per measure, one column per reading and the date of each reading at the top. The fourth is the review count and rating shown on the business's public profile, read on the same date as the rest. The fifth is a calendar entry for the next reading, placed a year ahead on the day the last one was taken, because a measurement that is not scheduled is a measurement that will be taken when somebody remembers.

9.5 Consistency and Distinctiveness

Consistency is the measure a small business controls most directly, and it is therefore the one that should be on the scorecard first. The consistency audit described in Chapter 7 counts the touchpoints a business has and checks each one against the guide: the same logo, the same blue, the same typefaces, the same words, the same promise. It produces a fraction, and a fraction is a health measure because it can be taken again. Nordec's audit found 13 of 36 touchpoints consistent with one another before the rollout, which is 36.1%, and 33 of 36 six months after it. The second figure is the strongest result in the whole scorecard, and it is strong for a reason that has nothing to do with statistics: it is a count of objects rather than a sample of people, so there is no margin of error to argue about.

The reason consistency deserves the first place is that it is the precondition of everything else. A business whose logo appears in three versions and whose blue appears in five shades is teaching its market that the details do not matter, and it is spending the recognition it has earned on re-teaching the same name. Distinctive brand assets — logos, colors, fonts, characters, taglines and jingles — are a researched field, and the work in it examines which branding elements succeed and which fail over the long term (Romaniuk, 2018). The practical question the field asks is not whether an asset is attractive but whether people link it to the brand, and that question can be asked of a small business's own customers with a picture and a single question. A color that customers cannot name as Nordec's is not yet an asset, whatever the guide says.

The audit and the asset question are two halves of the same measure. The audit asks whether the business is using its assets consistently; the asset question asks whether the market has learned them. A business can score 33 of 36 on the first and still fail the second, because consistency applied for six months is not the same as recognition built over years. That is why the case's scorecard places the consistency fraction beside the awareness figures rather than above them, and why the honest reading of the whole table is that the business has put its house in order and has not yet been rewarded for it in the market.

Redesigns are where consistency and distinctiveness meet, and where the measurement has something uncomfortable to say. A field experiment with 632 participants on two sportswear brands' logos found that when a logo was redesigned from angular to more rounded, strongly committed consumers judged it more negatively and their attitude to the brand fell, while weakly committed consumers responded positively (Walsh et al., 2010). The finding does not say that redesigns are wrong; it says that the people who care most about a brand are often the people most attached to what it looks like now, and that a business should expect the reaction to come from its best customers first. Nordec's refresh was an evolution rather than a revolution, keeping the name, the blue family and the letter N, and the choice was made partly with that finding in mind. The scorecard cannot prove the choice was right, and it can show that the customers who were most attached had nothing taken from them.

The measure that ties the section together is the count of touchpoints that carry the promise rather than the logo. A business can be perfectly consistent and still say nothing, because consistency is a property of execution and distinctiveness is a property of meaning. The audit in Chapter 7 counts both: the touchpoints that match the guide, and the touchpoints that carry a distinctive element a customer could name. Nordec's 33 of 36 counts the first; the second is checked once a year by asking a handful of customers to name the color, the shape and the line, and by writing down how many can.

9.6 Brand Valuation and Standards

There is a formal world of brand measurement that a small business should know exists and should not confuse with its own scorecard. The first piece of it is the standard that sets requirements for the procedures and methods of measuring the monetary value of a brand, covering objectives, bases, approaches, data and reporting, and defining a brand as a marketing-related intangible asset — names, terms, signs, symbols, logos and designs — that identifies goods, services or entities and creates distinctive images and associations among stakeholders; a value produced under it may be a single figure or a range (International Organization for Standardization, 2010). The second piece is the framework for non-monetary brand evaluation, first published in 2019 and replaced by a first edition in November 2021, which sets out input elements, output dimensions and sample indicators and can be used for internal or external evaluation (International Organization for Standardization, 2021). Neither is a marketing exercise. Both are instruments for moments when a brand has to be described to somebody outside the business.

Those moments are rare for a business of seven people, and they are almost always the same moment: a sale, a financing, a dispute or a succession. A monetary valuation of a small service brand is a professional engagement, and it belongs with the advisers who handle the transaction rather than with the owner's annual routine; Book M1 (Preparing Your Business for Sale) treats the preparation of a business for sale, and the valuation of its brand sits inside that work. What a small business can take from the standard without commissioning anything is its definition, because the definition is a checklist in disguise. Names, signs, symbols, logos and designs; distinctive images and associations among stakeholders. A business that can point to each of those and say where it is written down, who owns it and how it is used is a business whose brand could be valued if it ever had to be.

The global rankings belong to a different scale entirely, and reading them as a benchmark would be a mistake. One ranking combines financial analysis with brand-equity measures taken from consumers, and its 2026 study covered 4.6 million respondents and 22,000 brands, with the top 100 valued at US$13.1 trillion (Kantar, 2026). The other values brands on financial performance, the role of the brand in driving choice and loyalty, and a brand strength score, and its eligibility rules require among other things a substantial share of revenue outside the home region, public financial data and a brand strength score of 50 or more, so that such rankings concern large global brands only (Interbrand, 2025). Both are serious work and neither says anything about a maintenance business with 52 contracts. The useful thing a small business can take from them is the shape of the exercise: financial performance, the role of the brand in choice, and a measure of strength. That is a three-part structure a small business can imitate with its own figures, at its own scale, without pretending to a valuation.

The warning that belongs at the end of this section is about time horizons. The work on balancing long-term brand building against short-term activation examines the tension between the two and warns against using very short-term metrics as primary measures of performance (Binet & Field, 2013). A small business is unusually exposed to that warning, because its short-term metrics are excellent and immediate: calls answered, invoices sent, contracts signed this month. A brand scorecard that is read every month would be read against those numbers and would always lose. The remedy is the cadence rather than the metric: the scorecard is read once a year, in the same sitting as the annual brand review of Chapter 10, and the monthly numbers are read for what they are.

The following table sets out seven ways a brand can be measured, what each one shows, how a small business can perform it and what its limit is. Read it as a menu rather than a list of obligations: a business of seven people can carry four or five of these measures comfortably, and the ones it chooses matter less than the fact that it repeats them.

Table 9.1 — Ways to measure a brand

THE RULE Measuring monetary brand value

The standard that governs the monetary valuation of a brand is ISO 10668:2010, Brand valuation — Requirements for monetary brand valuation, first edition September 1, 2010, which sets requirements for the procedures and methods of measuring the monetary value of a brand, covering objectives, bases, approaches, data and reporting, and which defines a brand as a marketing-related intangible asset — names, terms, signs, symbols, logos and designs — that identifies goods, services or entities and creates distinctive images and associations among stakeholders (International Organization for Standardization, 2010). The rule for a small business is not that it must value its brand. The rule is that when a brand has to be valued — for a sale, a financing or a dispute — the valuation is a professional engagement conducted under a published standard, and the figure it produces may be a single amount or a range. A small business rarely needs this outside the sale of the business, and Book M1 (Preparing Your Business for Sale) treats that moment. The book explains the standard and does not replace the advice of a qualified valuation professional.

WHAT CHANGED A standard for brand evaluation, 2019 and 2021

The framework for evaluating a brand without putting a monetary figure on it was first published as ISO 20671 in 2019, and it was replaced by a first edition in November 2021 under the title Brand evaluation — Part 1: Principles and fundamentals (International Organization for Standardization, 2021). The change matters to a small business for one reason: the framework separates input elements from output dimensions and offers sample indicators, which is the same separation a small business makes when it distinguishes what it does — the promise, the guide, the templates, the training — from what comes back — awareness, associations, renewals, referrals. The framework is usable for internal evaluation as well as external, which means a business can adopt its logic without commissioning a study or hiring an assessor. What the framework does not do is supply numbers, thresholds or a pass mark, and this book does not list its elements or dimensions. Read as a structure rather than as a score, it is the closest thing in the standards world to the scorecard this chapter describes.

9.7 Small Samples and Honest Reading

The arithmetic of a small sample is the part of measurement that a small business is most tempted to skip, and it is the part that decides whether a scorecard is useful or misleading. The margin of error at 95% for a proportion p measured on n answers is 1.96 multiplied by the square root of p times (1 − p) divided by n. The formula is not an opinion and it does not depend on the subject being measured; it says how much a proportion can be expected to move from one sample to another simply because a different sixty people were asked. A business that computes it before reading a change will avoid the most common error in small-business brand measurement, which is treating a rise of a few points as a result.

The case's own numbers show what the formula does. Before the rollout, 13 of 60 owners named Nordec when prompted, which is 21.7%, and the margin of error is about ±10.4 points, so the interval runs from roughly 11.3% to 32.1%. Six months later, 19 of 60 named it, which is 31.7%, with a margin of about ±11.8 points, giving an interval from roughly 19.9% to 43.5%. The two intervals overlap substantially, which means that a rise of ten points measured on sixty calls each time cannot be distinguished from chance. The rise may be real. The measurement does not show that it is, and the honest thing to write beside the figure is exactly that.

The same caution applies to the smaller counts, and more sharply. Unprompted mentions moved from 3 to 5 out of 60, which is a change of two people. Two people is within the range of ordinary variation in who happens to answer the telephone, and no interval computed on five mentions will be narrow enough to support a claim. The count of prospects who knew Nordec before the first contact moved from 5 of 15 to 8 of 14, which is a change of three people on a base of fifteen, and the base is the number of proposals decided rather than a sample drawn for the purpose. It is a useful operational figure and it is not evidence of a shift in the market.

One measurement is not a trend, and this is the principle that governs the whole chapter. A trend requires at least two readings taken the same way, and a defensible trend requires more than two, because two readings can differ for reasons that have nothing to do with the brand: the season, the economy, a competitor's departure, a change in the list. The case has two readings of most measures and one reading of the recommendation score, and the correct description of the recommendation score is that it is a baseline rather than a result. A business that writes "56.1, first measurement" beside the figure has a number it can compare against next year. A business that writes "56.1, strong" has a compliment.

The last discipline is the hardest, and it concerns attribution. Nordec's buildings under contract moved from 52 to 56 over the six months, and the temptation is to credit the brand refresh. The honest reading is that the offer, the Bilan Nordec entry product and referrals all moved at the same time, and that the brand work cannot be separated from them. The evidence for the brand's contribution is indirect and worth stating as such: the consistency fraction moved from 13 of 36 to 33 of 36, which is a fact about the business's own materials; the review count rose, which is a fact about customers' willingness to speak; and the awareness figures rose by an amount the sample cannot confirm. Three facts of different strength, pointing the same way, and none of them proving that the design sold a contract.

WATCH OUT Reading a rise that the sample cannot show

The most common error in small-business brand measurement is to read a rise of a few points as a result, and it is common because the arithmetic is invisible. A business asks sixty people one year and sixty people the next, sees the number go from 21.7% to 31.7%, and writes in its plan that awareness rose by ten points. What the sample actually supports is a statement about two intervals that overlap, which is to say that the two readings are consistent with no change at all. The error is expensive in a specific way: it produces confidence in work that may not be working, and it removes the pressure to keep measuring. The remedy is a habit rather than a statistic. Before reading any change, compute the margin of error for each reading and write both intervals beside the figures. If the intervals overlap, describe the movement as a rise the sample cannot confirm, and say what would be needed to confirm it — a larger sample, or another year of the same measurement. A scorecard that states its own limits is more useful than one that does not, because it can be trusted where it is strong.

DO THE MATH The margin of error of sixty calls

Illustrative figures; the case is fictional. Step 1 — Before the rollout, 13 of 60 owners named Nordec when prompted, so p = 13 ÷ 60 = 0.217, or 21.7%. Step 2 — The margin of error is 1.96 × √(0.217 × 0.783 ÷ 60) = 1.96 × 0.0532 = 0.104, about ±10.4 points, giving an interval from about 11.3% to 32.1%. Step 3 — Six months later, 19 of 60 named it, so p = 19 ÷ 60 = 0.317, or 31.7%, and the margin is 1.96 × √(0.317 × 0.683 ÷ 60) = 1.96 × 0.0601 = 0.118, about ±11.8 points, giving an interval from about 19.9% to 43.5%. Step 4 — The two intervals overlap between 19.9% and 32.1%, so the difference between the readings is smaller than the uncertainty in each of them. Step 5 — The rise of about ten points cannot be told from chance with sixty calls each time; the honest statement is that awareness rose by an amount this sample cannot confirm, and that a larger sample or a third reading would be needed to confirm it.

9.8 The Case's Brand Scorecard

The scorecard of the case is short, and its shortness is deliberate. It holds seven measures, taken before the rollout and six months after it, by the same person using the same questions and the same list criteria. The measures are prompted awareness, unprompted mentions, the consistency fraction, the count of public reviews, the number of prospects who knew Nordec before the first contact, the recommendation score and the number of buildings under contract. Nothing on the list requires a specialist, and everything on it can be taken again next year by the administrator in a few days of work.

Figure 9.1 shows the scorecard as a picture rather than as a table. It draws prompted awareness as two bars, 21.7% and 31.7%, each with an error bar of about eleven points, and the two error bars overlap across most of their length; beside the bars, the other measures are listed with their before and after values. The purpose of the drawing is to make the overlap visible at a glance, so that a reader who sees only the picture cannot come away believing that the rise is established.

Figure 9.1 — Nordec's brand scorecard. Consistency moved beyond doubt; awareness rose, but with sixty calls each time the two error bars overlap, so the rise cannot be told from chance.

Read as a whole, the scorecard tells a story with one strong result and several weak ones. The consistency fraction moved from 13 of 36 to 33 of 36, which is a count of objects rather than a sample of people and therefore carries no margin of error. The review count rose from 23 to 31, which is a fact about customers' willingness to speak in public rather than a sample. The recommendation score of 56.1 is a baseline, measured once, with nothing to compare it against. The awareness figures rose by an amount the sample cannot confirm. The unprompted mentions rose from 3 to 5, which is two people. The prospects who knew Nordec before the first contact rose from 5 of 15 to 8 of 14, which is three people on a base of fifteen proposals.

The honest reading of the whole table is that the business put its own house in order and has not yet been rewarded for it in the market. That is a normal sequence and it is worth naming, because a business that expects the market to move in six months will conclude that the work failed. The consistency result is the one the business controls, and it is the one that moved decisively. The market measures are the ones the business does not control, and they moved in the right direction by amounts that a sample of sixty cannot confirm. The correct conclusion is not that the brand work did not work; it is that six months and sixty calls are not enough to know, and that the next reading, taken the same way, is what will begin to answer the question.

The scorecard also produced two findings that no measure on the list was designed to produce. The first is the gap between prompted and unprompted awareness, which is wide in both readings and points to a salience problem rather than an awareness problem: the name is recognized far more often than it is remembered. The second is the four detractors among the 41 customers who answered the recommendation question, which is a small enough group to telephone individually and a large enough group to matter in a business with 52 contracts. Neither finding would have appeared in a scorecard built only from financial figures, and both are actionable in the following quarter.

The following table sets out the seven measures with their values before and six months after, and with the honest reading of each. Read it as the model for a small business's own scorecard: the last column is the one that makes the table useful, because a figure without a reading is a number rather than a measurement.

Table 9.2 — Nordec's brand scorecard

CASE IN POINT Nordec Entretien inc.: the survey that did not prove the rise

Nordec Entretien inc. is a fictional business created for this encyclopedia; its owner, its employees, its customers, its competitors, its figures and the events described are invented, and any resemblance to a real business is coincidental. The administrator made sixty calls before the rollout and sixty calls six months later, from the same kind of list, with the same three questions in the same order. The prompted figure rose from 13 to 19, and the first draft of the internal summary described that as a rise of ten points. Sophie read it and asked what the margin of error was. The administrator computed it: about ±10.4 points before and about ±11.8 after, with the two intervals overlapping. The summary was rewritten to say that awareness rose by an amount the sample could not confirm, and the sentence was kept in the document that went to the seven people. The rewritten version is the more useful one, because it told the business what to do next: keep the measurement, keep the same questions, and take a third reading before drawing a conclusion.

IN LA MACHINE A form for the survey, nothing more

The application copy read for this book offers one page that serves this chapter directly (La Machine, 2026). Forms, at business/marketing/forms, can carry a short survey — the likelihood-to-recommend question, or the awareness questions of section 9.2 — and the answers are kept with the form, at business/marketing/forms/responses, where they can be read and counted. That is the whole of what the platform does for brand measurement, and it is worth saying what it does not do. There is no brand health tracking, no awareness survey tool and no net promoter tool beyond the general form builder; there is no brand consistency audit and no way to compare a scorecard across two readings. The scorecard itself therefore lives in a spreadsheet or a table outside the platform, and the form is the place where the answers arrive. A business that wants the sixty calls made by telephone will use the form for the customers it can reach by email and the telephone for the rest, and will keep both sets of answers in the same table.

Costly Mistakes

• Asking the prompted question first. A prompted question plants the name, and an unprompted question asked after it measures nothing but the effect of the question before it.
• Surveying the customers who already like the business. A list drawn by convenience produces a number that rises every year and describes the people who were asked rather than the market.
• Reading a rise without computing the margin of error. A movement of a few points on sixty calls is usually smaller than the uncertainty in each reading, and treating it as a result produces confidence the sample does not support.
• Changing the questions between two readings. A new question, a new list criterion or a new procedure makes the second reading incomparable with the first, and the change in the figure belongs to the change in the instrument.
• Treating one measurement as a trend. A recommendation score taken once is a baseline, and a business that reads it as a result has complimented itself rather than measured itself.
• Crediting the brand for growth that other work produced. When the offer, the entry product and referrals move in the same six months, the brand cannot be separated from them, and a scorecard that claims otherwise teaches the wrong lesson.
• Auditing consistency once and stopping. The consistency fraction is the measure a business controls most directly, and it decays as soon as new materials are made without being checked against the guide.

Key Takeaways

• Brand health is the state of what people know and feel about a business; brand knowledge has two components, awareness and image, and equity is positive when customers respond more favorably to the branded offer (Keller, 1993).
• Unprompted awareness, prompted awareness and salience are three different things, and salience — the propensity of a brand to come to mind in buying situations — is distinct from awareness and from attitude (Romaniuk & Sharp, 2004).
• A small business can measure awareness, associations, renewals, referrals, reviews and consistency, and it cannot isolate the brand's contribution to revenue when the offer, the price and the season move at the same time.
• The margin of error at 95% for a proportion is 1.96 times the square root of p times (1 − p) divided by n; for Nordec's 13 of 60 it is about ±10.4 points and for 19 of 60 about ±11.8 points, and the two intervals overlap, so the rise cannot be told from chance.
• The consistency audit is the strongest measure on a small business's scorecard because it counts objects rather than people; Nordec moved from 13 of 36 touchpoints to 33 of 36, which carries no margin of error.
• Monetary brand valuation under ISO 10668:2010 and the global rankings belong to large businesses and to transactions; a small business measures its brand for decisions, and it reads the scorecard once a year rather than every month (Binet & Field, 2013).

Your Action Plan

1. Write the questions down and fix the list criterion. Write the unprompted question, the prompted question and the open association question in their fixed order, and write the criterion that draws the list — the population, the size band, the territory and the exclusion of current customers. Deliverable: one page holding three questions and one list criterion. Verification: the questions are in the order that puts the unprompted one first, and the list criterion can be applied by somebody who was not in the room.
2. Draw the list and make the first sixty calls. Draw sixty names by the written criterion, make the calls from the business rather than from the owner, and record every answer verbatim, including the unflattering ones. Deliverable: sixty recorded answers with their dates. Verification: every answer is recorded in the respondent's words rather than in a category, and the list was drawn before the first call.
3. Compute the margins of error before reading anything. For each proportion, compute 1.96 times the square root of p times (1 − p) divided by n, and write the interval beside the figure. Deliverable: an interval beside every proportion on the scorecard. Verification: no change is described as a rise or a fall unless the intervals support it.
4. Build the scorecard with its honest reading column. Place the seven measures in one table with a column for each reading and a final column for the honest reading of each measure, and add the consistency fraction from the touchpoint audit of Chapter 7. Deliverable: one scorecard with dated readings and a reading column. Verification: every row states what the figure can and cannot support.
5. Book the next reading a year ahead and keep the instrument still. Place the next survey and the next audit on the calendar on the same dates, and make no change to the questions, the list criterion or the procedure before the second reading is taken. Deliverable: a dated entry for the next reading and a frozen questionnaire. Verification: the second reading uses the same three questions in the same order, drawn from the same population by the same criterion.

In La Machine

One route in the application copy read for this book serves the measurement work directly (La Machine, 2026). Forms, at business/marketing/forms, can carry a short survey — the likelihood-to-recommend question, or the awareness questions of section 9.2 — and the answers are kept with the form at business/marketing/forms/responses, where they can be read and counted. That is the whole of the platform's contribution to brand measurement, and it is a real one: the answers arrive in one place, with the date attached, and the business does not have to build a collection mechanism of its own. A business that makes its calls by telephone will use the form for the customers and prospects it can reach by email and keep the telephone answers in the same table, so that the two sets are read together rather than separately.

What the platform does not do belongs in the same breath, because a business that expects more will look for it and conclude that something is missing. La Machine has no brand health tracking, no awareness survey tool and no net promoter tool beyond the general form builder; it has no brand consistency audit, no way to compare a scorecard across two readings, and no brand portal or guidelines publisher. The scorecard therefore lives outside the platform, in a spreadsheet or a table, and the platform holds the answers rather than the analysis. The same is true of the touchpoint audit of Chapter 7, which is a count kept by hand, and of the review counts, which are read from the business's public profile. Book B7 (Marketing Analytics, Attribution and Dashboards) treats the wider question of which numbers a small business keeps and how it reads them, and Book B2 (Market Research and Competitive Intelligence) treats the design of the questions themselves.

Self-Check

1. What are the two components of brand knowledge, and why does a small business have to measure them separately?
2. What is the difference between unprompted awareness, prompted awareness and salience, and why is the gap between the first two useful?
3. Why is the consistency fraction the strongest measure on a small business's scorecard, and what does it not tell the business?
4. What is the margin of error at 95% for a proportion, and what does it show about Nordec's awareness figures?
5. Why does this book refuse to attribute Nordec's growth in buildings under contract to the brand refresh?

Answers. 1. Brand knowledge has two components, brand awareness and brand image, and customer-based brand equity is the differential effect of that knowledge on how customers respond to the marketing of the brand (Keller, 1993); a business must measure them separately because awareness is the easy half that a van or a sign can produce, while image is made of the associations that experience builds, and a business can be widely known and known for the wrong thing. 2. Unprompted awareness is the share who name the business when asked an open question about the category; prompted awareness is the share who recognize the name when it is read to them; salience is the propensity of the brand to be noticed or come to mind in buying situations, reflecting the quantity and quality of memory associations, and it is distinct from awareness and from attitude (Romaniuk & Sharp, 2004); a wide gap between the two awareness figures says the name is stored but not connected to the moment of need, which is a salience problem rather than an awareness problem. 3. The consistency fraction counts objects rather than people, so it carries no margin of error, and it is the measure the business controls most directly; Nordec moved from 13 of 36 touchpoints to 33 of 36, which is a fact about the business's own materials; what it does not say is whether anyone in the market has learned the assets, which is a separate question asked of customers. 4. The margin of error at 95% for a proportion p measured on n answers is 1.96 times the square root of p times (1 − p) divided by n; for 13 of 60 it is about ±10.4 points and for 19 of 60 about ±11.8 points, giving intervals of roughly 11.3% to 32.1% and 19.9% to 43.5%, which overlap, so the rise of about ten points cannot be told from chance with sixty calls each time. 5. Because the offer, the Bilan Nordec entry product and referrals all moved in the same six months, and no small sample can separate the brand's contribution from theirs; the evidence for the brand work is indirect — the consistency fraction moved decisively, the review count rose, and the awareness figures rose by an amount the sample cannot confirm — and the honest conclusion is that six months and sixty calls are not enough to know.

Chapter References

Aaker, J. L. (1997). Dimensions of brand personality. Journal of Marketing Research, 34(3), 347–356. https://doi.org/10.1177/002224379703400304

Binet, L., & Field, P. (2013). The long and the short of it: Balancing short and long-term marketing strategies. Institute of Practitioners in Advertising.

Interbrand. (2025). Best global brands 2025. Retrieved September 25, 2026, from https://interbrand.com/best-global-brands/

International Organization for Standardization. (2010). Brand valuation — Requirements for monetary brand valuation (ISO Standard No. 10668:2010). https://www.iso.org/standard/46032.html

International Organization for Standardization. (2021). Brand evaluation — Part 1: Principles and fundamentals (ISO Standard No. 20671-1:2021). https://www.iso.org/standard/81739.html

Kantar. (2026). Kantar BrandZ most valuable global brands 2026. Retrieved September 25, 2026, from https://www.kantar.com/campaigns/brandz/global

Keller, K. L. (1993). Conceptualizing, measuring, and managing customer-based brand equity. Journal of Marketing, 57(1), 1–22. https://doi.org/10.1177/002224299305700101

Keller, K. L. (2001). Building customer-based brand equity: A blueprint for creating strong brands (Report No. 01-107). Marketing Science Institute.

La Machine. (2026). Documentation — route reference inventory. Internal product documentation, read September 25, 2026, together with the code of the pages this book describes: business/marketing/forms and business/marketing/forms/responses.

Reichheld, F. F. (2003). The one number you need to grow. Harvard Business Review, 81(12), 46–54.

Romaniuk, J. (2018). Building distinctive brand assets. Oxford University Press.

Romaniuk, J., & Sharp, B. (2004). Conceptualizing and measuring brand salience. Marketing Theory, 4(4), 327–342. https://doi.org/10.1177/1470593104047643

Walsh, M. F., Winterich, K. P., & Mittal, V. (2010). Do logo redesigns help or hurt your brand? The role of brand commitment. Journal of Product & Brand Management, 19(2), 76–84. https://doi.org/10.1108/10610421011033421

Chapter 10 — Typical Tasks, Cadence and Workflows

A brand stays coherent not because it was designed well once but because somebody does a handful of small, dull things on a rhythm — checking what goes out, keeping the templates current, spot-checking the touchpoints and updating the guide — and this chapter turns the work of Chapters 2 to 9 into that rhythm.

What You'll Be Able to Do

• Describe the four rhythms of brand management — weekly, monthly, quarterly and annual — and say what each one is for.
• Run the weekly check of a new material in five minutes, using the logo, the colors, the typefaces, the voice and the French version as its five steps.
• Prepare the quarterly review by drawing ten of the thirty-six touchpoints at random and reading the new online reviews for words that match or contradict the promise.
• Plan the annual review, including the full audit of the touchpoints, the awareness survey, the review of the platform and the revision of the guide.
• Route a new material through the right approval, from a template the brand keeper can check to a brief that goes to the design studio.
• Recognize the four things that break the cadence and put in place the calendar, the backup and the templates that keep it running.

Why It Matters

A brand is a promise repeated, and repetition is a task before it is anything else. The work of Chapters 2 to 8 produced a platform, a name, a verbal identity, a visual identity, a design system, a guide of twenty-four pages and a plan for the rollout. Every one of those is an artifact, and an artifact does not hold a brand in place. What holds it in place is somebody noticing, week after week, that a document about to leave the business carries the right logo, the right blue, the right words and the right promise. A business that has done the strategy work and skipped the maintenance has bought the expensive half of the job and left the cheap half undone.

Drift arrives through four ordinary events, and none of them looks like a mistake on the day it happens. A new person joins with the habits of another business and produces documents the way they always have. A new document is needed by Friday and nobody thinks to design it. An urgent job is answered by reusing a file that is two years old. A print shop or a sign maker receives an ambiguous file and improves it, because improving it seemed helpful. Each of the four is reasonable on its own, and together they are how a brand that was consistent in March is inconsistent by November. The inventory of Chapter 7 is what makes the sum visible, and its first reading at Nordec was blunt: three versions of the logo in circulation, five different blues, four typefaces, and only thirteen of the thirty-six touchpoints consistent with one another, or 36.1 per cent.

The cost of leaving the work undone is not the cost of a rebrand, which is large and occasional, but the slow loss of the recognition the business paid for. Six months after the rollout, the same inventory read thirty-three of thirty-six touchpoints consistent. That number is a measurement, not a mechanism, and nothing inside it keeps the figure from falling again. The rhythm that keeps it is small: at Nordec it costs sixty-one hours of the administrator's time in a year, which is $2,196 at thirty-six dollars an hour, against a one-time rollout of $24,120. That work sits inside the identity process itself, as the fifth of five phases — research, strategy, design, launch and governance — rather than as an afterthought that follows the launch (Wheeler & Meyerson, 2024). A small business that treats governance as a phase rather than a chore is doing what the process asks, and it is protecting the spending it has already made.

In This Chapter

Section 10.1 explains why a brand drifts even when nobody makes a mistake, and what a brand keeper, a release and a rhythm are. Section 10.2 describes the weekly check of new materials and the monthly reading of a sample, with the brand calendar that holds them. Section 10.3 covers the quarterly review, its ten touchpoints drawn at random and its one-page note of findings. Section 10.4 covers the annual review, from the full audit of the thirty-six touchpoints to the awareness survey and the revision of the guide. Section 10.5 sets out how a new material is requested, produced, checked, approved and filed. Section 10.6 covers the onboarding of people to the brand, from the day-one summary to the first visit sheets of a new technician. Section 10.7 divides the work in three sizes of business. Section 10.8 names what breaks the cadence and what keeps it running anyway.

10.1 Brand Management as a Recurring Practice

A brand is not a document, and it is not a moment. It is a promise that a business repeats, in the same words and the same shapes, across hundreds of small occasions in a year, and repetition is a task before it is anything else. The work of Chapters 2 to 8 produced a platform, a name, a verbal identity, a visual identity, a design system, a guide of twenty-four pages and a plan for the rollout. Every one of those is an artifact, and an artifact does not hold a brand in place. What holds it in place is somebody noticing, week after week, that a document about to leave the business carries the right logo, the right blue, the right words and the right promise.

Drift arrives through four ordinary events, and none of them looks like a mistake on the day it happens. The first is a new person, who joins with the habits of another business and produces a document the way they always have. The second is a new document, a form or a report that nobody thought to design because it was needed by Friday. The third is an urgent job, in which the fastest way to answer a customer is to reuse a file that is two years old. The fourth is a supplier, a print shop or a sign maker who receives a file and improves it, because the file was ambiguous and the supplier wanted to be helpful. Each of the four is reasonable. Together they are how a brand that was consistent in March is inconsistent by November.

Drift is invisible from inside the business, and that is its most dangerous property. The people who work in a company see each decision separately: this report, this van, this proposal. Nobody sees the sum, because the sum is spread across thirty-six touchpoints and several months. The inventory of Chapter 7 is what makes the sum visible, and its first reading at Nordec was blunt: three versions of the logo in circulation, five different blues, four typefaces, and only thirteen of the thirty-six touchpoints consistent with one another, or 36.1 per cent. Six months after the rollout the same inventory read thirty-three of thirty-six. That second number is a measurement, not a mechanism. Nothing in it keeps the figure from falling again, and the only thing that will is a routine.

The consistency audit of Chapter 7 and the scorecard of Chapter 9 are both necessary and neither is sufficient. The audit is a photograph: it is accurate on the day it is taken and it begins to age immediately. The scorecard is a measurement: it tells a business whether the brand is moving, and it does so four times a year at best. What sits between them is the work this chapter describes, which is the ordinary maintenance that keeps the photograph from going stale and gives the scorecard something to measure. That work sits inside the identity process itself, as the fifth of five phases — research, strategy, design, launch and governance — rather than as an afterthought that follows the launch (Wheeler & Meyerson, 2024). Governance, in that account, is a phase of the work, with its own tools and checklists, and a small business that treats it as a phase rather than a chore is doing what the process asks.

A rhythm is cheaper than a rescue, and the comparison is not close. A rescue is a rebrand: a design fee, a rollout, a set of documents to redo, a set of customers to explain the change to, and the risk that the recognition built over four years is spent rather than extended. The rhythm costs a fraction of that, and it spends its money on prevention rather than on repair. It also has a second advantage that is easy to overlook: it is performed by people who already work in the business, in pieces of half an hour and an hour, rather than by a project that has to be scheduled, budgeted and finished.

What governance means in a business of seven people is smaller than the word suggests. It means a named person who holds the templates and the asset folder, a calendar with the four rhythms on it, a checklist that fits in five minutes, a folder where every released material is filed with its version, and a short log of who approved what. Five things, none of them expensive, and together they are the difference between a brand that is managed and a brand that is remembered. The rest of this chapter describes each of the five in the order in which a business meets them.

DEFINITION Brand drift, brand keeper and release

Brand drift is the slow divergence of what a business publishes from what its brand platform says, produced not by a decision but by the accumulation of small, reasonable ones. The brand keeper is the person inside the business who holds the templates and the asset folder, checks every new material before it goes out, keeps the calendar of the four rhythms and files what has been released; at Nordec the role is held by the administrator, and it is a role rather than a title. A release is any material that leaves the business and can be seen by somebody outside it: a proposal, a report, a visit sheet, an invoice, a letter, an email signature, a social post, a page of the website, a van, a uniform, a sign. The asset folder is the single place where the current version of every identifier and every template is kept, so that the question of which file is the right one has one answer.

KEY CONCEPT A rhythm beats a rescue

A brand is maintained by small checks performed often, not by large corrections performed rarely. The reason is arithmetic rather than philosophy: a material checked before it leaves costs five minutes, and the same material discovered six months later costs a reprint, a reissue, an apology and a conversation about why the business cannot keep its own identity straight. The rhythm also has a property that a rescue cannot have, which is that it is continuous. A rescue restores consistency to the state it had on the day of the rescue, and then the same four events begin again. A rhythm never lets the gap grow large enough to need a rescue, and it does so at a cost the business can carry in an ordinary month. The rule that follows is simple: when a choice has to be made between a check and a correction, choose the check.

10.2 The Weekly and Monthly Rhythm

The weekly rhythm is thirty minutes and it is the one that does most of the work, because it is the only one that runs often enough to catch a material before it leaves the building. The brand keeper sees every new material before it goes out: a proposal, a report, a social post, a letter, a page of the website, a proof from a supplier. The check is not a review of the work itself; it is a check of the identity, and it takes about five minutes per material. The remaining time in the half hour goes to the requests that arrive from colleagues, which are usually requests for a template or for a file, and to filing what has been released.

The monthly rhythm is one hour and it looks backward rather than forward. The brand keeper takes a sample of what actually went out in the month — the monthly reports, the visit sheets, the social posts — and reads them against the guide. The monthly reports are the largest group: forty-eight buildings on the Preventive and Portfolio tiers produce five hundred and seventy-six reports in a year, and a sample of them shows whether the template is being used as designed or whether it has been quietly modified by the people who fill it in. The visit sheets are the second group, and they matter because they are produced on site, by technicians, at the end of a long day, which is exactly the condition under which a template gets abandoned.

What the check looks at is fixed, and it is fixed so that it can be performed by somebody who is not a designer. The first thing is the logo: which version is on the material, and whether it has the clear space the guide asks for. The second is the colors: whether the blues and the amber are the ones in the guide, and whether the text on them meets the contrast the guide records. The third is the typefaces: whether the material uses the humanist sans-serif for everything and the monospaced face only for building record codes. The fourth is the voice and the promise: whether the wording says the time, the cost and what comes next, and whether anything in it contradicts the promise that Nordec knows the building and tells the owner ahead. The fifth is the French version, where the material needs one.

The four rhythms are drawn in Figure 10.1. The figure shows four rows, one for each rhythm. The first row is the weekly rhythm: the check of every new material before release and the template requests, half an hour, performed by the brand keeper. The second row is the monthly rhythm: a sample of the reports, the visit sheets and the social posts that went out, one hour. The third row is the quarterly rhythm: a spot-check of ten of the thirty-six touchpoints and a reading of the new online reviews with the owner, two hours. The fourth row is the annual rhythm: the full audit of the thirty-six touchpoints, the awareness survey, and the review of the guide and of the platform with the whole team, about two days of the brand keeper's time and a half-day workshop.

Figure 10.1 — The brand's working rhythm at Nordec. Small checks often, a full audit once a year.

The following table sets out the brand calendar at Nordec in the order in which the year meets it, with the task, the person who performs it, the time it takes and the output it leaves behind. Read it as the business's own document rather than as a description of one: the value of the calendar is that the same tasks fall on the same dates every year, so that a missed week is visible against fifty-one identical ones.

Table 10.1 — Nordec's brand calendar

METHOD Checking a new material in five minutes

First, look at the logo: confirm that it is the current version, that it is the one drawn with the horizon line through the letter N, and that it has the clear space the guide asks for around it. Second, look at the colors: confirm that the blues and the amber are the ones in the guide, and that any text sitting on a color meets the contrast the guide records — white on Nordec Blue, Nordec Blue on Mist, Slate on white, and never amber text on white. Third, look at the typefaces: the humanist sans-serif for everything, the monospaced face only for building record codes, and no third face introduced by a supplier. Fourth, read the wording aloud and listen for the voice: does it say the time, the cost and what comes next, and does anything in it contradict the promise that Nordec knows the building and tells the owner ahead? Fifth, confirm the French version where the material needs one, then file the material in the asset folder with its version number and write the check into the approval log.

DO THE MATH What the brand rhythm costs Nordec in a year

Illustrative figures; the case is fictional. Step 1 — The weekly check takes half an hour and runs about fifty weeks a year, which is 0.5 × 50 = 25 hours. Step 2 — The monthly reading of a sample takes one hour and runs twelve times, which is 12 hours. Step 3 — The quarterly review takes two hours and runs four times, which is 8 hours. Step 4 — The annual review takes about two days of the brand keeper's time, which is 16 hours, plus a half-day workshop with the whole team. Step 5 — The brand keeper's year is 25 + 12 + 8 + 16 = 61 hours, which at the administrator's rate of $36 an hour is $2,196 a year; set against the one-time rollout of $24,120, the rhythm protects that spending for about 9.1 per cent of it each year, and it is the only part of the brand work that never ends.

TOOLS OF THE TRADE What a brand calendar needs

Four functions carry the calendar, and a business should look for them rather than for a product. The first is a shared calendar with recurring entries for the weekly check, the monthly reading, the quarterly review and the annual day, visible to the people who perform them and to the owner who attends part of them. The second is a checklist for new materials, short enough to be performed in five minutes and fixed enough that two people performing it reach the same answer. The third is an asset folder with versions, where the current logo, palette, typefaces and templates sit in one place and superseded files are removed rather than kept beside them. The fourth is a simple log of approvals, one line per release, with the date, the material, the person who checked it and the level at which it was approved. No brand names belong on this list, and no prices; the requirement is the function.

10.3 The Quarterly Brand Review

The quarterly review takes two hours and it is the rhythm at which the business stops looking at individual materials and starts looking at the brand as a whole. The brand keeper draws ten of the thirty-six touchpoints at random, checks each one against the guide, and then sits down with Sophie to read the new online reviews and the comments customers have left in the quarter. The two halves of the review answer different questions. The spot-check answers whether the identity is holding. The reading of the reviews answers whether the promise is landing.

The spot-check is random on purpose. A business that checks the same touchpoints every quarter checks the ones it already knows are fine, because those are the ones somebody thought of. Drawing ten at random from the inventory of thirty-six means that a touchpoint nobody has looked at in a year will come up, and it means that the review cannot be satisfied by tidying the same three documents. The ten are checked against the same five things the weekly check uses, with one addition: whether the touchpoint still exists at all. A touchpoint that has quietly stopped being used is a finding, because the inventory is a description of the business and a description that has drifted from the business is worth correcting.

The reading of the reviews is the part of the quarter that most often produces a change. Nordec's online reviews stood at twenty-three with an average of 4.7 before the rollout and thirty-one with an average of 4.8 six months after it, and the numbers matter less than the words. The brand keeper reads the new ones for language that matches the promise and for language that contradicts it. "On time" and "fast" appeared in fifteen of the twenty value interviews of Book A6, "knows my building" in eleven, "tells me ahead" in nine; a review that says the technician arrived when they said they would is the promise being repeated by a customer, and a review that says nobody warned the owner before a part was replaced is the promise being contradicted. Both are findings, and the second is the more valuable of the two.

The output of the quarterly review is one page. It lists what was checked, what was found, what will be fixed and who will fix it, with a date. The page is deliberately short, because a long report is a report nobody reads, and because the review's purpose is to produce a small number of decisions rather than a complete account of the quarter. The findings that concern a single material are fixed by the brand keeper. The findings that concern a template are fixed by updating the template and telling the people who use it. The findings that concern the promise itself are not fixed at the quarterly review at all; they are carried to the annual review, because a promise is not changed in a two-hour meeting.

One habit makes the quarterly review worth its two hours, and it is the habit of writing the findings down even when they are small. A finding that is not written down is a finding that will be made again in the next quarter by somebody who has forgotten it, and the third time it is made the business begins to treat it as a fact of life rather than as a defect. The one-page note is also what makes the annual review cheap: the annual day reads four of those pages and starts from a list of what has already been noticed rather than from nothing.

CASE IN POINT Nordec Entretien inc.: the door hanger made at the print shop

Nordec Entretien inc. is a fictional business created for this encyclopedia; its owner, its employees, its customers, its competitors, its figures and the events described are invented, and any resemblance to a real business is coincidental. A technician had a building to visit on a Friday and wanted to leave something behind for the owner, so they took an old file from a shared folder to a print shop and came back with a box of door hangers carrying the wrench logo inside a circle, in one of the five blues the business no longer used. The weekly check caught it on the invoice rather than on the door, because the print shop's bill arrived before the hangers were distributed. The stock was not thrown away; it was kept for the depot, where only Nordec's own people would see it, and the file was removed from the shared folder so that the next person in a hurry would not find it. The finding was written into the quarterly note, and the fix was not a conversation about care but a template: a door hanger that the brand keeper can produce in ten minutes, which is faster than the trip to the print shop.

10.4 The Annual Brand Review

The annual review is the largest of the four rhythms and the only one that asks whether the brand is still the right one. It takes about two days of the brand keeper's time and a half-day workshop with the whole team, and it contains four pieces of work: the full audit of the thirty-six touchpoints, the awareness survey, the review of the platform, and the revision of the guide. The four are performed in that order, because each one feeds the next.

The full audit is the spot-check performed on everything. Every one of the thirty-six touchpoints is checked against the guide, and the result is written as a single number beside the previous year's number. At Nordec the first audit read thirteen of thirty-six consistent, or 36.1 per cent, and the audit six months after the rollout read thirty-three of thirty-six. The annual audit is what turns those two readings into a series, and a series is what tells a business whether its routine is working. An audit that produces the same number two years running is not a failure; it is a stable brand, and stability is what the routine is for. An audit that produces a falling number is a signal that something in the business has changed faster than the guide has.

The awareness survey is the second piece of work, and it is described in full in Chapter 9. It is sixty telephone calls to owners of buildings not under contract in the two territories, asking first whether they can name a maintenance provider unprompted and then whether they have heard of Nordec. The first reading was thirteen of sixty, or 21.7 per cent, with a margin of error of about plus or minus 10.4 points at 95 per cent; the second was nineteen of sixty, or 31.7 per cent, with a margin of about plus or minus 11.8 points. The rise cannot be distinguished from chance, and the honest reading of the two figures is that the survey has not yet shown a movement. The annual review keeps the question rather than the answer, because a measurement taken once is an anecdote and a measurement taken every year is a trend.

The review of the platform is the third piece of work and the most uncomfortable. The platform is the purpose, the promise, the three values and the four personality traits set out in Chapter 2, and the annual question is whether each of them is still true of the business as it is now. Nordec's promise is that it knows the building and tells the owner ahead, and the annual review asks whether the business still does that at fifty-six buildings as it did at fifty-two. The three values each carry a decision they change: Punctual puts the ninety-minute commitment before a more profitable job, Straight puts a two-hour call before exceeding the included work, Thorough puts the record update on site before leaving. A value that no longer changes a decision is a value that has become a word on a wall, and the annual review is where that is noticed.

The revision of the guide is the fourth piece of work and it closes the loop. The findings of the audit, the survey and the platform review become changes to the guide, and the changes are made once a year rather than continuously, so that the document the business works from is stable for twelve months at a time. The decisions available at the annual review are three: fix the touchpoint, update the guide, or brief the design studio. A touchpoint that has drifted is a fix. A rule that no longer matches the way the business works is an update to the guide. A need that the guide cannot meet — a new tier, a new service, a new market — is a brief to the studio, with a fee and a date. The half-day workshop with the whole team is where the three are separated, and it is the only occasion in the year when the seven people who deliver the brand are in the same room to talk about it.

WHAT CHANGED What the refresh changed in the team's week

Before the refresh, nobody in the business checked a material before it went out. The three versions of the logo, the five blues and the four typefaces were not the result of a decision; they were the result of nobody having the task. Since the rollout, dated in this book as six months after the refresh of Book A6, the weekly check exists, it has a named owner, and it takes half an hour of the administrator's week. The change is small enough to be described in one sentence and large enough to alter what the business publishes: the same inventory that read thirteen of thirty-six touchpoints consistent before the rollout read thirty-three of thirty-six after it, and the weekly check is the part of the routine that keeps the second figure from sliding back toward the first. The lesson is not that a check is powerful. It is that a task with a name, a time and a place on the calendar is performed, and a task without them is not.

10.5 Producing and Approving New Materials

Every material a business publishes follows a path, and the path is worth writing down because the alternative is a path that changes with whoever is in a hurry. The path has seven steps: the request, the choice between a template and a brief, the draft, the check against the guide, the approval at the right level, the release, and the filing of the material in the asset folder with its version. The seven steps take less time than they sound, because six of them are minutes and only the draft is work.

The request is the first step and it is deliberately small. A colleague needs something: a proposal for a Preventive contract, a page for the website, a card for a new technician, a sign for the depot. The request names what it is for, who will see it, when it is needed and whether anything like it already exists. The last question is the one that saves the most time, because a business of seven people produces the same document many times a year, and the answer is usually that a template already exists. The brand keeper answers the request within a day, which matters more than it sounds: a request that waits is a request that gets answered by somebody improvising.

The second step is the fork in the path. If the material fits a template, the brand keeper produces it or hands the template to the person who asked, and the material goes to the check. If the material does not fit a template, it does not get improvised; it goes to the design studio with a brief. The brief is short and has four parts: what the material is for, who will see it, what it must say, and which existing template it is closest to. The studio's fixed fee of $7,800 covered the logo, the palette, the typography, eleven templates and the guide, and work outside that scope is quoted separately, which is why the brief exists. A business that sends a vague request to a studio pays for the studio's time in discovering what was wanted.

The third step is the draft, produced by whoever is closest to the content. The fourth is the check, performed by the brand keeper against the five things of section 10.2, and recorded in the approval log. The fifth is the approval at the right level, which is the step most often skipped and the one that decides whether the business can answer for what it published. The rule at Nordec is simple and it is set out in Table 10.2: anything that changes the promise, the name or a core identifier is approved by Sophie; anything built from a template is approved by the brand keeper; anything new that does not fit a template goes to the design studio with a brief. The sixth step is the release, and the seventh is the filing, which is what makes the next request cheap.

The following table sets out who approves what at Nordec, with an example of each kind of change and the evidence the business keeps. Read it as the answer to a question that arrives in the middle of a busy week, when somebody needs to know whether they can decide alone: the column of evidence is what makes the answer checkable a year later.

Table 10.2 — Who approves what

QUÉBEC & CANADA NOTE The French version is part of the check

In Québec, the Charter of the French language requires French in commerce and business, including commercial documents, signs and product inscriptions, and this book states its rules in general terms rather than by article. For the routine in this chapter the consequence is practical: the French version of a material is not a translation task performed after the material is approved, it is one of the five things the weekly check verifies before release. A proposal, a report, a letter, a page of the website and a social post all leave the business and all reach people in Québec, and the check confirms that the French version exists, that it says the same thing as the English one, and that the names the business uses in French — Essentiel, Préventif, Portefeuille, Bilan Nordec — are the ones the guide sets out. The check is performed by the brand keeper, in the same five minutes as the rest, and it is recorded in the approval log like every other step. This book explains the law; it does not replace the advice of a lawyer.

THE RULE Review what you publish

The Competition Act governs what a business may say to the public. Section 52 makes it an offense to knowingly or recklessly make a representation to the public that is false or misleading in a material respect, and paragraph 74.01(1)(a) treats a representation to the public that is false or misleading in a material respect as reviewable conduct, a civil matter; both apply to sales to businesses as well as to sales to consumers. A brand promise or a tagline is a representation to the public, and so is every material that repeats it. The practical consequence for this chapter is that the weekly check is not only an aesthetic exercise. A material that claims a ninety-minute response must be one the business can meet, and a report that says a part was inspected must describe an inspection that happened. The check is where a claim is read before it is published rather than after a customer has relied on it. This book explains the law; it does not replace the advice of a lawyer.

10.6 Onboarding People to the Brand

The people of a business deliver its brand, and Chapter 7 makes the point at length: a customer meets the promise through a technician at a door, a dispatcher on the telephone and an administrator in an email, long before they meet a logo. That is why onboarding is a brand task and not only a human resources task, and why it belongs in this chapter rather than in a policy manual. A new person who has never been told what the business promises will deliver whatever they brought with them from the last place they worked, and they will do it with the best intentions.

The first day carries three things. The new hire receives the one-page summary of the guide, which is the whole of the identity on a single sheet: the promise, the three values, the four personality traits, the logo, the palette with its hexadecimal codes, the two typefaces and the tagline in French and in English. They receive their email signature, already built from the template, so that the first message they send carries the identity of the business rather than a blank line. And they receive the uniform, which at Nordec is issued in the first week: seven people wear it, and the two vans carry the same identity on the road every working day.

The technicians receive a second session, thirty minutes long, on the voice principles and on the visit sheet. The voice has four principles — say the time, say the cost, say what comes next, and no jargon without its plain word — and they are taught with examples rather than with rules, because a technician writing a visit sheet at the end of a long day needs to know what a good one looks like rather than what a bad one violates. The visit sheet is the material that matters most in this session, because it is the one the customer keeps: six hundred and twenty-four preventive visits a year produce six hundred and twenty-four of them, and each one is a page of the brand in a building's file.

The check that closes the onboarding is the first visit sheets of the new technician, read by the brand keeper in the weekly check for the first month. The purpose is not to grade the person but to catch the two things that go wrong most often: a sheet filled in with abbreviations only the technician understands, and a sheet that reports what was done without saying what comes next. Both are corrected in a two-minute conversation, and both are corrected before the customer has seen a month of them. A business that waits for the annual audit to discover that a new technician has been writing visit sheets in their own style has spent a year teaching a customer a different brand.

The onboarding also has a second direction, which is the one a small business forgets. A person who has been in the business for years and takes on a new task needs the same thirty minutes as a new hire, because the task is new even if the person is not. The administrator who becomes the brand keeper is the clearest case: the role is new, the templates are new, and the five-minute check is a skill rather than an instruction. At Nordec the brand keeper learned the check by performing it beside Sophie for the first month, on real materials, before performing it alone.

WHO DOES THIS The brand keeper, with the owner's signature

At Nordec the brand keeper is the administrator, and the role is defined by what they hold rather than by what they are called. They hold the templates and the asset folder, they perform the weekly check and the monthly reading, they prepare the quarterly review and the annual audit, they answer the requests that arrive from colleagues, and they keep the approval log. They do not decide what the brand promises, and they do not change an identifier. The owner's part is the signature: Sophie approves anything that changes the promise, the name or a core identifier, attends the quarterly review and three hours of the annual day, and is the person the brand keeper escalates to when a request does not fit a template. The division is what makes the role sustainable in a business of seven people, because a brand keeper who also owns the promise is a brand keeper who cannot be replaced, and a brand keeper who owns nothing is a person whose checks are advice.

10.7 Who Does What in Three Sizes of Business

The four rhythms do not change with the size of the business; what changes is how many people touch them and how much of the work can be carried by one person. In a solo business the owner performs every step and is the single point of failure for all of them. In a team of about seven, like Nordec, one person holds the brand keeper role and the owner holds the signature. In an established company, the work divides further, and a marketing manager owns the guide, an asset library and the yearly audit, with an agency performing the design work that falls outside the templates.

The solo business runs two rhythms rather than four, and it does so deliberately. The monthly reading and the annual review are the ones that survive, because the monthly reading catches the materials that have gone out and the annual review catches the direction of the brand. The weekly check is folded into the act of producing the material, since the person who makes a document and the person who checks it are the same person, and the quarterly review is folded into the annual one. What the solo business cannot do is skip the annual review, because it is the only occasion on which anybody asks whether the promise is still true. A solo owner who keeps the templates in one folder, checks a material against the guide before sending it, and takes half a day once a year to read the whole identity has the discipline in the form that fits.

The team of about seven is the size at which the brand keeper role becomes worth naming. One person holds the templates and the folder, performs the weekly check and the monthly reading, prepares the quarterly and annual reviews, and answers the requests. The role takes sixty-one hours a year at Nordec, which is a little over an hour a week, and it is performed by somebody who already works in the business rather than by a new hire. The owner's part is the signature and the attendance, and the design studio is called when a need falls outside the templates. This is also the size at which the onboarding described in section 10.6 becomes a routine rather than an event, because a business of seven people hires often enough for the first day to need a procedure.

The established company divides the work along the lines of the four rhythms and adds a fifth person outside the business. A marketing manager owns the guide, the asset library and the yearly audit, and holds the approval authority that Sophie holds at Nordec. A designer or an agency performs the work that falls outside the templates, on a retainer or by the project. The brand keeper's weekly check becomes a review performed by the marketing team rather than by one person, and the asset library becomes a system rather than a folder, because the number of materials published in a month has grown past what one person can hold. The four rhythms are the same four, and the difference is that each of them is performed by somebody whose job it is.

The following table sets out seven recurring brand tasks and who performs each of them in a solo business, a team of about seven and an established company. Read it as a distribution of responsibility rather than a hierarchy: the tasks are the same in every column, and what changes is whether one person carries several of them or several people carry one each.

Table 10.3 — Who does what in three sizes of business

IN LA MACHINE Where the rhythm can live

The platform can hold the calendar of the four rhythms, and it cannot hold the brand itself (La Machine, 2026). Projects, Tasks, Meetings and the Calendar can carry the brand project and its recurring checks, and in the Control center the Plan tasks page gives each task a verification criterion, which is the closest the platform comes to the checklist this chapter describes. Marketing → Templates stores message templates by channel with variables and a preview, which is useful for the wording of a message and holds no brand colors, no logo and no footer, so a message built there is not a released material until it has been through the check. What the platform does not have belongs in the same breath: there is no brand audit, no approval workflow, no saved brand kit that other pages read, no brand portal, no asset library with versions and no brand health tracking. The Branding page computes the contrast of a primary color and shows a pass or fail badge at the 4.5:1 threshold, and its values are not saved, so it is a place to try a palette rather than a register of the brand.

10.8 What Breaks the Cadence

Four things break the cadence in a small business, and each of them has a remedy that can be put in place before the break rather than after it. The first is the busy season, when the work that pays arrives all at once. The second is the departure of the brand keeper, whether to another role or out of the business. The third is a supplier who improves a file. The fourth is the urgent material made outside the templates. None of the four is unusual, and none requires a large response; each requires a decision taken in a quiet week, because the decision is easy then and impossible in the week the break happens.

The busy season is the break every business expects and few plan for. The remedy is to decide in advance which parts of the routine are protected and which are reduced. The weekly check is the part that must be protected, because it is the only one that catches a material before it leaves, and a busy season is exactly when materials leave in a hurry. What can be reduced is the depth of the monthly reading, which can look at a smaller sample, and the length of the quarterly review, which can hold its agenda in an hour. What is most dangerous is to reduce everything at once, because the habit is harder to rebuild than to keep, and a business that has skipped a quarter has a folder of unchecked materials to work through in the month it decides to start again.

The departure of the brand keeper is the break that a named backup prevents, and the backup has to exist before the departure rather than being improvised on the first morning without them. The backup's job is deliberately narrow: perform the weekly check, file what is released, and report anything that needs a decision to Sophie. The backup does not need to decide whether a template should change; the backup needs to notice that a material does not match the guide and to say so. A backup who has performed the check once, on an ordinary week, with the real folder in front of them, is worth more than a document describing the check, because the questions a document does not answer are the questions that arrive in the first ten minutes.

A supplier who improves a file is the break that arrives with good intentions. A print shop receives a logo with an ambiguous clear space and centers it differently; a sign maker receives a color specification and matches it by eye; a vehicle-wrap shop receives a file and redraws it at a larger size. The remedy is not to distrust suppliers but to remove the ambiguity before the file leaves: the guide carries the logo in the formats a supplier needs, with the clear space drawn, the colors written with their hexadecimal codes, and a note saying that the file is not to be redrawn. The check on the proof, before production begins, is the second half of the remedy, and it is the step that catches the supplier who improved the file anyway.

The urgent material made outside the templates is the fourth break and the most common. A deadline appears, a customer needs something by Friday, and the fastest answer is an old file or a new one made from nothing. The remedy is not to forbid it, which would trade a small risk for a missed deadline; the remedy is to make the template faster than the improvisation. A door hanger the brand keeper can produce in ten minutes beats a trip to the print shop, and a proposal template that takes twenty minutes beats a document assembled from an old one. Every finding of this kind recorded in a quarterly note is a request for a template, and the templates that exist at Nordec — eleven of them, from the visit sheet to the social post — are the accumulated answer to the times somebody was in a hurry.

WATCH OUT The urgent flyer made outside the templates

The urgent material is dangerous because it is justified. The deadline is real, the customer is real, and the person making it is trying to help. What makes it a trap is that it is invisible: it leaves the business, it is seen by a customer, and it never passes through the check, so nobody knows it exists until a quarterly spot-check draws its touchpoint at random or a customer mentions it. Three habits reduce the risk without slowing anybody down. The first is to make the template faster than the improvisation, which is a design problem rather than a discipline problem. The second is to allow the material to go out and require the line: whoever made it tells the brand keeper in the same week, so that it is checked after the fact and a template is built from it. The third is to keep the old files out of reach, because a shared folder full of superseded versions is an invitation to the person in a hurry, and the person in a hurry will always take the file that is easiest to find.

Costly Mistakes

• Treating the brand as a project that finished. A platform, a guide and a rollout are artifacts, and artifacts do not keep a brand consistent; the work that does is the small check performed every week by somebody whose task it is.
• Leaving the check to whoever produces the material. The person who makes a document in a hurry is the person least able to see that it carries an old logo, and a check performed by its author is not a check.
• Checking the same touchpoints every quarter. A spot-check that always looks at the same three documents confirms what is already known; the ten touchpoints are drawn at random so that the ones nobody thought of come up.
• Approving a change to the promise at the wrong level. A promise, a name or a core identifier changed by whoever was closest to the request is a brand changed without a decision, and the change will be discovered in a customer conversation rather than in a review.
• Sending a vague request to the design studio. A brief that does not say what the material is for, who will see it, what it must say and which template it is closest to pays the studio to discover the answer.
• Onboarding a new person without the brand. A technician who has never been shown the voice principles and the visit sheet will write in their own style for a year, and the customer will have learned a different brand by the time anybody notices.
• Letting the brand keeper leave without a backup. A role held by one person with no second is a routine that stops on the day that person is away, and the materials produced during the gap are the ones the next audit finds.

Key Takeaways

• A brand drifts through four ordinary events — a new person, a new document, an urgent job and a supplier who improves a file — and none of them looks like a mistake on the day it happens, which is why the rhythm rather than the rescue is the remedy.
• The rhythm has four speeds: a weekly check of every new material before release, a monthly reading of a sample, a quarterly spot-check of ten of the thirty-six touchpoints with a reading of the new reviews, and an annual audit of everything with the awareness survey and the revision of the guide.
• At Nordec the rhythm costs the brand keeper sixty-one hours a year — twenty-five weekly, twelve monthly, eight quarterly and sixteen annual — which is $2,196 at thirty-six dollars an hour, about 9.1 per cent of the $24,120 one-time rollout, and it is the part of the brand work that never ends.
• Every material follows the same path: request, template or brief, draft, check against the guide, approval at the right level, release, and filing with its version; the promise, the name and the core identifiers are approved by the owner, and anything built from a template by the brand keeper.
• The people deliver the brand, so onboarding is a brand task: the one-page summary on the first day, the email signature, the uniform in the first week, a thirty-minute session on the voice and the visit sheet for technicians, and a reading of the first visit sheets by the brand keeper.
• The cadence breaks through a busy season, the departure of the brand keeper, a supplier who improves a file and the urgent material made outside the templates; the remedies are a calendar, a named backup, a proof checked before production and templates that are faster than improvising.

Your Action Plan

1. Name the brand keeper and write the role down. Choose the person who will hold the templates and the asset folder, perform the weekly check and prepare the reviews, and write the five things the role holds. Deliverable: a named brand keeper and a one-paragraph role description. Verification: the person can say which materials they check and which decisions they escalate to the owner.
2. Put the four rhythms on the calendar for the next twelve months. Place the weekly check, the monthly reading, the quarterly review and the annual day as recurring entries, with the owner's attendance marked inside the quarterly and annual ones. Deliverable: dated entries for about sixty-one hours of work. Verification: the entries are on the calendar of the person who performs them, and the annual day is booked before the busy season rather than after it.
3. Write the five-minute checklist and use it on the next material. Set out the five steps — the logo version and its clear space, the colors and their contrast, the typefaces, the voice and the promise in the wording, and the French version where it is required — and perform it on the next material that leaves the business. Deliverable: a one-page checklist and one checked material. Verification: the check is recorded in the approval log with its date and the name of the person who performed it.
4. Write the approval table and the release path. Set out who approves what, with an example of each kind of change and the evidence kept, and write the seven steps a material follows from request to filing. Deliverable: a one-page approval table and a described path. Verification: the next request that arrives is answered within a day, and the material is filed with its version.
5. Onboard the next person to the brand and check their first materials. Give the one-page summary on the first day, the email signature and the uniform in the first week, and the thirty-minute session on the voice and the visit sheet for a technician. Deliverable: a dated onboarding and a first-month check of the new person's materials. Verification: the brand keeper has read the first visit sheets and corrected anything that did not match the guide.

In La Machine

Four routes in the application code read for this book touch the routine directly (La Machine, 2026). Projects (productivity/projects), Tasks (productivity/tasks), Meetings (productivity/meetings) and the Calendar (productivity/calendar) can carry the brand project and its recurring checks, which is where the four rhythms can live as dated work rather than as intentions, and in the Control center the Plan tasks page (/taches) gives each task a verification criterion, which is the closest the platform comes to the five-minute checklist. Marketing → Templates (business/marketing/templates) stores message templates by channel with variables and a preview, useful for the wording of a message and holding no brand colors, no logo and no footer, so a message built there is not a released material until it has been through the check. Read together, the four routes hold the calendar and the wording of the routine, and they do not hold the brand.

What the platform does not do belongs in the same breath. There is no brand audit, no approval workflow, no saved brand kit that other pages read, no brand portal or guidelines publisher, no asset library with versions, usage rights or expiry dates, and no brand health tracking. The Branding page (business/branding) offers a logo drop zone, a brand name, a tagline, a website address, a primary and a secondary color with ten preset swatches, a choice among five fonts, an email footer and three social links, and it computes the contrast of the primary color with a pass or fail badge at the 4.5:1 threshold; its values are not saved, so it is a place to try a palette and see its contrast rather than a register of the brand. The asset folder, the approval log and the audit this chapter describes are therefore kept outside the platform, and Book J2 (Processes, SOPs and the Definition of Done) is where the routine becomes a written procedure with a definition of done.

Self-Check

1. Why does a brand drift even when nobody makes a mistake, and which four events produce the drift?
2. What are the four rhythms, how long does each take at Nordec, and what does each one produce?
3. What does the quarterly review do that the weekly check cannot, and why are the ten touchpoints drawn at random?
4. What are the four pieces of work in the annual review, and what three decisions can come out of it?
5. What is the path a new material follows, and who approves a change to the promise, to a template and to a supplier's proof?

Answers. 1. A brand drifts because it is published one material at a time, each one reasonable on the day it was made, and the four events that produce the drift are a new person who brings the habits of another business, a new document needed by Friday, an urgent job answered with an old file, and a supplier who improves an ambiguous file. 2. The weekly check of every new material before release takes thirty minutes and produces a checked material and a line in the approval log; the monthly reading of a sample of reports, visit sheets and social posts takes one hour and produces a short note; the quarterly spot-check of ten of the thirty-six touchpoints with a reading of the new reviews takes two hours and produces a one-page note of findings and fixes; the annual review takes about two days of the brand keeper's time and a half-day workshop and produces a dated audit, the awareness survey result, and a decision to fix, to update the guide or to brief the studio. 3. The quarterly review looks at the brand as a whole rather than at one material, and it reads the words customers use in their reviews for language that matches or contradicts the promise; the ten touchpoints are drawn at random because a business that checks the same ones every quarter checks the ones it already knows are fine. 4. The four pieces of work are the full audit of the thirty-six touchpoints, the awareness survey of sixty telephone calls, the review of the platform to ask whether the promise is still true, and the revision of the guide; the three decisions are to fix the touchpoint, to update the guide, or to brief the design studio. 5. The path is request, template or brief, draft, check against the guide, approval at the right level, release, and filing with its version; the promise is approved by the owner, a material built from a template by the brand keeper, and a supplier's proof by the brand keeper against the guide before production begins.

Chapter References

Charter of the French language, CQLR c C-11 (Québec). LégisQuébec, updated to June 10, 2026 — as read for Book C2 (Websites, Landing Pages and Conversion); its rules on trademarks, signs, products and business names are stated in this book through the explanations of the Office québécois de la langue française and of the Registraire des entreprises.

Competition Act, R.S.C. 1985, c. C-34, ss. 52 and 74.01 (Canada). Justice Laws Website, current to July 21, 2026 (last amended March 26, 2026) — as read for Book A6 (Designing a Value-Added Offer) on September 22, 2026.

La Machine. (2026). Documentation — route reference inventory. Internal product documentation, read September 25, 2026, together with the code of the pages this book describes: business/branding, business/marketing/templates, productivity/projects, productivity/tasks, productivity/meetings, productivity/calendar and /taches.

Wheeler, A., & Meyerson, R. (2024). Designing brand identity: A comprehensive guide to the world of brands and branding (6th ed.). Wiley.

Chapter 11 — Expertise, Roles and When to Get Help

Most of a small business's brand is made in-house by people who are not designers, and the skill worth learning is knowing which pieces to buy, how to brief the people who supply them, and what to keep.

What You'll Be Able to Do

• Name the seven roles behind a brand — strategist, namer, designer, writer, brand keeper, legal adviser and producer — whether they are held by seven people or by one.
• Decide which parts of the work the owner and the team keep in-house and which parts are bought, using the platform and the touchpoint inventory as the test.
• Brief a designer or a studio in two pages, covering the platform, what must be kept, the touchpoints, the deliverables, the budget, the timeline and who approves.
• Compare a freelance designer, an independent studio, an agency and a template-based design tool used by the owner on the same criteria, with the same questions about rights.
• Commission trademark work from a registered trademark agent, and recognize the moment when the question belongs to a lawyer instead.
• Secure the rights in work that is bought, through a written assignment of copyright, a waiver of moral rights, and the delivery of vector masters and editable templates.

Why It Matters

A brand is not built by a department, because a small business does not have one. It is built by a handful of roles that exist in every business whether or not anybody is named for them, and by a handful of suppliers who are chosen once and then, too often, forgotten until something needs reprinting. The reason this chapter exists is that the choice of suppliers is where most of the risk in a small brand actually sits, and it is not the risk most owners watch for. The visible risk is that the design will be disappointing. The hidden risk is that the business will not own what it paid for, will not be able to reprint it, will not be able to stop using it, and will discover all three at the worst possible moment.

The business followed through this book reached the question with a clear picture of what it had. Its inventory of thirty-six touchpoints found three versions of the logo in circulation, five different blues, four typefaces, and only thirteen of the thirty-six touchpoints consistent with one another, which is 36.1% of them. It had seven people, two vehicles, fifty-two buildings under contract, revenue of $37,251 a month and a run-rate of $447,012 a year. It had a platform from Chapter 2, names from Chapter 3 and a visual identity from Chapter 4, but none of it was yet attached to anybody's signature, anybody's folder or anybody's contract. That is the point at which expertise is either bought well or bought badly, and the difference is almost entirely made of paper: a brief, a proposal, a contract and a folder.

What Nordec bought and what it kept is worth setting down before the chapter goes any further, because the shape of it is common. It kept the platform, the approvals and the daily use of the templates. It bought an independent design studio on a fixed fee of $7,800 for the logo, the palette, the typography, eleven templates and the guide; a registered trademark agent for $1,200 of agent fees covering the review of a knockout search and an application in Nice class 37; vehicle wraps at 2 × $4,100; uniforms at $1,470; a website update at $2,900; a depot sign at $1,900; and printed items at $650. The whole rollout came to $24,120, which is 5.4% of the run-rate, one time, plus sixty-four hours of the administrator's time at $36, or $2,304.

In This Chapter

Section 11.1 sets out the seven roles behind a brand and shows which of them Nordec held in-house and which it bought. Section 11.2 examines the two roles that cannot be given away — the owner who owns the promise and the brand keeper who guards the materials day to day. Section 11.3 looks at brand strategists and naming specialists, at when a workshop of the owner's own is enough and at what a bought strategy should deliver. Section 11.4 compares the freelance designer, the independent studio, the agency and the template-based design tool, and looks at portfolios, references, contests and marketplaces. Section 11.5 covers trademark agents and lawyers, the search, the application, the fees and the moment when a dispute belongs to a lawyer. Section 11.6 looks at printers, sign makers, wrap shops and the other producers who turn files into objects. Section 11.7 sets out the brief, the comparison of proposals and the contract terms that keep the rights in the business. Section 11.8 names what a small business should never buy out, and what it will have to buy again at the next refresh.

11.1 The Roles Behind a Brand

A brand is made by roles, not by job titles, and the first useful thing a small business can do is name the roles it is already performing. Seven of them appear in every brand, from the largest corporation to a business of one person with a van. There is a strategist, who decides what the brand promises and to whom. There is a namer, who puts words on that promise for the market. There is a designer, who turns it into identifiers that can be recognized without reading. There is a writer, who decides how the business sounds when it speaks. There is a brand keeper, who holds the materials and checks that they are used the same way twice. There is a legal adviser, who secures the names and the marks. And there are producers, who turn the files into printed pages, wrapped vehicles, embroidered garments and signs. None of the seven is optional, and any of them can be held by a person who also does something else.

The strategy role at Nordec was held by the owner, and it was held deliberately. Sophie Nadeau ran two half-day workshops with the seven people in the business, using the twenty interviews of Book A6 as the input, and the platform that came out of those conversations fits on a page: keep commercial buildings running, with no surprises for the people who own them; the promise "We know your building, and we tell you ahead"; three values with the decision each one changes; four personality traits; and a tagline in French first, « Votre bâtiment, un an d'avance. ». That work was not bought, and it should not have been, because the person who has to live with a promise is the person who should make it. Chapter 2 treats the workshop in full; what matters here is that the role of strategist was performed, not that it was performed by an outsider.

The naming role was also held in-house, with one piece of bought help at the end. Twelve candidates were produced, seven passed a screen that checked whether the words worked in French and in English and whether they meant what Nordec wanted them to mean, and four survived a knockout search of the trademark databases. The choice was « Bilan Nordec » in French and "Nordec Building Check" in English, with the tiers taking French names first — Essentiel, Préventif, Portefeuille — and Essential, Preventive and Portfolio in English. The bought help was the search review that a registered trademark agent performed, and section 11.5 returns to it.

The design role was bought. An independent design studio, on a fixed fee of $7,800, produced the logo, the palette, the typography, eleven templates and a guide of twenty-four pages with a one-page summary. The chosen route keeps the letter N and draws a horizon line through it, the building seen a year ahead, and it drops the generic wrench that had been drawn on a template site four years earlier. The palette is Nordec Blue #1F3A5F, Signal Amber #E8A33D, Slate #5B6770, Mist #F4F6F8 and white, with white text on Nordec Blue at 11.48:1 contrast and Signal Amber carrying no text on white at 2.16:1.

The writing role stayed inside. The voice of Nordec has four principles — say the time, say the cost, say what comes next, no jargon without its plain word — and a glossary of twelve words to use and twelve to avoid. The brand keeper role belongs to the administrator, who holds the templates and the asset folder, checks every new material before it leaves the business, and keeps the supplier files. The legal adviser role was bought from a registered trademark agent for the search review and the application in Nice class 37. The producing roles were bought from four suppliers: a vehicle-wrap shop for two vans, a uniform supplier for seven people, a sign maker for the depot sign, and a printer for the printed items, with the web developer of the site working alongside them.

Figure 11.1 shows how those roles divide at Nordec. It is drawn as two columns: on the left, the people inside the business — Sophie, the brand keeper, the technicians and the dispatcher — and on the right, the suppliers who were bought. Between the columns run the two things that move in both directions during an engagement: what each supplier receives, which is a brief or a set of files, and what each supplier delivers, which is a route, a mark, a wrapped van, a printed sheet or a registered application.

Figure 11.1 — The roles behind Nordec's brand. The work that is bought comes with a written brief and leaves with the rights.

The table below lists the same roles one by one, with the person who holds each at Nordec, what that role delivers, and whether it is held in-house or bought. Read it as an assignment of responsibility rather than as an organization chart: the roles do not disappear when a business is small, and the only question a small business has to answer is who holds each one and whether the answer is written down.

Table 11.1 — Nordec's brand roles

The list is short, and that is the lesson of it. A business of seven people can hold four of the seven roles and buy three, and the four it holds are precisely the four that cannot be transferred: what the business promises, what it is called, how it sounds, and who checks the work. Everything else is a supplier relationship, which means it can be renegotiated, replaced or ended, provided the business kept the files and the rights.

DEFINITION Brand strategist, identity designer, brand keeper and trademark agent

A brand strategist is the person who decides what the brand promises, to whom, and at what it will refuse to be good; the work produces a platform the business can decide with, not a document it files. An identity designer turns that platform into identifiers that carry recognition without reading — a mark, a palette, a typographic choice, a set of templates — and is judged by whether the identifiers survive being applied a hundred times by people who are not designers. A brand keeper is the person inside the business who holds the master files, hands out the templates, checks each new material against the guide and keeps the record of what was produced by whom; the role is small, daily and decisive. A trademark agent is the professional who searches the registers, assesses the risk of confusion, files the application, answers the examiner and manages renewals. In Canada, only a registered trademark agent may act for an applicant before the registrar, and the regulator of that profession is not named in this book.

11.2 The Owner and the Brand Keeper

Two of the seven roles cannot be given away, and a business that tries to give either of them away pays for it later. The first is the owner's ownership of the promise. The second is somebody's daily guardianship of the materials. They are different in kind: the first is a decision, taken rarely and held for years; the second is a habit, performed in minutes and repeated every week. A business that confuses them ends up with a promise nobody remembers making and a folder nobody can find.

The promise belongs to the owner because the promise is a commitment the business can be held to. Nordec's promise, "We know your building, and we tell you ahead", is a sentence a customer can repeat, a technician can be measured against and a competitor can be compared with. If a supplier rewrites it, if a designer shortens it for a layout, if a printer sets it in a different order, the commitment the business makes to its customers has changed without the business deciding. That is why the rule at Nordec is simple and written down: Sophie approves the promise, the tagline and every identifier, and no supplier may alter any of them. The identifiers are the logo, the palette, the typography, the templates and the voice. A change to any of them is a change to how the promise is seen, and it is decided by the person who answers for the promise.

The approval rule is worth stating as two lists rather than as a principle, because a principle is easy to agree with and hard to apply on a Tuesday. The first list is what needs the owner's approval: the promise and the tagline; any new use of the logo; any new color or typeface; any new template; any change to a tier name or to the Bilan Nordec name; and any material that leaves the business carrying a claim the business has not made before. The second list is what does not: the use of an existing template with new content; the filling in of the monthly report; the printing of a document from a master; the sending of an email signature; the ordering of a reprinted item from a file that was approved. The purpose of the division is not to keep the owner busy. It is to keep the brand keeper fast, so that the check that protects the promise never becomes the reason a job is late.

The brand keeper is the role that makes the rest of this book survivable. At Nordec, the administrator holds the templates and the asset folder, checks every new material before it goes out, and keeps the supplier files: which studio produced which file, which printer holds the plate, which wrap shop has the vector master, which registration is in force and when it renews. The role is not creative and it is not senior, and it is the difference between an identity that holds for years and an identity that drifts back into three logos and five blues within eighteen months. Chapter 5 sets out the system the keeper maintains: twenty-one design tokens, eleven templates, a guide of twenty-four pages and a one-page summary. Chapter 10 sets out the rhythm in which the keeping happens, because a role without a cadence is an intention.

The time this takes is modest, and it should be budgeted in hours rather than left to goodwill. The rollout at Nordec consumed sixty-four hours of the administrator's time at $36 an hour, which is $2,304, on top of the $24,120 of bought work. After the rollout, the recurring work is the check on new materials, the ordering of reprints, the update of the supplier files and the review of what the next season needs; Chapter 10 places the whole cadence and its annual rhythm. A business that does not name the keeper will find that the templates quietly stop being used, that the two vans begin to look like two different companies, and that the first person to notice is a customer.

Two failure modes are worth naming so that a business can watch for them. The first is the keeper who becomes a bottleneck, holding every file until a person gives up and makes their own version. The remedy is the two lists above: the use of a template needs no approval, and only its change does. The second is the keeper who leaves and takes the folder with them, not out of malice but because the folder was in a personal account. The remedy is that the asset folder belongs to the business, is reachable by two people, and is checked at the annual review of Chapter 10 like any other asset. The keeper is a role, not a person, and the business should be able to hand it over on a Friday.

WHO DOES THIS The owner briefs, the brand keeper follows through

In a business of about ten people, the owner usually holds the strategy, the naming and the writing, and one other person holds the keeping. In a business of one, the same person holds everything, and the discipline that replaces a second person is a folder with an agreed structure and a rule that nothing leaves without being checked against the guide. At Nordec, the division is explicit. Sophie holds the platform, approves the promise and every identifier, sits in the first meeting with any new supplier, and approves the studio's three routes. The administrator holds the templates and the asset folder, checks every new material, orders the reprints, keeps the supplier file with the license records, and reports at the monthly pass of Chapter 10 anything that needed the owner's decision. The technicians and the dispatcher hold the part of the brand that no document can carry: the uniform, the van, the arrival on site and the words used at the door.

11.3 Brand Strategists and Naming Specialists

A brand strategist is bought for a disagreement, not for a document. That is the plainest way to decide whether the help is needed. If the seven people in the business agree on what the brand promises and on what it will refuse, the workshop is an afternoon and the output is a page; the owner can run it with the material Chapter 2 describes. If the seven people do not agree, if the question keeps returning to the same meeting without being settled, or if the answer changes depending on who is asked, then an outside facilitator buys something a document cannot supply: a process, a neutral voice in the room, and a deadline.

Nordec's own workshop is the case for keeping the work. Two half-days with the seven people, built on the twenty interviews of Book A6, produced a purpose, a promise, three values with the decision each one changes, four personality traits and a tagline. The words customers used about Nordec were already collected and counted: "on time" or "fast" from fifteen people, "knows my building" from eleven, "tells me ahead" from nine, "straight with me" from seven and "friendly" from five. The platform did not need an outside strategist to be discovered, because the evidence was already in the business and the person who could decide was already in the room. A small business with a clear market, a clear offer and a handful of interviews has most of what a strategist would ask for.

The situations that change the answer are specific. A move into a market the business does not know — a new territory, a new kind of customer, a different size of building — is a situation where the platform may need to be tested before it is committed. A merger or an acquisition brings two promises into one business, and the question of which one survives is a strategy question with legal consequences attached (Chapter 8). A platform that nobody agrees on after two attempts is a platform that needs a facilitated decision rather than a third attempt. And an architecture question — whether the entry offer should carry its own brand, which Chapter 6 answers for Nordec — is the kind of question where an outside view is cheap compared with the cost of being wrong twice.

What a bought strategy should deliver, and how to judge it, is easier to state than most engagements admit. The deliverable is not a slide deck; it is a page the business decides with. It should say what the business promises, to whom, and what it will refuse; it should give the decision rules that follow, in the form of concrete choices rather than adjectives; and it should be written in the words the customers use, not in the words of the category. A strategist who cannot reduce the work to a page has produced a document. A strategist whose page survives three months of real decisions has produced a platform. The test is not whether the owner likes the page, but whether the business has used it to refuse something.

Naming is a narrower purchase, and it has two halves that are easy to confuse. The first half is generation and preference: producing candidates, screening them for meaning in the languages the market speaks, listening to how they sound, and choosing. The second half is availability: searching the registers, assessing the risk of confusion, and filing. The first half can be done in-house, and at Nordec it was. The second half cannot, because it turns on a search of the trademark databases and on a professional judgment about the risk of confusion, which is the work of a registered trademark agent. A naming specialist who produces an elegant shortlist and does not arrange the search has delivered half of the work.

The rule that protects the money is therefore blunt: never pay to adopt a name before a knockout search has been run. Nordec's knockout search found two registered marks close to "Nordec" in unrelated classes of goods, which meant the risk existed but was manageable in the class the business actually occupies, building maintenance and repair services. If the search had found a live mark in that same class, the four candidates that survived the language screen would have been four names the business could not use, and the money spent choosing among them would have been spent on a decision that had already been made by the register. Chapter 3 sets out the naming method; this chapter's point is only about the order of the spending.

KEY CONCEPT Buy expertise, keep ownership

The central idea of this chapter is that buying expertise and owning the result are two separate acts, and that a small business must perform both. The first act is the engagement: a brief, a supplier, a fee and a deliverable. The second act is the transfer: the master files, the editable templates, the color codes, the written assignment of copyright and the waiver of moral rights. A business that performs only the first has rented its identity for as long as the supplier chooses to be available, and it will discover the arrangement when it needs a reprint, a new van or a second language. A business that performs both can change supplier at any time, because what the supplier made is now an asset of the business. The test is simple and it can be applied before any money moves: at the end of this engagement, will the business hold the files it needs to produce the next piece without asking anybody?

11.4 Graphic Designers, Studios and Agencies

Four options are open to a small business that needs an identity designed, and they differ less in talent than in shape. There is the freelance designer, one pair of hands with a portfolio. There is the independent studio, a small team with a process. There is the agency, a larger organization with more layers and a broader scope. And there is the template-based design tool used by the owner, which is not a designer at all but is an option that many businesses take, and it deserves to be described honestly rather than dismissed. The right question is not which one is best, but which one fits the scope of the work, the size of the business and the way the identity will be used.

A freelance designer suits a defined scope that one person can carry: a logo refinement, a template, a set of print pieces, a document that has to look like the rest of them. The advantage is directness, because the person who designs is the person the business speaks with, and the decisions are taken in one conversation. The risk is capacity: a single person has one calendar, and a business whose reprint depends on one calendar is exposed to illness, vacation and a busy month. A freelance designer is also the option most likely to be engaged without a contract, which is how businesses acquire a logo they cannot prove they own.

An independent studio suits the scope Nordec had: a whole identity, produced once, in a defined number of weeks, with more than one pair of eyes on it. The studio Nordec chose answered the same two-page brief as two others, and it proposed three routes, eleven templates, two rounds of revisions and the transfer of the rights, on a fixed fee of $7,800. What the studio supplied that a single designer might not is a process: routes to choose between, templates to apply, a guide to follow, and a named moment at which the files and the rights change hands. The trade is coordination, because there is a small team to brief rather than a person, and the brief has to be good enough to survive being read by more than one of them.

An agency suits a scope that goes beyond the identity: a launch with a campaign attached, media to buy, several markets, or a business that wants one supplier for everything. The advantage is that the pieces are coordinated, and the disadvantage is that the identity may be designed to serve the campaign rather than the other way round, with the layers between the owner and the work multiplying. For a business of seven people, an agency is usually the wrong shape for the identity itself, and may be the right shape for a later moment when the identity has to be carried into a market the business does not yet know.

The fourth option is the template-based design tool used by the owner, and it is worth being precise about what it can and cannot do. It can produce a first logo in an evening, in a business with no identity and a van that needs something on its side. What it cannot do is make that logo exclusive: the templates and the elements offered to one user are offered to the others, and a logo assembled from shared elements may be held by another business somewhere, in the same market or in another one. A business that uses this route should treat the result as a temporary identifier, use the tool's own license terms in writing, and plan the day the real identity is designed. Nordec's first logo was drawn this way in 2022, on a template site, and by the time it was replaced there were three versions of it in circulation.

Portfolios and references are how the choice is actually made, and four questions do most of the work. The first is whether the supplier has done work for a business like this one, in the same category, at the same size, with the same constraints on production. The second is what files were delivered: a supplier who shows only finished photographs of printed pieces has not shown that the vector masters and editable templates exist. The third is who owns the rights in the work shown, which is a question most portfolios cannot answer and every contract should. The fourth is what the process was: how many routes, how many rounds of revision, who approved and when the files changed hands. A supplier who can answer all four in a first conversation is a supplier who has done this before.

The table below compares the four options on four criteria: what each one suits, what it typically delivers, the main risk attached to it, and the rights question that has to be checked. Read it as a filter rather than a ranking, because the right answer for a business of one with a new van is not the same as the right answer for a business with fifty-two buildings under contract, a fleet and a French-language market.

Table 11.2 — Choosing who designs the identity

WATCH OUT The logo you paid for but do not own

The most expensive line in a small brand is often invisible on the invoice: the work was paid for, and the copyright was never assigned. By default, the person who creates a work is its first owner, so a logo drawn by a supplier belongs to that supplier unless the business has a written assignment signed by the owner of the right. The consequence is not theoretical. A business without the assignment cannot stop the supplier from reusing the work, may be limited in how it can modify the mark it commissioned, and may find that the files it needs to reprint sit with somebody it no longer speaks to. Moral rights add a second layer: the author's right to the integrity of the work and to be associated with it cannot be assigned, and can only be waived, so a work whose author has not waived them can be defended even after the copyright has moved. Both points are settled by a single paragraph in the contract, agreed at signature, for free.

CASE IN POINT Nordec Entretien inc.: three answers to one brief

Nordec Entretien inc. is a fictional business created for this encyclopedia; its owner, its employees, its customers, its competitors, its figures and the events described are invented, and any resemblance to a real business is coincidental. Nordec sent the same two-page brief to three studios and received three different answers. The first answered with a price only, which told the business that the brief had not been read as a brief. The second proposed a full rebrand including a new name, which was a serious answer to a question Nordec had already settled: the name was kept, and the platform of Chapter 2 had decided it. The third answered the brief as written, with three routes, eleven templates, two rounds of revisions and the transfer of the rights, on a fixed fee. Nordec chose the third, and the choice was not about the lowest price. It was the only proposal that named what would be delivered, when the business would approve it and who would own it afterwards.

11.5 Trademark Agents and Lawyers

Legal support for a brand arrives in two forms, and they are bought at different moments. The first is a registered trademark agent, engaged while the business is still deciding what to file. The second is a lawyer, engaged when something is contested or when a document has to be drafted that the business intends to rely on. Most small businesses need the first once and the second rarely, and the mistake is to buy neither until a problem has already appeared.

A registered trademark agent does four things that matter to a small brand. The first is a search and a written opinion on availability, which is a judgment about the risk of confusion rather than a list of results. The second is the application: identifying the goods and services in ordinary commercial terms, grouping them by class, and filing through the appropriate channel. The third is the correspondence with the examiner, because applications are frequently met with an objection that has to be answered with an argument rather than a payment. The fourth is the management of the registration afterwards: the renewal dates, the record of what is covered, and the watch for anything that threatens it. At Nordec the agent reviewed the knockout search and filed the word mark NORDEC in Nice class 37, for building maintenance and repair services, for $1,200 of agent fees, an illustrative figure of this case.

The knockout search and the comprehensive search are different purchases and should not be confused. A knockout search is the quick screen that removes the candidates that cannot be used, and it is what took Nordec's twelve candidates down to seven on language and meaning and then to four on availability. It found two registered marks close to "Nordec" in unrelated classes of goods, which is a risk to record rather than a reason to stop when the business's own class is clear. A comprehensive search is a wider piece of work, covering the registers, the usage of marks in the market and the international sources that could bear on the question. The United States Patent and Trademark Office does not search for the applicant and recommends a clearance search of its own database, state registries, use on the internet and international sources, so that refusals, oppositions and infringement suits are avoided; the principle is the same on this side of the border.

The official fees are published and they change, and a business should budget from the published page rather than from a memory of what somebody paid. Filing online in 2026 costs $491.06 for the first class of goods or services and $149.04 for each additional class, and those amounts rise on January 1, 2027 to $499.41 and $151.57. A renewal in 2026 costs $595.06 for the first class and $185.49 for each additional class, rising to $605.18 and $188.64 in 2027. There is no separate registration fee for applications filed after June 17, 2019, and a registration lasts ten years from the date of registration and is renewable for further periods of ten years, which is the Trademarks Act, R.S.C. 1985, c. T-13, s. 46(1), as the book states it. The book explains these rules and does not replace the advice of a lawyer or a trademark agent.

A lawyer is needed when the question stops being administrative. A dispute over a mark, a license granted to another business, a cease-and-desist letter received or contemplated, a coexistence agreement with a business whose name is close to this one, or a contract that will be relied on in a dispute: those are legal questions rather than filing questions, and they belong with the lawyer who knows the business's other agreements. Book H5 (Intellectual Property) treats the intellectual property questions in full, and Book H1 (Contracts and Commercial Agreements) treats the drafting of the agreements themselves. The division of labor with the brand keeper is worth writing down: the agent holds the register and the dates, the brand keeper holds the certificates and the supplier licenses in the asset folder, and the owner decides what the business will spend to defend.

WHAT CHANGED Trademark fees for 2026 and 2027

The Canadian Intellectual Property Office publishes its trademark fees and revises them, which makes the filing decision a dated one rather than a permanent one (Canadian Intellectual Property Office, 2026a). In 2026, filing an application online costs $491.06 for the first class of goods or services and $149.04 for each additional class. On January 1, 2027, those amounts rise to $499.41 and $151.57. A renewal filed online in 2026 costs $595.06 for the first class and $185.49 for each additional class, and those amounts rise in 2027 to $605.18 and $188.64. There is no separate registration fee for applications filed after June 17, 2019. The practical consequence for a small business is not that the increase is large; it is that the cost of a filing is knowable in advance, published and dated, and that a decision to file is therefore a decision that can be planned rather than a surprise that arrives with the invoice.

11.6 Printers, Sign Makers and Other Suppliers

Producers are the suppliers who turn files into objects, and they are the point at which a brand either becomes visible or quietly drifts. The designer's work exists as a set of files until a van is wrapped, a garment is embroidered, a sign is mounted or a page is printed. Every one of those steps is performed by somebody who is not a designer, working from a file, under a deadline, with the materials they have in the shop. The identity that survives that journey is the identity that was specified properly at the start, and the specification is a short list of things the business has to hand over and check.

The vehicle-wrap shop is the producer with the largest visible surface at Nordec: two vans at $4,100 each, $8,200 in total, which is the single biggest line of the rollout after the design itself. A wrap is produced from a vector master, and it has to be adapted to the shape of the vehicle rather than scaled to it; a door handle, a wheel arch and a rear panel each break the layout differently. The file a wrap shop needs is a vector format, with the color codes stated by name and by code, and with the layout adapted to the vehicle template. The check is the first van: view it from ten metres, from three metres and from the side mirror, in daylight and in the rain, because a stripe that reads well in the shop can disappear at a red light.

The uniform supplier produces the identity seven people wear, at 7 people × 5 garments × $42, which is $1,470. Textiles take color differently from a screen, and a blue that is right on a monitor can arrive on a shirt as something else; the practical answer is a physical sample before the run, approved by the brand keeper and kept as the reference. Nordec's amber is used for accents and safety details rather than for large panels or text, because amber text on white measures 2.16:1, which is far below what a reader needs. That rule travels with the order to the supplier, which is the point: a specification that exists only in the guide is not a specification.

The sign maker produces the depot sign at $1,900, and the printer produces the printed items at $650. Both work from print-ready files, and both will produce a proof if asked. The proof is the cheapest quality control the business has: a printed sheet, a mounted sample or a scaled photograph of the sign in place, checked before the full run is produced. The check has four items on it, and they are always the same four. The color, compared with the approved sample rather than with a memory. The spelling, including the accents on a French name. The size of the logo relative to the surface, because a logo that is correct at one size can be unreadable at another. And the placement of the tagline, since « Votre bâtiment, un an d'avance. » is part of the mark rather than a caption to be trimmed for space.

The files a producer receives carry a licensing question that most small businesses never ask. Nordec's typography uses two open-license faces, and the license under which they are released permits use, study, copying, embedding, modification and redistribution of the font software, with selling permitted only when it is bundled with other software, and it reserves the font names against modified versions (SIL International, 2007). The important half of the rule for a brand is the other half: documents and other works made with the fonts are not covered by the license, so a logo set in such a font is not itself under it. That is why the contract with a designer should say that any font or image used in the work is licensed for the use the business intends, and why the record of those licenses belongs in the brand keeper's folder. Images carry the same question in a different form, and the answer there is a record of the license, its address and the attribution text.

The last supplier risk is the one that does the most damage over time, and it is the supplier who improves the file. A wrap shop that redraws the mark to fit a panel, a printer who substitutes a similar blue because the exact one is out of stock, a sign maker who adjusts the spacing to fit the plate: each of those is a small kindness that produces a new version of the identity in circulation. Nordec reached the design studio in Chapter 4 with three versions of its logo and five different blues in the market, and none of them had been created deliberately. The remedy is a sentence in the purchase order and a check on the first piece: the file is used as supplied, and any change is approved by the brand keeper before production.

TOOLS OF THE TRADE What a design engagement needs

Five things carry a design engagement from the first conversation to the last reprint, and none of them is a product. The first is a brief template: a document with fixed headings, so that the platform, the constraints, the deliverables, the budget, the timeline and the approvals are stated the same way for every supplier who is asked. The second is a comparison grid: one row per proposal and one column per criterion, so that three answers to one brief can be read side by side rather than compared from memory. The third is a contract with a rights clause, naming the deliverables and their formats, the number of revision rounds, the assignment of copyright on final payment, the waiver of moral rights and the warranty of originality. The fourth is a shared folder for masters and proofs, holding the vector masters, the editable templates, the approved color codes, the approved samples and the proofs as they were approved. The fifth is a supplier record, one line per supplier with what they produce, which file they hold, which license applies and when the last order was placed.

11.7 Buying Help: Briefs, Deliverables and Rights

Buying help well is a matter of three documents, and a business that produces all three will get a better result at a lower price than a business that produces none. The first is the brief, which states the problem once and sends it to everybody. The second is the comparison of proposals, which reads the answers against the same criteria. The third is the contract, which says what will be delivered, when it will be paid for and who will own it. The three are cheap to produce and they are the whole difference between commissioning work and hoping for it.

The brief Nordec sent was two pages, and its shape is worth copying because it answers every question a supplier would otherwise ask in a meeting. It opens with the platform of Chapter 2 in five lines: the purpose, the promise, the values with the decisions they change, the personality traits and the tagline. It states what must be kept: the name Nordec, a blue family that the market already associates with the vans, and the letter N. It lists the thirty-six touchpoints found in the inventory of Chapter 1, with the ones that must be produced first. It states the deliverables — a logo, a palette, a typographic choice, eleven templates and a guide — and it states the budget, the timeline of six weeks and the person who approves. Chapter 4 describes the routes that came back; the point here is that the brief is what made them comparable.

Comparing proposals is easier than most owners expect, because the criteria are the same every time. What is being delivered, item by item, including the formats. How many routes and how many rounds of revision are included, and what happens after the last round. What the timeline is, and what the business has to supply for it to hold. What the price covers, and what it does not. And what the business will hold at the end: the master files, the templates, the color codes, the guide, the copyright and the waiver. A proposal that answers four of those five in writing is a proposal that can be accepted; one that answers two is a conversation. The process of building an identity, from the research that precedes it to the governance that follows it, is described as five phases — research, strategy, design, launch and governance — in a reference work on the subject, and the last of those phases is a phase of the work rather than an afterthought (Wheeler & Meyerson, 2024). That is the argument for putting the ownership and handover terms in the contract at the beginning: governance is what the engagement is built towards.

The contract terms that matter for a small brand are six, and they can be agreed in a page. The first is the deliverables and the file formats: vector masters, color codes stated by name and by code, and the templates in editable form rather than as flat images. The second is the number of revision rounds, and what a round is, because a studio and an owner can disagree about that word at the worst possible moment. The third is the assignment of copyright on final payment, in writing and signed, with a statement that the business may use, modify and reproduce the work without further payment. The fourth is the waiver of moral rights, in whole or in part, given by the author. The fifth is the supplier's warranty that the work is original and that any font or image used is licensed for the use the business intends. And the sixth is the right to show the work in the supplier's portfolio, which costs the business nothing and is usually worth granting.

The copyright points behind those terms are set out in the Copyright Act, R.S.C. 1985, c. C-42, as the book states it. The author of a work is its first owner, under s. 13(1). Where a work is made by an employee in the course of employment, the employer is the first owner unless there is an agreement to the contrary, under s. 13(3), which is why a business's own employees can produce material the business owns without a separate assignment and a commissioned supplier cannot. An assignment or grant of copyright is valid only if it is in writing and signed by the owner of the right, under s. 13(4), which is the clause that turns a paid invoice into an owned asset. And moral rights — the integrity of the work and the right to be associated with it — cannot be assigned but can be waived, in whole or in part, under s. 14.1, and an assignment of copyright does not by itself waive them. This book explains these rules and does not replace the advice of a lawyer or a trademark agent.

Owning the master files is the practical half of the same idea, and it is worth stating what a business should hold at the end of an engagement. The vector masters of the mark, in the versions the guide specifies, including the small-size version and the one-color version. The palette, with each color named and given its code, as in Nordec Blue #1F3A5F, Signal Amber #E8A33D, Slate #5B6770 and Mist #F4F6F8. The typefaces and the record of their licenses. The templates in editable form, in the formats the business actually uses. The guide, with its twenty-four pages and its one-page summary. And the signed assignment, kept with the certificates, because it is the document that makes the rest of the folder useful.

METHOD Writing a two-page design brief

First, put the platform in five lines: the purpose, the promise, the values with the decision each one changes, the personality traits and the tagline. Second, state what must be kept, because a brief without constraints invites a proposal for a different business; Nordec's brief named the name, a blue family and the letter N. Third, list the touchpoints and the deliverables, from the thirty-six touchpoints of the inventory, with the ones that must be produced first and the formats they have to arrive in. Fourth, state the budget, the timeline and who approves: a fixed fee or a range, six weeks or twelve, and the name of the single person who will answer within a day. Fifth, state what the business expects to hold at the end — the vector masters, the editable templates, the color codes, the guide, the written assignment of copyright and the waiver of moral rights — so that the ownership terms are discussed before the work begins rather than after it is finished.

THE RULE Put the assignment in writing

Under the Copyright Act, R.S.C. 1985, c. C-42, s. 13(4), an assignment or grant of copyright is valid only if it is in writing and signed by the owner of the right, and under s. 14.1 moral rights cannot be assigned but can be waived, in whole or in part, and an assignment of copyright does not by itself waive them. Two consequences follow for a business buying design work, and they cost nothing to arrange. The first is that the contract must contain an assignment of copyright signed by the supplier, with a date, describing the work it covers. The second is that it should contain a waiver of moral rights, so that the business may modify its own mark without the author's consent, which otherwise remains a right the author can exercise. Section 13(1) puts the author as the first owner by default and s. 13(3) makes the employer the first owner of work made by an employee, which is why commissioned work needs the clause and employed work usually does not. This book explains the law and does not replace the advice of a lawyer or a trademark agent.

QUÉBEC & CANADA NOTE Rights in a commissioned logo

The federal rules apply in Québec as elsewhere in Canada, and they are the reason a commission is not a purchase until the paperwork says so. The Copyright Act, R.S.C. 1985, c. C-42, provides at s. 13(1) that the author of a work is its first owner, and at s. 13(4) that an assignment or grant of copyright is valid only if it is in writing and signed by the owner of the right. The practical effect is that a logo drawn by an independent studio belongs to the studio unless the business holds a signed assignment, however much was paid for it. Section 14.1 adds that moral rights cannot be assigned but can be waived, in whole or in part, and that an assignment of copyright does not by itself waive them, so a business that wants to modify its own mark over the years should obtain the waiver in the same document. A small business commissioning an identity should therefore have two sentences in the contract, agreed before the first route is drawn, and should keep the signed document with its registration certificates. This book explains the law and does not replace the advice of a lawyer or a trademark agent.

11.8 What You Keep In-House

The dividing line between what a small business buys and what it keeps is drawn by a single test: is this a decision the business must be able to make again by itself, or is it a piece of work that is produced once and then used? The decisions stay inside — what the brand promises, what it is called, how it sounds, what may carry the logo, and who checks each new material. The production is bought — the identity design, the trademark work, the wrapped vans, the printed pages, the sign and the website build. A business that buys a decision loses the ability to make it again, and a business that keeps production ends up with work it cannot produce at the quality the market sees.

Four things are kept, and each has a reason that is practical rather than sentimental. The platform is kept because Chapter 2's work only pays if the business can use it to refuse something, and the person who must use it is the owner. The approvals are kept because the identifiers are the promise in visible form, and a supplier cannot decide what the business is committing to. The use of the templates is kept because the daily production of the monthly report, the visit sheet and the capital planning note has to happen without a supplier in the loop, or it will not happen. And the voice is kept because the words customers hear on the telephone, at the door and in the email are spoken by the seven people in the business, not by anybody who can be briefed.

Three things are bought once and used for years. The identity design is bought once, and Nordec paid $7,800 for it. The trademark work is bought once, with $1,200 of agent fees and the official filing fee, and then renewed every ten years. The templates are built once, in the middle of that engagement, and the business edits them from then on. What is bought again is the production: the wraps when a van is replaced, the uniforms as garments wear out, the printed items each time the stock runs low, the sign when the depot moves, the website when the offer changes. A business should be able to tell the difference at a glance, and Chapter 10's cadence is where the recurring purchases are seen coming.

There is one more thing to keep, and it is the record. The asset folder, the guide, the signed assignment, the certificates, the licenses for the typefaces and the images, the approved color samples, and the list of which supplier holds which file: this is what makes the difference between a brand the business owns and a brand it rents from the people who made it. A business that keeps the record can change studio, printer or wrap shop without renegotiating its identity. A business that does not keep the record discovers, at the moment it needs a reprint, that the only person who can produce the file is the person it has stopped paying.

The arithmetic of the rollout shows where the money actually goes, and it is worth reading once before a business plans its own. The calculation below is not a table but a sum, and it ends with a lesson that surprises most owners: designing the identity is the smaller part of the cost, and making it visible is the larger one.

DO THE MATH What Nordec bought and what it kept

Illustrative figures; the case is fictional. Step 1 — The work bought outside, in the rollout budget of Chapter 7: the design studio $7,800, the trademark agent $1,200, the vehicle wraps 2 × $4,100 = $8,200, the uniforms $1,470, the website update $2,900, the depot sign $1,900 and the printed items $650, for a total of $24,120. Step 2 — The work kept in-house: 64 hours of the administrator's time at $36, which is $2,304, bringing the rollout to $26,424 in cash and internal time. Step 3 — The studio's share of the bought work is $7,800 ÷ $24,120 = 32.3%, so designing the identity accounts for roughly a third of what was spent. Step 4 — The production share, being the wraps, the uniforms, the sign and the printed items, is $8,200 + $1,470 + $1,900 + $650 = $12,220, which is $12,220 ÷ $24,120 = 50.7%. Step 5 — The lesson is in the two percentages read together: the identity is the smaller cost and the visibility is the larger one, so a business that spends its whole budget on design and none of it on production has paid for a brand nobody sees.

IN LA MACHINE What the platform does and does not do for suppliers

The application copy read for this chapter gives the supplier side of a brand three places to live (La Machine, 2026). The Gallery at business/marketing/gallery is a library of royalty-free stock images that records, for each image, its license, the license's address and the attribution text, and it adds a provenance form — source, address, license, attribution — for the business's own images, which is the closest the platform comes to a license register. Companies at business/marketing/companies and Contacts at business/marketing/contacts hold one record per customer and the people attached to it, with notes, so a design studio, a wrap shop or a printer can be kept as a company record with its contacts. Projects at productivity/projects can carry the design engagement with its milestones, and Tasks can carry the steps that follow it. What the platform does not have is the thing this chapter's folder actually needs: there is no brand portal or guidelines publisher to share the guide with suppliers, and no digital asset management with versions, usage rights or expiry dates. The record therefore lives outside the platform, and the supplier lines live inside it.

The reason to keep the record outside the platform is not a shortcoming of the platform; it is that the record answers a different question. The Gallery answers what a license says about an image. The supplier file answers who holds which file, which license applies to which typeface, what was approved at which proof, and when the registration renews. Those are the questions that decide whether a business can produce its next van, its next print run and its next sign without asking anybody's permission.

The last point is the one to carry out of the chapter. Expertise is bought in small, defined pieces, and every one of those pieces is a supplier relationship with a beginning, an end and a set of rights attached. A business that writes the brief, compares the proposals, signs the assignment and keeps the folder has bought everything it needs and owns all of it. A business that skips any one of the four has paid for work it may not be able to use, which is the most expensive outcome available in branding and the easiest one to avoid.

Costly Mistakes

• Commissioning a logo without a written assignment of copyright. The work is paid for and still belongs to the person who drew it, and the business discovers the arrangement when it needs a reprint, a modification or a second language.
• Paying to choose a name before a knockout search. Money spent selecting among candidates that a register has already ruled out is money spent on a decision that was never the business's to make.
• Sending a brief that describes the outcome instead of the problem. A brief that asks for a look invites three unrelated answers, and the comparison of proposals becomes a matter of taste rather than of scope.
• Choosing a supplier on the portfolio alone. A portfolio shows finished photographs, not the vector masters, the editable templates, the revision rounds or the rights, and those are what the business will need in eighteen months.
• Using a shared template as a permanent logo. Elements and templates offered to one user are offered to the others, so the mark may be held by another business in the same market.
• Letting a producer improve the file. A wrap shop, a printer or a sign maker who redraws the mark to fit a surface creates a new version of the identity, and version drift is how three logos and five blues end up in circulation.
• Leaving the asset folder in one person's account. The brand keeper is a role rather than a person, and a folder that cannot be reached by a second person stops being the business's the day that person leaves.

Key Takeaways

• Seven roles stand behind every brand — strategist, namer, designer, writer, brand keeper, legal adviser and producer — and a small business holds four of them and buys three.
• Two roles cannot be given away: the owner's ownership of the promise, and the brand keeper's daily guardianship of the templates, the asset folder, the checks and the supplier files.
• A brand strategist is bought for a disagreement rather than for a document, and Nordec's own two half-day workshops were enough because the evidence was already in the business and the person who could decide was already in the room.
• The four ways to have an identity designed — a freelance designer, an independent studio, an agency and a template-based design tool used by the owner — differ less in talent than in shape, and the rights question has to be asked of all four.
• A registered trademark agent runs the search, files the application, answers the examiner and manages the renewal; filing online in 2026 costs $491.06 for the first class and $499.41 from January 1, 2027 (Canadian Intellectual Property Office, 2026a).
• The assignment of copyright is valid only if it is in writing and signed by the owner of the right, and moral rights cannot be assigned but can be waived, so both sentences belong in the contract before the first route is drawn.

Your Action Plan

1. Name the seven roles and write down who holds each one. Set out the strategist, namer, designer, writer, brand keeper, legal adviser and producer roles, mark each as in-house or bought, and name the person who answers for it. Deliverable: one page listing the roles and the holders. Verification: every bought role has a named supplier or a dated plan to find one, and every in-house role has a named person.
2. Write the two-page brief before speaking to any supplier. Put the platform in five lines, state what must be kept, list the touchpoints and the deliverables with their formats, and state the budget, the timeline and the single approving person. Deliverable: a two-page brief. Verification: the same brief can be sent unchanged to three suppliers and read as a comparable question by each.
3. Compare three answers on one grid and check the rights in every proposal. Read what is delivered, how many routes and revision rounds are included, the timeline, the price and what the business will hold at the end. Deliverable: one comparison grid with three rows. Verification: each row names the master files, the editable templates, the assignment of copyright and the waiver of moral rights.
4. Put the assignment and the waiver in the contract before the first route is drawn. Add the deliverables and formats, the revision rounds, the assignment on final payment, the waiver of moral rights, the warranty of originality and the portfolio right. Deliverable: one signed contract with a rights clause. Verification: the signed assignment is filed with the registration certificates and reachable by two people.
5. Open the asset folder and hand it to the brand keeper. Create the folder for the vector masters, the editable templates, the color codes, the approved samples, the proofs, the licenses and the supplier file, and name the person who checks every new material. Deliverable: one asset folder with an agreed structure and a named keeper. Verification: the keeper can produce the vector master, the approved color and the license record for the last three materials the business produced.

In La Machine

Three routes in the application copy read for this book touch the supplier side of a brand directly (La Machine, 2026). The Gallery at business/marketing/gallery is a library of royalty-free stock images that records, for each image, its license, the license's address and the attribution text, and adds a provenance form for the business's own images with the source, the address, the license and the attribution — which is the part of a license register a small business is most likely to need and least likely to keep. Companies at business/marketing/companies and Contacts at business/marketing/contacts hold one record per customer and the people attached to it, with notes, so the design studio, the wrap shop, the uniform supplier, the printer and the sign maker can each be held as a company record with their contacts and the history of what they produced. Projects at productivity/projects can carry the design engagement itself, with its milestones, and the tasks that follow it. Book N1 maps the platform's modules.

What the platform does not do belongs in the same breath, because a business should not look for it twice. There is no brand portal and no guidelines publisher, so the guide cannot be shared with a supplier from inside the platform. There is no digital asset management with versions, usage rights or expiry dates, so the asset folder and the supplier file live outside it. There is no brand consistency audit, no brand health tracking and no trademark search. The four routes above are places to keep the record of suppliers, licenses and engagements; they are not the record itself, and the record is a folder the business owns and a person the business has named.

Self-Check

1. What are the seven roles behind a brand, and which of them can a small business not give away?
2. When is a brand strategist worth buying, and what should the deliverable be?
3. What are the four ways to have an identity designed, and which question about rights has to be asked of all four?
4. What does a registered trademark agent do, and at what point does the question belong to a lawyer instead?
5. What must the contract with a designer say about the files and the rights, and which sections of the Copyright Act stand behind those clauses?

Answers. 1. The seven roles are the strategist, the namer, the designer, the writer, the brand keeper, the legal adviser and the producer; the two that cannot be given away are the owner's ownership of the promise and the brand keeper's daily guardianship of the templates, the asset folder, the checks and the supplier files. 2. A strategist is worth buying for a disagreement — a move into an unknown market, a merger or acquisition, a platform nobody agrees on, or an architecture question — and the deliverable should be a page the business decides with, written in the customers' words, rather than a slide deck. 3. The four ways are a freelance designer, an independent studio, an agency and a template-based design tool used by the owner; all four must be asked whether the copyright is assigned in writing, whether moral rights are waived and whether the working files are delivered, because a template-based tool's license is not an assignment. 4. A registered trademark agent searches the registers and gives an opinion on the risk of confusion, identifies the goods and services in ordinary commercial terms and files the application, answers the examiner's objections and manages the renewal dates; a lawyer is needed when a dispute arises, when a license is granted, when a cease-and-desist letter is received or contemplated, or when an agreement has to be drafted that the business will rely on (Book H5). 5. The contract must name the deliverables and their formats — vector masters, color codes stated by name and by code, and editable templates — the number of revision rounds, the assignment of copyright on final payment, the waiver of moral rights, the supplier's warranty that the work is original and that any font or image is licensed for the intended use, and the portfolio right; s. 13(1) makes the author the first owner, s. 13(4) makes an assignment valid only in writing and signed by the owner of the right, and s. 14.1 provides that moral rights cannot be assigned but can be waived.

Chapter References

Canadian Intellectual Property Office. (2026a). Fees for trademarks (page modified July 31, 2026). Retrieved September 25, 2026, from https://ised-isde.canada.ca/site/canadian-intellectual-property-office/en/trademarks/fees-trademarks

Copyright Act, R.S.C. 1985, c. C-42, ss. 13 and 14.1 (Canada). Justice Laws Website, current to September 3, 2026 (last amended November 7, 2024) — read September 25, 2026.

La Machine. (2026). Documentation — route reference inventory. Internal product documentation, read September 25, 2026, together with the code of the pages this book describes: business/marketing/gallery, business/marketing/companies and productivity/projects.

SIL International. (2007). SIL Open Font License (Version 1.1, February 26, 2007). Retrieved September 25, 2026, from https://openfontlicense.org/open-font-license-official-text/

Wheeler, A., & Meyerson, R. (2024). Designing brand identity: A comprehensive guide to the world of brands and branding (6th ed.). Wiley.

Chapter 12 — Tools: Functions and Popular Tools

Branding tools do four jobs — create, store and share, protect, and measure — and a small business chooses them by the job to be done and by what its platform already covers, not by the length of a feature list.

What You'll Be Able to Do

• Name the four jobs a branding tool can do — create, store and share, protect and measure — and place any tool in front of you in one of them.
• Tell a master file from an export, and a template from a master, and say which tool each one belongs in.
• Choose between a shared folder, a brand portal and a digital asset management system by the size of the asset library and the number of people who touch it.
• Search the official trademark databases before a name is adopted, and read a search result as information rather than as permission.
• Measure what a small business can learn about its brand with a form, a telephone and the reviews its customers already write, as Chapter 9 sets out.
• Decide what to buy and what to leave alone by starting from the job, then from what the platform already covers, and only then from the features on offer.

Why It Matters

A brand is a set of decisions, and no tool makes a decision. What tools decide is which decisions are easy to repeat. A palette of five blues is not a design failure; it is what happens when five people each have their own copy of a file and no place to look up the right one. A logo that changes shape between a proposal and a van is not a matter of taste; it is what happens when the master file lives on the computer of a designer who is no longer answering messages. The identity a business can actually maintain is the identity its tools let it repeat, which is why the choice of tools belongs in a book about the brand at all.

The market in front of a small business is loud and crowded, and almost all of it is sold by feature. Tools advertise what they can do rather than what they are for, and a business that buys by feature list ends up with four tools that each do part of one job and none of them well. The remedy is a short set of questions asked in a fixed order: which job is this for, what does the platform already cover, who will open this tool every week, and how does the business get its files out again. A tool that survives those questions is usually modest, and modest tools get used.

The case followed through this book arrived at the question from a measurement rather than from an opinion. An inventory of Nordec's thirty-six touchpoints found three versions of the logo in circulation, five different blues, four typefaces, and only thirteen of the thirty-six touchpoints consistent with one another, or 36.1%. Nothing had failed in an obvious sense; each person had simply solved their own problem with whatever was at hand. What changed afterwards was not a larger number of tools but a smaller and better placed set: master files held by the design studio and handed over, eleven editable templates, a shared asset folder with one folder per identifier and a folder for retired files, and one person — the administrator, as brand keeper — who holds the assets and checks every new material. The whole set was already partly covered by the platform the business paid for, and the parts that were not covered were done by telephone, by a form and by the official registers.

In This Chapter

Section 12.1 sets out the four jobs a branding tool can do and maps the families of tools underneath them. Section 12.2 covers the tools that create the identity — vector drawing, layout and image editing — and the files and color codes they produce. Section 12.3 covers templates and self-service design, and the table of popular tools by family. Section 12.4 covers the shared folder, the brand portal and digital asset management, with versions, rights and expiry dates. Section 12.5 covers name and trademark searching, from the official databases to domain and handle checks. Section 12.6 covers the tools of measurement: surveys, reviews and listening. Section 12.7 covers generative image tools and the rights attached to what they produce. Section 12.8 closes with how to choose and combine tools, and how to leave them.

12.1 What Branding Tools Do

A branding tool is not bought because it is a branding tool; it is bought because it does a job. Every tool a small business meets in this field falls into one of four jobs, and a tool bought without naming its job is a tool that will be used for a few weeks and then quietly forgotten. The four jobs are create, store and share, protect and measure. A single tool may serve more than one of them, and no tool serves all four well, which is why the useful question in front of any offer is not what the tool can do but which of the four jobs it is for.

The first job is to create. Creating means drawing the mark, choosing and setting the type, building the layouts and producing the files the business will need. It is the job people think of when they hear the word branding, and it is the job for which the largest number of tools exists. This work belongs on the whole to a designer, because the decisions inside it — the curve of a letter, the space around a mark, the weight of a rule under a heading — are made with the eye and with practice. An owner who attempts this work in a general tool produces something that looks approximately right, and approximately right is precisely the quality that fails on a van seen at thirty metres on a highway.

The second job is to store and share. A brand is repeated thousands of times a year by people who are not designers: the administrator who sends a proposal, the lead technician who needs the logo on a slide, the printer who needs a file with its color codes. Storing and sharing means that each of those people finds the current file in a place they know, without asking anybody and without receiving a stale version forwarded from a colleague's inbox. This job is the least discussed and the most neglected, and the case followed through this book shows what neglecting it costs. Three versions of the logo, five blues, four typefaces and thirteen consistent touchpoints out of thirty-six is not a story about bad design. It is a story about a business with no place to look up the right answer.

The third job is to protect. Protecting means checking that a name is free to use before it is painted on a vehicle, filing it where the business can defend it, and holding the rights that attach to the work that has been paid for. This job is done in official registers and in written contracts rather than in software, and the tools it needs are the ones a state provides: a trademark database that can be searched, an application that can be filed, a fee that has to be paid, a certificate that lasts a fixed term. A small business can perform part of this work itself, and should not perform all of it alone, for the reasons Chapter 11 gives in full.

The fourth job is to measure. Measuring means finding out whether the brand is noticed, whether it is recognized, whether what people say about the business matches what the business promises, and whether anything has moved since the question was last asked. It is the job most often skipped, because it produces no object. A logo and a set of guidelines can be shown to a customer; a survey result cannot. Yet a brand that is never measured is a promise that is never checked, and Chapter 9 sets out what a business of seven people can measure with a telephone, a form and the reviews its customers already write.

Four jobs, then, and one habit in front of any purchase: name the job first, then ask what the platform the business already pays for covers, and only then read the features. A feature list describes everything a tool is able to do and says nothing about the work it was bought for. A template tool with two hundred capabilities that nobody can find is a worse purchase than a plain one the administrator opens every Monday morning. The order of those questions is what keeps a tool set small, and a small tool set is what keeps an identity consistent.

Figure 12.1 groups the families of tools a small business actually meets under the four jobs. The figure shows four columns, one for each job — create, store and share, protect and measure — and under each column the families that belong to it, from drawing and layout applications under create to the official registers under protect, with a mark beside the jobs the platform covers in part and a different mark beside those it does not cover at all. Read it as a map of the field rather than as a shopping list, because the useful question is which column is empty in the business at hand.

Figure 12.1 — Branding tools by job. Choose by the job to be done, and note what your platform already covers before buying.

The table that follows takes the same four jobs and sets out eight tool families inside them, with the function each family must perform and the person who uses it in the case followed through this book. Read it as a checklist of functions rather than of products: the columns name a job, a family, a function and a role, and no column names a vendor.

Table 12.1 — Branding tool families by job

DEFINITION Master file, template, brand portal and digital asset management

A master file is the original from which every other version of an element is produced: the drawing that holds the logo, the document that holds the layout. It is the file that must not be edited by accident, because every export made from it inherits its quality. An export is a copy made for a use: a vector file for the printer, a raster image for a screen, a small file for a signature. A template is a document whose fixed parts — the palette, the type, the placement — are already decided, and whose open parts are the words and the numbers a user supplies. A brand portal is a page or a small site that presents the identity and hands out its files, usually with the rules that govern them. Digital asset management goes further: it holds the assets with their versions, their usage rights and their expiry dates, and it records who downloaded what, which matters when a license to use a photograph runs out or when a campaign must be withdrawn.

12.2 Design and Layout Tools

The tools that create an identity are three families, and a small business will pay for none of them directly. The first family is the vector drawing application, in which the mark itself is drawn as paths and curves. Vector work is what allows a logo to be enlarged to the side of a building or reduced to the width of a favicon without losing its edge, because a vector drawing describes shapes rather than pixels. The file it produces — a vector format such as SVG, EPS or PDF — carries the geometry and the color codes and can be opened by a printer, a sign maker or a wrap shop years later.

The second family is the layout application, in which documents are built: the proposal, the report, the one-page monthly summary, the twenty-four-page guide. Layout work is where a visual identity stops being a mark and becomes a system of pages, and it is the work that decides how a heading sits above a paragraph, how a table is ruled, how much white space a page is allowed. In the case followed through this book, eleven templates came out of this family: a visit sheet, a one-page monthly report, a capital planning note, a consolidated portfolio report, a proposal, an invoice, the assessment report, a letter, an email signature, a slide and a social media post. Each of them is a page whose fixed parts were decided once.

The third family is image editing, which prepares photographs and screens: cropping, correcting tone, resizing a file for a screen or for print, and producing the different versions a website and a printed piece each need. A small business rarely needs this family as a standing tool. It needs the work, and the work is bought by the hour or arrives inside the studio's fixed fee, which in the case was $7,800 for the logo, the palette, the typography, the eleven templates and the guide.

Two technical points decide more arguments than any other in this area. The first is that color behaves differently on a screen and on paper. A color code written for a screen is a light value; a color intended for a printed sheet is built from ink, and the two do not translate exactly. This is why the palette of a brand should be held as named codes with their hexadecimal values — Nordec Blue #1F3A5F, Signal Amber #E8A33D, Slate #5B6770 and Mist #F4F6F8, in the case — and why the printed proof, not the monitor, is the final word on a print job. A printer works to its own color reference, and the conversation about a proof is a conversation with the printer.

The second point is the difference between the designer's tools and the owner's tools. The studio that built Nordec's identity works in its own vector and layout applications and delivers master files; the business does not need to own or learn those applications, and buying them would be a purchase of capability nobody in the business would use. What the business needs is the master file, the exports and the templates in a form it can open, which is the subject of section 12.3. A small business that buys a professional drawing application after a refresh, because it feels like part of the project, has bought a subscription and a learning curve rather than an asset.

Rights belong in this section because they attach to files. Under the Copyright Act, the author of a work is its first owner, and an assignment or grant of copyright is valid only if it is in writing and signed by the owner of the right; moral rights — the integrity of the work and the right to be associated with it — cannot be assigned but can be waived, in whole or in part, and an assignment of copyright does not by itself waive them. The contract with Nordec's studio assigns the copyright in writing and includes a waiver of moral rights, which is what makes the master files the business's to use, to alter and to hand to a printer. A business that pays for a logo without that clause pays for a license it has never read. The same Act provides that an employer is the first owner of a work made by an employee in the course of employment, unless the parties agree otherwise, which is the rule that applies to a template an administrator builds on a Tuesday afternoon on the employer's time.

One more file-level point belongs here. Master files should be delivered in the formats the businesses downstream will ask for, and the list is short: a vector format for the sign maker and the wrap shop, a document format for the printer, and screen formats for the website and social posts. A delivery that consists only of images is not a delivery of an identity; it is a delivery of pictures of one, and it will be redrawn badly the first time a partner needs a larger version.

THE RULE An open font license covers the font, not the logo

The SIL Open Font License, version 1.1 of February 26, 2007, permits using, studying, copying, embedding, modifying and redistributing the font software, and it permits selling the font only bundled with other software; reserved font names may not be used by modified versions without permission. The clause a business most often misreads is the one that sits beside those permissions: documents and other works made with the fonts are not covered by the license, so a logo set in a font under this license is not itself under it (SIL International, 2007). The practical consequences are three. A logo set in an open-license typeface can be registered and used as a mark, because the logo is a work of its own. The typeface file itself must still be used within the terms of its license. And a typeface bought under a commercial license must have a license that covers each use the business makes of it, including embedding it in a document that leaves the business. Nordec's two typefaces are open-license: a humanist sans-serif for everything and a monospaced face for building record codes. That choice keeps the license question simple, and it is a choice a business should record in the guide.

12.3 Templates and Self-Service Design

The templates of a visual identity live in two places, and confusing them is the most common cause of a drifting brand. They live as editable documents — the proposal, the report, the letter — which are opened, filled in and saved by the people who write them. And they live in a template-based design tool, which is a page-driven application where a person who is not a designer assembles a post or a sheet from fixed shapes: text boxes, color swatches, image frames and bands. The first kind of template is a document; the second is a set of constraints with a canvas. A business needs both, and it needs them to hold the same palette and the same type.

The value of a template is not that it saves time, though it does. The value is that it removes decisions that should not be made twice. When a template carries the palette and the type, the person using it makes the decisions that belong to them — what the post says, which photograph to use, which number to put in the table — and none of the decisions that belong to the brand. Chapter 5 sets out how a template is built from the design system and how the eleven templates of the case were chosen; this section is about the tools those templates live in and how they are kept honest.

Three practices keep templates honest. The first is that each template is locked in the parts that must not move, so that a user can change text and images but cannot change a color or a typeface. The second is that every template carries the same identifiers: the same mark in the same position, the same palette, the same heading rules. The third is that the template belongs to one person — the brand keeper — who is the only one who edits it, so that twelve versions of a template do not appear in twelve folders. In the case followed through this book the administrator became the brand keeper, holds the templates and the asset folder, and checks every new material before it leaves the business; Sophie approves any change to the promise or to the identifiers, which is a different kind of decision from a change to a template.

The risk of the template-based family is the one its name does not advertise. Templates are shared, and a template that thousands of other businesses use will make a business look like thousands of other businesses. The failure is not the tool; it is the picture a customer forms when a post looks exactly like the post of a competitor two streets away. The remedy is not to avoid templates, which would return the business to the problem templates solved. The remedy is to have one's own: either a template built from the business's own identifiers, or a starter layout that has been through the hands of a designer and altered until the palette, the type and the mark are the business's own. A generic template used unchanged is a borrowed identity, and it is borrowed from strangers.

Two smaller traps belong here. The first is the export nobody can reopen: a design saved only inside a template-based tool cannot be taken to another tool, which is the exit question of section 12.8. The second is the file made by a person who leaves, in a personal account that the business cannot reach, which is the reason templates are held centrally rather than by the person who uses them most.

The table that follows lists four families of tools with examples and with what each vendor publishes about its own pricing as of September 25, 2026, whether a free plan exists, and one note about the family. Read the columns in this order: the family first, then the example, then what the vendor actually publishes. The table records only what the vendors' own pricing pages stated on that date; it prints no amount, and a business should read the page itself before deciding.

Table 12.2 — Popular tools by family

The first two families in that table are the ones a small business meets in its first year of a refresh, and the second two are the ones it meets when the asset library has grown beyond a folder. The distinction that matters is not the size of the vendor but the number of people and the number of brands: one brand and ten users is a folder, several brands or many users with rights to manage is a portal or a managed library.

WHAT CHANGED Portable design tokens, October 2025

On October 28, 2025, version 2025.10, the first stable version of the design tokens specification, was published as a report of a World Wide Web Consortium Community Group; it is a Community Group report and not a W3C Standard (Design Tokens Community Group, 2025). The change matters to the tools of this chapter because of what the specification covers: theming and multi-brand support, modern color spaces, aliases between tokens, and one token file that can produce code for several platforms. A palette and a type scale held as tokens are no longer a description in a document that someone retypes into a website; they are a file that a website, an application and a set of templates can each read. For a small business the practical gain is the one this chapter keeps returning to: one source of truth for color and type, which several tools consume rather than a document each tool ignores. The specification is young and not every tool reads it, so a business should treat it as a direction rather than a requirement, and keep the values written in the guide as well.

TOOLS OF THE TRADE What a small business really needs

Five functions carry a small business's identity, and a business should look for the function rather than the product. The first is a set of editable templates that a non-designer can open, fill in and export, with the palette and the type locked and the text and the images open. The second is a shared asset folder with one folder per identifier, a separation between master files and exports, and a folder for retired files, reachable by everyone who needs it and writable by one person. The third is a form that can carry a short survey question and keep the answers with the form. The fourth is the official trademark database, searched before a name is filed or printed, free of charge, and treated as information rather than as an opinion. The fifth is a calendar that holds the checks of Chapter 10 and the recurring reading of Chapter 9, because a tool that is never opened on a date is a tool the business has stopped using. No brand name belongs in this list, and no price; the requirement is what the function does.

12.4 Brand Portals and Asset Management

The assets of an identity have to live somewhere, and the three places they can live correspond to three sizes of problem. The smallest is a folder on a shared drive, which costs nothing beyond the drive itself and is enough for one brand, one palette, eleven templates and a handful of people. The middle is a brand portal, a page that presents the identity — the mark, the palette, the type, the rules — and hands out the files, usually with a short set of instructions. The largest is a digital asset management system, which holds the assets with their versions, their usage rights and their expiry dates, and records who downloaded what. Each of the three answers a different question, and a business that buys the third to solve the problem of the first has bought a subscription and an obligation to keep it full.

The shared folder is the right answer more often than the market admits. It needs structure rather than software: one folder per identifier, a separation between the master files and the exports made from them, a folder for retired files, a naming convention that carries the version and the date, and one person with write access while everybody else reads. The method below sets out how to build it in an afternoon, and the case followed through this book is built on exactly that arrangement: an asset folder on the business's shared drive, one folder per identifier, and a retired folder for the old files. That last folder is the one that stops a business from using a retired logo, because a file that has been moved is harder to open by accident than a file that has been deleted and is still in somebody's downloads.

A brand portal earns its place when the files have to be handed to people outside the business: a sign maker, a wrap shop, a printer, a partner who needs the logo for a co-branded event. Its value is not storage but instruction — the portal presents the rules beside the files, so that a supplier reads what may and may not be done. Its limit is the same as the folder's: it holds files, not rights, and it does not know when a license to use a photograph ends. The case's arrangement points the same way, since the manufacturer's certified-installer mark may appear on the website only under the manufacturer's own usage rules, an arrangement a portal can describe but cannot enforce.

Digital asset management becomes worth its cost when the business has several brands, or many users, or assets whose rights expire. The functions that justify it are specific: versions, so that a file has a history rather than a replacement; usage rights, so that a photograph knows where it may be used; expiry dates, so that a license that has lapsed can be found before a customer finds it; and a record of downloads, so that a withdrawal can be traced. A business with one brand, eleven templates and seven people does not have those problems yet. It has the problem of a folder that is about to become untidy, and the answer to that problem is a method rather than a system.

METHOD Setting up an asset folder in an afternoon

First, create one folder per identifier: one for the mark, one for the palette, one for the type, one for the document templates, one for the social templates. An identifier is a single element people must be able to find by name alone. Second, inside each folder, separate the master files from the exports, so that the file that must not be edited sits apart from the files made to be used, and give the exports names that say what they are for. Third, create a folder called retired and move into it every file that has been replaced, rather than deleting it, so that a file in circulation elsewhere can be recognized as old when it appears. Fourth, adopt a file name that carries the version and the date, in the same order every time, so that the newest file sorts to the top and two files with the same name are impossible. Fifth, give read-only access to everyone but the brand keeper, who holds the write permission, so that the folder cannot be changed by seven people acting separately. The deliverable is a folder structure; the verification is that a colleague can find the current logo in under a minute without asking.

CASE IN POINT Nordec Entretien inc.: the retired folder

Nordec Entretien inc. is a fictional business created for this encyclopedia; its owner, its employees, its customers, its competitors, its figures and the events described are invented, and any resemblance to a real business is coincidental. The inventory that started this book's work found three versions of the logo in circulation across thirty-six touchpoints, and the reason was mundane: the oldest of the three was the original drawn by a friend in 2022, it was still in the template folder of the administrator's mail application, and it was still on a supplier's copy of a co-branded sheet. The new identity arrived with a fourth file, and the risk of a fifth was real. The answer was not a system. The administrator built one folder per identifier on the shared drive, moved the master files into a folder of their own, made the write permission theirs alone, and created a retired folder into which the original logo and its exports were moved rather than deleted. Six months later, the touchpoint inventory found thirty-three of thirty-six touchpoints consistent, and the three that were not were the three that were scheduled last. The retired folder is what made the difference on the day a supplier asked for the old file and was told, with a link, that it was retired.

IN LA MACHINE Visual Studio, Gallery and a Branding page that saves nothing

The platform's pages for making and holding brand material are worth reading one by one (La Machine, 2026). The Branding page at business/branding takes a logo, a brand name, a tagline, a website address, a primary and a secondary color with ten preset swatches and a choice among five fonts, computes the contrast of the primary color with a pass or fail badge at the 4.5:1 threshold, and previews a client invitation card and a client email — and saves none of it, so no other page uses what was entered. The Visual Studio at business/social/visual-studio edits social, video, print and web formats, saves its designs and sends its exports to the Gallery at business/marketing/gallery, which holds royalty-free stock images and records each one's license, the license's address and the attribution text, with a provenance form for the user's own images. The Magazine Cover page at business/magazine-cover is a free-canvas editor with seven starter templates that keeps its covers in the browser only, and the website builder's Theme & Branding panel keeps its configuration in the browser too. None of these pages reads a stored brand palette or logo, and the platform has no brand kit, no guidelines publisher, no asset management with versions or rights, no consistency audit, no trademark search and no image generation.

12.5 Name and Trademark Search Tools

A name is searched before it is adopted, and the tools for searching are mostly public. The central one in Canada is the Canadian Trademarks Database, which can be searched free of charge in a basic search and in an advanced search (Canadian Intellectual Property Office, 2026b). A basic search answers the question a business usually starts with — whether a word appears as a registered or pending mark at all. The advanced search answers the more useful question: within a class of goods or services, and within a period, which marks are close to the one in mind. Neither search is an opinion, and the difference between a search result and a legal opinion is the difference this section exists to make plain.

A search has two uses, and a business should perform both. The first use is to sort candidates before money is spent on any of them. In the case followed through this book, twelve candidates were generated for what would become the assessment offer, seven passed a screen on language and meaning because they had to work in French and in English, and four passed a knockout search of the trademark databases; the choice from those four was « Bilan Nordec » in French and "Nordec Building Check" in English. A knockout search is fast and cheap, and its only job is to remove a candidate that cannot be used. The second use is to inform the filing itself: a search that finds two registered marks close to a name in unrelated classes of goods, as the case's agent found for Nordec, tells the business where the risk sits even when the classes are different.

The official sources differ in what they cover, and a business that reads only one will miss a country. The United States Patent and Trademark Office states that it does not search for the applicant and recommends a clearance search of its own database of registered and pending marks, of state registries, of use on the internet, where rights can arise from use alone, and of international sources, so that refusals, oppositions and infringement suits are avoided (United States Patent and Trademark Office, 2023). Three lessons follow for a small business that exports or sells online. Rights can exist without a registration, which is why the internet is part of the search. A registration in one country is not a registration in another. And the search itself is the applicant's work, because no office performs it.

Two more checks belong beside the register, and they are performed in minutes. The first is the internet address: whether the name is free as a domain and in the spelling a customer would type. The second is the social handle: whether the name is available on the platforms the business uses, and, if it is not, whether a close variant is acceptable. Both checks are performed through the services that sell those names, and both should be done before the register search rather than after, because a name that is free as a mark and taken as an address is a name the business will not use.

Commercial search services form a fourth family, and they are used mostly by agents and lawyers. Their value is breadth — more registers in one query, watch services that report new filings after a search — and their limit is the same as every other tool's: a result is a fact about a database, not a conclusion about the risk of confusion. A business should treat the free registers as the place to begin, a commercial service as a way to widen the search when money is at stake, and an agent's opinion as the only thing that should decide a filing.

One sentence of law belongs here, because a name that is filed is a name that is claimed. In Canada, a person may apply to register a trademark for goods or services if they use it or propose to use it and are entitled to use it in Canada; the goods or services are stated in ordinary commercial terms and grouped by class of the Nice Classification, each group preceded by its class number; and a registration lasts ten years from the date of registration, renewable for further periods of ten years. Nordec's application for the word mark NORDEC was filed in Nice class 37 for building maintenance and repair services through a trademark agent, whose fee in this illustrative case was $1,200. This book explains the law and does not replace the advice of a lawyer or a trademark agent.

QUÉBEC & CANADA NOTE Searching Canadian trademarks

In Canada, the search that a business performs itself is free and takes minutes. The Canadian Trademarks Database of the Canadian Intellectual Property Office can be searched in a basic search and in an advanced search at no charge (Canadian Intellectual Property Office, 2026b). The same source gives a forecast wait of about 8.2 months to examination for an application filed in October 2026, which is the number to keep in mind when timing a launch around a filing. Two practical readings follow for a Québec business in particular. Because a trademark that is registered with the Canadian office enjoys a recognized status under the province's language rules, a business that intends to display a mark in another language on a sign has a reason to file early. And because the search is free, there is no reason to skip it: a knockout search before a name is printed on a vehicle costs twenty minutes, while a name that has to be withdrawn after the wrap is installed costs the wrap. A search does not replace the opinion of a trademark agent, and this book does not give legal advice.

12.6 Brand Monitoring and Surveys

The fourth job — measure — is performed with the least expensive tools of the four, and it is the job a small business postpones longest. Three families belong to it, and they sit in order of how much a business controls them. The first is the survey, in which the business asks its own customers a question and keeps the answers. The second is review reading, in which the business reads what customers publish on its public profile. The third is listening, in which the business reads what is said about it on social platforms it does not own. The three answer different questions, and a business of seven people can perform the first two thoroughly and the third only occasionally.

The survey family is where the case's measurement lives. The awareness question among owners of buildings not under contract in the two territories was asked by telephone, sixty people each time, and the answers were entered in a form; the count moved from thirteen of sixty to nineteen of sixty, with a margin of error of about ±10.4 and ±11.8 points at 95%, which means the rise cannot be distinguished from chance. Chapter 9 explains what that result does and does not permit a business to conclude, and the tool point here is narrower: the survey was performed with a telephone and a form, without a survey platform, because the sample was sixty people and the questions were few.

The same arrangement carried the customer question about the likelihood to recommend, on a scale from 0 to 10, answered by forty-one of fifty-two customers, with twenty-seven at the top of the scale, ten in the middle and four at the bottom, giving a net score of 56.1 measured once with no earlier value to compare it to. A single reading is a baseline, not a trend, and the tool that matters for a baseline is the one that makes the second reading possible: the same question, the same scale, the same period, the same way of asking. A form with an export the business can open is enough, and the discipline of asking the same way twice is worth more than any dashboard.

Review reading is the second family and the cheapest of all, because the content is already public. The case's online reviews moved from twenty-three with an average of 4.7 to thirty-one with an average of 4.8, and the reading of them is a task rather than a purchase: the reviews are read on the business profile, the recurring complaints are noted, and the specific facts — a missed window, a document not sent — are routed to the person who owns the work. Reading reviews has a second use that a survey cannot provide, which is the exact words a customer uses. The twenty interviews of Book A6 produced "on time" or "fast" fifteen times and "knows my building" eleven times, and those words became the material of the brand platform in Chapter 2.

Listening is the third family, and it is bought as software that collects mentions of a name from platforms the business does not own. It is the right tool for a business whose customers discuss it in public, and it is the wrong first purchase for a business of seven people whose customers call on the telephone. The honest position for a small business is that a listening service is a decision for later, taken when mentions have become frequent enough that reading them by hand is no longer possible. Until then, the same attention spent on the business's own profile, and on the two or three platforms where its customers actually are, covers the same ground for nothing.

WHO DOES THIS The brand keeper owns the tools, the studio owns its own

In the case followed through this book, the tools of the identity are held in two places, and the split is deliberate. The design studio holds the applications in which the mark, the palette and the templates were made, and it delivers master files and exports; the business does not own those applications and does not need to learn them. The administrator, as brand keeper, holds everything the business uses: the asset folder with its folders per identifier, the retired folder, the eleven templates, the template-based design tool for social posts, the survey form and the review reading. Sophie approves any change to the promise or to the identifiers. The trademark agent holds the search and the filing, and the official registers are public. Nobody in the business is asked to become a designer, and nobody outside it is asked to keep the business's files. That division is what makes the tool set small enough to maintain: two people hold assets, one person decides, and one supplier is paid for the work the business cannot do.

12.7 Generative Image Tools and Their Rights

Generative image tools produce pictures from a written instruction, and they have changed one part of the branding work more than any other: the speed at which drafts exist. A mood, a rough layout, three directions for a campaign image, a background for a slide — all of it can be in front of a person in minutes, which shortens the distance between an idea and something to look at. That is a real gain for a small business, and it is a gain in the part of the work where nothing is final. The difficulty begins at the moment an image stops being a draft and becomes an identifier: something a customer will see a hundred times and associate with the business.

Four questions have to be answered before that point, and none of them is answered by the tool's user interface. The first is what the tool's own terms say. Every generative image tool carries terms of use, and those terms state what the vendor claims about the outputs, what the business is permitted to do with them, and what the vendor reserves. Those terms change, and a business that reads them once at the beginning of a project is reading a document that may not still be in force at the end. The second question is the ownership of what comes out. The Copyright Act provides that the author of a work is its first owner and that an assignment or grant of copyright is valid only if it is in writing and signed by the owner of the right, which means a business that intends to own an image should establish who the author is and obtain a written assignment where one is possible. A vendor's claim about outputs is a claim, not a registration, and it should be read as carefully as any other contract term.

The third question is the risk of a lookalike. A generative image tool can produce, from a short instruction, something close to an existing mark or an existing character without any intention of copying it, and a mark that resembles a registered one raises exactly the risk of confusion that the search in section 12.5 exists to reduce. An image that a tool produces is not searched against a register before it is used, and a logo drawn from an instruction has not been through the clearance work a name goes through. Chapter 4 sets out what makes a mark distinctive and defensible, and Chapter 11 says who to consult; neither of those questions is answered by a prompt.

The fourth question is disclosure, and it is a question of the business's own standards rather than of any register. Some audiences accept an image made with a generative image tool without asking; others expect to be told. A brand whose promise is plain speech — Nordec's promise is that it knows the building and tells the owner ahead — has a reason to extend that plainness to the pictures it publishes, and the simplest rule is the one the guide can carry: images made with a generative image tool are not used as identifiers, and where they are used as illustration, the business knows what the tool's terms permit and applies them.

None of this argues against the family. It argues for a boundary that is easy to state and easy to keep: generative image tools are used for drafts, for backgrounds and for illustration, and the identifiers of the brand — the mark, the palette, the type, the templates — are made by people, under a contract that assigns the rights. Book K4 (Artificial Intelligence for Business) treats the use of these tools in a business, and Book H5 (Intellectual Property) treats the rights, the searches and the filings; this section is only the branding chapter's contribution to the question, which is that an identifier carries a promise and should be owned as deliberately as it is shown.

WATCH OUT A logo from a generator is not automatically yours

The error is easy to make and expensive to undo: a business asks a tool for a logo, likes the second suggestion, uses it on a van, a website and eighty invoices, and only later asks who owns it. Several separate questions have been collapsed into one, and each has to be answered on its own. What does the tool's terms of use say about the outputs, and have those terms changed since the image was made? Who is the author of the work, since the Copyright Act makes the author its first owner and requires an assignment to be in writing and signed? Does the output resemble a registered mark, given that nothing searched it against a register? And can the same image be produced again for somebody else, from a similar instruction? A business that cannot answer those questions does not hold an identifier; it holds a picture it is using on trust. The safe route for a mark is the one the case took with an independent studio: a fixed fee, a written assignment of copyright, and a waiver of moral rights.

12.8 Choosing and Combining Tools

Choosing a tool is easier when it is done in a fixed order, and the order is the whole of the method. The first question is which job the tool is for: create, store and share, protect or measure. The second is what the platform the business already pays for covers, because a function the platform performs is a function the business should not buy twice, and because a tool the platform already carries has no new account, no new password and no new renewal date. The third question is who will open the tool every week, which is a test of whether the tool will be used at all rather than of the tool's quality. The fourth is how the business gets its files out again. A tool that answers only the first question and ignores the other three is a subscription with a learning curve attached.

The platform coverage is worth stating plainly, because it is the part of the decision a small business most often skips. A branding page in a business platform may let a person enter a brand name, a tagline, a website address, a primary and a secondary color with ten preset swatches, choose among five fonts and see a contrast badge at the 4.5:1 threshold with a live preview of a client card and a client email; and it may still save none of it, so that the page is a place to try a palette and see its contrast rather than a register of the identity. A website builder inside the same platform may hold a gallery of themes, a brand name, a tagline, a logo, a favicon, a banner image, five color overrides and two fonts chosen from a catalog of two hundred, and keep that configuration in the browser. Read together, those two pages show what the platform is good for: trying a palette, previewing it and putting it on a site. Neither of them is a store of assets that other tools read.

One source of truth is the principle that keeps a combined tool set from splitting. The identity has values — a mark, a palette with its codes, a type scale, a set of templates — and those values must live in exactly one place that the business controls, with everything else made from that place. A second copy is not a backup when it can be edited separately; it is a rival. The asset folder of section 12.4 is that place for a business of seven people, and the guide is its written description. When a value changes, it changes there first, and everything downstream is regenerated.

Access and permissions are the second principle, and they are a design decision rather than an administrative one. The person who holds the write permission holds the identity, which is why the case's administrator is the brand keeper while the seven people read. Chapter 10 sets out the recurring checks that keep the arrangement alive, and the same logic applies to a portal or a managed library: an account that can change an asset is an account that can change the brand by accident.

Exit is the third principle, and it is asked at the beginning rather than at the end. Before a tool is adopted, the business should confirm that it can produce its files in a format it can keep: a document format for templates, a vector format for the mark, an export for anything held in a system. The design tokens specification points in a helpful direction here, since a token file that one tool can read is closer to a portable record than a palette written in a document somebody else maintains, but the specification is a Community Group report and not a standard, and the safest exit remains a copy of every asset in a format the business can open without the tool that made it (Design Tokens Community Group, 2025).

The tool set that results from all of this is deliberately small, and it is worth naming in functions rather than products: the studio's drawing and layout applications, which the business does not own; a template-based design tool for the posts and sheets a non-designer makes; an asset folder with one folder per identifier and a retired folder, held by one person; the official registers for the name; a form for the surveys; and a review page for what customers publish. Six functions, one of them outside the business, and none of them bought because of a feature list.

KEY CONCEPT One source of truth for the identity

The rule that holds a tool set together is that the identity has one place where its values live and are edited, and every other file in the business is made from that place. The mark, the palette with its codes, the type scale and the templates exist once, in a folder that one person may change and everybody may read; the exports, the documents, the website and the vehicle wraps are outputs. The rule sounds obvious and is violated constantly, because a copy is easy to make and a rival copy is easy to forget. When a color changes in the folder and not in a template, the business has two palettes. When a logo is replaced in the folder and not on a supplier's co-branded sheet, the business has two logos, in public. One source of truth is not a software feature, although some software helps; it is a decision about where the answer lives, taken once, by the person who holds the write permission, and announced to everybody who will ever need it.

DO THE MATH What Nordec's tool set costs in hours

Illustrative figures; the case is fictional. Step 1 — Setting up the asset folder takes 4 hours: one folder per identifier, master files separated from exports, a retired folder and the read-only permissions. Step 2 — Loading the eleven templates into the template-based design tool and locking their palette and type takes 6 hours. Step 3 — Building the survey form for the awareness calls and the recommendation question takes 2 hours. Step 4 — The three steps together come to 4 + 6 + 2 = 12 hours of the administrator's time, at $36 an hour, which is $432. Step 5 — The comparison is the point of the arithmetic: the software itself was already in place, so the cost of the arrangement is the administrator's time and not a new subscription, and $432 is spent once whereas a subscription is paid every month. A business that buys a new tool to avoid those twelve hours will pay more in the first year and own less at the end of it.

Costly Mistakes

• Buying by feature list instead of by job. A tool chosen for what it can do rather than for the job it is bought to perform ends up performing none of the four jobs well and occupies the place where a simpler tool would have worked.
• Keeping the identity only in the designer's files. A business that has never received the master files owns pictures of its identity rather than the identity, and it will pay for the work a second time the first time a sign maker asks for a vector.
• Letting every person keep their own copy of the logo. Three versions of a mark, five blues and four typefaces across thirty-six touchpoints is what separate copies produce, and no amount of checking at the end repairs it.
• Using a generic template unchanged. A template that thousands of other businesses also use makes a business look like those businesses, which is the opposite of what a template was supposed to do.
• Treating a free database search as a legal opinion. A search reports what a register holds; it does not say whether a name may be used, and the two questions are answered by different work.
• Assuming a font license covers the logo set in it. Under the SIL Open Font License, documents and other works made with the fonts are not covered by the license, so the logo and the font file are governed by different terms (SIL International, 2007).
• Using an image whose license and attribution were never recorded. A photograph without its source, its license and its attribution is a photograph the business cannot prove it was allowed to publish.

Key Takeaways

• Branding tools do four jobs — create, store and share, protect and measure — and a tool should be chosen by the job, then by what the platform already covers, and only then by its features.
• Master files, exports and templates are three different things, and a business owns its identity when it holds the master files under a written assignment of copyright, with moral rights waived as the Copyright Act allows.
• The template-based family is the one a non-designer uses; its value comes from templates built from the business's own identifiers, and its risk is the generic template that thousands of others use unchanged.
• A shared folder with one folder per identifier, separated master files, a retired folder and a single write permission is enough for one brand and a handful of people; a brand portal is for handing files and rules to outsiders; digital asset management is for several brands, many users, and assets whose rights expire.
• Searching a name is done in the official databases — in Canada, the Canadian Trademarks Database, free in a basic and an advanced search — and the search is information rather than permission, which is why a trademark agent assesses registrability and the risk of confusion (Canadian Intellectual Property Office, 2026b).
• Generative image tools shorten the distance between an idea and a draft, and they do not settle ownership, resemblance or disclosure; an identifier should be made under a contract that assigns the rights rather than borrowed from a tool's output.

Your Action Plan

1. Name the job before you look at any tool. Write, in one line, which of the four jobs the next purchase is for, and what the platform already covers of that job. Deliverable: one line naming the job and the gap. Verification: the line names a job, not a tool, and the gap is a function and not a preference.
2. Build the asset folder and the retired folder this week. Create one folder per identifier, separate the master files from the exports, add a retired folder, name files with their version and date, and make the write permission yours alone. Deliverable: a folder structure with a single writer. Verification: a colleague finds the current logo in under a minute without asking anybody.
3. Put the identity into the templates you actually use. Take the eleven materials the business sends most often, load them into the template tool with the palette and the type locked, and confirm that each one carries the same identifiers. Deliverable: templates that a non-designer can fill in and export. Verification: a filled-in template produced by somebody who is not the brand keeper passes the keeper's own check.
4. Search the name in the official database and record the result. Run a basic and an advanced search on the business's name and on the names of its offers, save the results with their date, and take the result to a trademark agent before anything is filed or printed. Deliverable: a dated search record and one question for the agent. Verification: the record names the database searched, the date, the classes looked at and the marks found close to the name.
5. Set the two measurement tools and put them on the calendar. Build the short survey form with the same question and scale used next time, confirm where the review page is read, and place both on the recurring check of Chapter 10. Deliverable: a form, a reading routine and two dates. Verification: the form holds the awareness question and the recommendation question with the scale, and the calendar holds the date on which each will be repeated.

In La Machine

The platform's treatment of the identity is visible in a few pages read in its code, and the picture is uneven in a way a business should know about (La Machine, 2026). The Branding page at business/branding offers a logo drop zone, a brand name, a tagline and a website address, a primary and a secondary color with ten preset swatches, a choice among five fonts, an email footer and three social links, and it computes the contrast of the primary color with a pass or fail badge at the 4.5:1 threshold, showing a live preview of a client invitation card and a client email. What it does not do is keep any of it: the save button shows a confirmation and no other page uses what was entered. Read plainly, the page is a place to try a palette and see its contrast, not a brand register. The Visual Studio at business/social/visual-studio is an editor for social, video, print and web formats whose designs are saved and whose exports go to the Gallery, and the Gallery at business/marketing/gallery is a library of royalty-free stock images that records each image's license, the license's address and the attribution text, with a provenance form for the user's own images. The Magazine Cover page at business/magazine-cover is a free-canvas editor with seven starter templates and formats from a portrait post to a link-preview image, exporting a PNG and keeping covers in the browser only.

The same reading shows what the platform does not cover, and this is the part that decides a tool set. There is no saved brand kit that other pages read; no brand portal or guidelines publisher; no digital asset management with versions, usage rights or expiry dates; no brand consistency audit; no brand health tracking, awareness survey or net promoter tool beyond the general form builder; no trademark search and no naming tool; and no image generation in the product. The website builder at business/website, with its Theme & Branding panel of twenty-eight themes, a logo, a favicon, a banner image, five color overrides and two fonts from a catalog of two hundred, keeps its configuration in the browser, which is a preview of the identity and not the identity itself. A business therefore keeps its asset folder, its guide, its search records and its survey results outside the platform, and uses the platform for the parts it does well. Book C2 (Websites, Landing Pages and Conversion) covers the website builder in full, and Book N6 (Social Media, Content and the Media Library) covers the media library.

Self-Check

1. What are the four jobs a branding tool can do, and why does the order of the questions — job, platform, features — matter?
2. What is the difference between a master file, an export and a template, and which of the three belongs in a tool a non-designer uses?
3. When is a shared folder enough, when is a brand portal worth having, and what does digital asset management add to either of them?
4. What do the official trademark databases do, what do they not do, and what did the Canadian office state about the wait to examination?
5. What does an open font license such as the SIL Open Font License cover, and where does a logo set in such a font stand under it?

Answers. 1. The four jobs are create, store and share, protect and measure, and the order matters because a tool chosen by its features is chosen for what it can do rather than for the work it was bought to perform, which is how a business ends with four tools doing part of one job. 2. A master file is the original from which every other version is produced, an export is a copy made for a use, and a template is a document whose fixed parts are decided and whose open parts are supplied by the user; a non-designer uses templates, and the double-sided practice of an editable document and a template-based design tool gives that person both. 3. A shared folder with one folder per identifier, separated master files, a retired folder and a single writer is enough for one brand and a handful of people; a brand portal earns its place when files and rules must be handed to suppliers and partners; digital asset management adds versions, usage rights, expiry dates and a record of downloads, which a business needs when it has several brands, many users or assets whose rights lapse. 4. The official databases let a business search registered and pending marks — in Canada, the Canadian Trademarks Database, free in a basic and an advanced search — and they do not give an opinion on whether a name may be used; the Canadian Intellectual Property Office forecast a wait of about 8.2 months to examination for an application filed in October 2026 (Canadian Intellectual Property Office, 2026b). 5. The license covers the font software, permitting its use, study, copying, embedding, modification and redistribution, and it states that documents and other works made with the fonts are not covered by it, so a logo set in such a font is a work of its own while the font file remains under the license (SIL International, 2007).

Chapter References

Brandfolder. (2026). Pricing. Retrieved September 25, 2026, from https://brandfolder.com/pricing

Canadian Intellectual Property Office. (2026b). Trademarks (page modified October 5, 2026). Retrieved October 29, 2026, from https://ised-isde.canada.ca/site/canadian-intellectual-property-office/en/trademarks

Copyright Act, R.S.C. 1985, c. C-42, ss. 13 and 14.1 (Canada). Justice Laws Website, current to September 3, 2026 (last amended November 7, 2024) — read September 25, 2026.

Design Tokens Community Group. (2025, October 28). Design tokens specification reaches first stable version. World Wide Web Consortium. Retrieved September 25, 2026, from https://www.w3.org/community/design-tokens/2025/10/28/design-tokens-specification-reaches-first-stable-version/

Figma. (2026). Pricing. Retrieved September 25, 2026, from https://www.figma.com/pricing/

Frontify. (2026). Pricing. Retrieved September 25, 2026, from https://www.frontify.com/en/pricing

La Machine. (2026). Documentation — route reference inventory. Internal product documentation, read September 25, 2026, together with the code of the pages this book describes: business/branding, business/social/visual-studio, business/marketing/gallery, business/magazine-cover and business/website.

SIL International. (2007). SIL Open Font License (Version 1.1, February 26, 2007). Retrieved September 25, 2026, from https://openfontlicense.org/open-font-license-official-text/

United States Patent and Trademark Office. (2023). Comprehensive clearance search for similar trademarks (last modified November 30, 2023). Retrieved September 25, 2026, from https://www.uspto.gov/trademarks/search/comprehensive-clearance-search-similar-trademarks

Chapter 13 — Practical Advice, Pitfalls and Your Action Plan

This chapter gathers the book into what matters most, the mistakes that cost the most, and a sixty-day plan that takes any small business from a scattered set of logos and documents to a coherent brand it can keep.

What You'll Be Able to Do

• Name the principles that decide whether a brand holds together, from the promise kept at every touchpoint to the honest sample behind every measure.
• Recognize the ten mistakes that cost small businesses the most money and time, and match each one to the chapter that treats it.
• Run the first fifteen days of a brand project, from the inventory of touchpoints to the platform workshop, with a deliverable at the end of each block.
• Commission the middle of the project — names, voice, a knockout search, a written brief and the choice of a designer — without ordering the vans.
• Build the identity, the design tokens and the first templates, and settle the architecture before a single vehicle is wrapped.
• Measure the result with a scorecard whose limits are stated as plainly as its figures, and keep the brand with a yearly rhythm rather than a yearly clean-up.

Why It Matters

A book about branding can leave a reader with more knowledge and no more clarity, because every chapter adds something that ought to be done. This chapter exists to remove that risk. It takes the twelve chapters that came before and reduces them to a small number of principles, a list of mistakes with their prices, and a plan of sixty days that a business of seven people can actually run. The plan is not a summary of the book; it is the order in which the book's work has to happen, and the order matters more than most owners expect. A business that writes its platform before it chooses a logo spends the same money and gets a mark that means something. A business that chooses the logo first spends the same money and gets a mark it will replace in four years.

The stakes are ordinary and concrete. A brand is what a customer remembers about a business when nobody from that business is in the room, and it is built out of small things repeated: the same blue on a van and on an invoice, the same promise in a proposal and in a monthly report, the same plain word for the same service in French and in English. None of those repetitions is expensive on its own. What is expensive is their absence, because a business whose materials disagree with one another asks its customers to do work the customer did not agree to do. The customer sees a van, a letter and a website that look like three companies, and the only thing that resolves the confusion is a price.

There is also a cost to moving too fast in the other direction. A rebrand that throws away the elements people already recognize asks the market to learn a business again, and the market is not obliged to accept the invitation. The research on rebranding found that structural change, above all mergers and acquisitions, is the most common trigger, and that a change in marketing aesthetics affects brand equity less than factors such as the behavior of employees (Muzellec & Lambkin, 2006). That finding is a reason to be calm about the visual side of a change and serious about the operational side. The colors and the letterforms are the visible part of a brand, and the visible part is the part that is easiest to fix. The promise, the response time, the report that arrives when it was promised — those are the parts that decide whether the brand is believed, and they are the parts a sixty-day plan must not postpone.

In This Chapter

Section 13.1 sets out the principles that matter most, drawn from Chapters 1 to 12, and defines the three words this chapter uses throughout. Section 13.2 lists the ten most costly mistakes, each with what it costs and the chapter that treats it. Section 13.3 opens the sixty-day plan with days 1 to 15, the inventory of touchpoints and the platform workshop. Section 13.4 covers days 16 to 30: names, voice, the knockout search, the brief and the choice of a designer. Section 13.5 covers days 31 to 45: the identity, the design tokens, the first templates and the architecture decision. Section 13.6 covers days 46 to 60: the rollout of documents and digital touchpoints, the one-page guide and the first measures. Section 13.7 explains what the scorecard can and cannot show. Section 13.8 reads the case six months after the rollout and records what its owner would do differently.

13.1 What Matters Most

Seven principles carry this book, and each of them was earned in a chapter rather than asserted in an introduction. They are worth stating together, because a reader who remembers only these will still build a brand that holds, and a reader who remembers every technique but forgets these will build a set of attractive files that disagree with one another within a year.

The first principle is that a brand is a promise kept at every touchpoint. Chapter 1 defined the brand as what a business promises beyond its offer and how that promise is seen, and Chapter 7 showed that the promise is kept or broken at the points where a customer actually meets the business: the van in the parking lot, the technician at the door, the invoice, the monthly report, the telephone call at eleven at night. A promise that is written in a strategy document and contradicted by a late arrival is not a weak promise; it is a different promise, and the customer believes the one they experienced.

The second principle is that the platform comes before the logo. Chapter 2 built the platform — purpose, promise, values, personality — and every chapter after it consumed that platform rather than replacing it. The logo is a container for meaning that the platform supplies. Chosen first, it is a shape that has to be justified afterwards, and the justification is always weaker than the decision would have been.

The third principle is that names are checked before they are loved. Chapter 3 described a knockout search of the trademark databases as a step that happens early, before a team becomes attached to a name, and the reason is not legal caution alone. A name that has to be abandoned after it has appeared on a proposal, a uniform and a website costs far more than the search that would have found the problem in an afternoon.

The fourth principle is that accessibility is part of the identity rather than a review that happens at the end. The Web Content Accessibility Guidelines set a contrast requirement of at least 4.5:1 for text and 3:1 for large text at level AA, with an exception for text that is part of a logo or a brand name (World Wide Web Consortium, 2024). A palette that fails those numbers is not a palette with a small flaw; it is a palette that will be corrected later, at the cost of every material printed with it.

The fifth principle is that a system beats a set of files. Chapter 5 replaced the idea of a logo file with the idea of design tokens and templates: named values that a document, a website and a report all read from the same source. A folder of files is a set of copies, and copies drift. A system is a set of references, and references stay true.

The sixth principle is that a business should evolve before it revolts. Chapter 8 argued for keeping the elements people already recognize and changing what has stopped working, and the research supports the caution: when a logo was redesigned from angular to more rounded in a field experiment with 632 participants, strongly committed consumers judged the new mark more negatively and their attitude to the brand fell, while weakly committed consumers responded positively (Walsh et al., 2010). The people most attached to a business are the least enthusiastic about a change they did not ask for.

The seventh principle is that measurement must use honest samples. Chapter 9 built a scorecard, and it insisted on stating the margin of error beside every figure taken from a small sample. The reason is not modesty. The research on long-term brand building and short-term activation warns against using very short-term metrics as the primary measures of performance (Binet & Field, 2013), and a small business that reads a two-point movement in a sample of sixty as a triumph has replaced one bad habit with another.

DEFINITION Coherence, consistency and distinctiveness

Coherence is the quality of a brand whose parts agree with one another: the promise, the name, the voice, the colors and the behavior all point in the same direction, so that a customer who meets the business in two places meets the same business twice. Consistency is coherence maintained over time and across materials, which is a matter of systems rather than of intention: the same token, the same template, the same rule applied by whoever happens to be producing the document. Distinctiveness is the quality of being recognizable without the name being read — the shape, the color, the letterform or the phrase that a customer can identify from across a parking lot or halfway down a page. The three are related and not identical. A brand can be consistent and not distinctive, which is the fate of a business that copies the conventions of its category carefully. A brand can be distinctive and not coherent, which is the fate of a business whose materials were made by four people who never met. Coherence is the foundation, consistency is the discipline, and distinctiveness is what the two of them buy over time.

KEY CONCEPT The platform before the logo

The central idea of this book is an order of operations. A business decides what it promises, to whom, and how it behaves when the promise is inconvenient; then it decides what to call that promise; then it decides how the promise looks; then it builds the system that repeats the look; then it puts the system in front of customers. Every one of those steps is cheaper than the step before it and more expensive than the step after it, which is why the order is not a matter of taste. A logo chosen before the platform has no criteria to satisfy, so it is judged by preference, and preference belongs to whoever is in the room. A logo chosen after the platform has a specification: it must carry the letterform or the color that the business will repeat, it must work at the size of a favicon and the size of a depot sign, it must survive being printed in one color, and it must not be confusable with the marks of the businesses a customer might also call. Those are checkable requirements, and a mark that satisfies them is a decision rather than an opinion.

13.2 The Ten Most Costly Mistakes

The mistakes in this section are the ones that cost small businesses real money, and the cost is rarely the invoice for the work itself. It is the second invoice, the one that pays for doing the same thing again after the first attempt has been printed, wrapped, installed or filed. Ten of them recur often enough to be worth naming, and each one is treated in full in the chapter beside it.

The first mistake is starting with the logo. Its cost is a mark that has to be replaced once the business discovers what it is trying to say, and the replacement is not only the design fee but every material that carried the first mark. The remedy is to write the platform first, in the two half-day workshops of Chapter 2, and to give the designer a brief rather than a preference.

The second mistake is letting the touchpoints drift. Its cost is measured in the case followed through this book: an inventory of 36 touchpoints found 3 versions of the logo, 5 different blues and 4 typefaces, with only 13 of the 36 consistent with one another. The remedy is the inventory and the templates of Chapters 5 and 7, which turn a set of one-off documents into a set of documents that read from the same source.

The third mistake is choosing a name without searching. Its cost is a rename at the worst possible moment, after the name has appeared on a proposal, a vehicle and a website, and the remedy is the knockout search of Chapter 3, performed before the team has become attached to anything.

The fourth mistake is writing a promise the business cannot keep. Its cost is not only a disappointed customer but a representation to the public that the business may have to defend, since a brand promise or a tagline made to the public is a representation under the Competition Act, and the remedy is to test every promise against a decision it would change, as Chapter 2 does with its three values.

The fifth mistake is treating the guide as a document. Its cost is a guide that is admired once and never opened, and the remedy is the system of Chapter 5: tokens that carry the values and templates that carry the layout, so that the guide explains a system rather than describing a set of files.

The sixth mistake is giving every offer its own brand. Its cost is that each brand must earn its own recognition from nothing, and the remedy is the architecture decision of Chapter 6, where a branded house with descriptive names keeps one reputation working for every offer.

The seventh mistake is rolling out the vans first. Its cost is that the most expensive and least reversible touchpoint carries an identity that is still moving, and the remedy is the order of Chapter 8: documents and digital first, then vehicles and uniforms, then the sign.

The eighth mistake is leaving the rights with the supplier. Its cost is that the business cannot reuse, adapt or defend what it paid for, and the remedy is a written assignment of copyright signed before the last payment, as Chapter 11 recommends, since the Copyright Act makes an assignment valid only in writing and signed.

The ninth mistake is measuring with a sample too small to read. Its cost is a decision taken on a movement that is indistinguishable from chance, and the remedy is the scorecard of Chapter 9, which states the margin of error beside every figure.

The tenth mistake is rebranding as a revolution. Its cost is the recognition the business has already paid for, thrown away in a single announcement, and the remedy is the evolution of Chapter 8, which keeps the name, the color family and the letterform that people already know.

The following table sets out the ten mistakes with what each one costs, the remedy that closes it and the chapter that treats it in full. Read it as a diagnostic list rather than a confession: a business that recognizes three of its own habits here has found the three places where its next year of brand spending is most likely to be wasted.

Table 13.1 — The ten most costly branding mistakes

WATCH OUT Ordering the vans first

The van is the touchpoint that feels most urgent and is the worst place to start. It is urgent because it is visible: two vehicles on the road every working day, seen by every owner and manager the business drives past, which is why the temptation to wrap them in the first month is strong. It is the worst place to start because it is the most expensive single item, the least reversible, and the one that carries the identity at the largest size and the longest distance. A wrap applied before the palette is settled freezes a color that may still change, and a wrap applied before the letterform is settled freezes a shape that may still be corrected. The order that works is the opposite of the order that feels urgent. Documents and digital materials are cheap, fast and easy to correct, and they are where the identity is tested against real content: an invoice with a long customer name, a report with a table, an email signature at the bottom of a plain message. Once those hold, the vehicles and the uniforms follow, and the depot sign follows them, because a sign is a permit application as well as a design.

13.3 A Coherent Brand in 60 Days: Days 1 to 15

The plan in this chapter takes sixty days, in four blocks of fifteen, and it assumes a business of the size followed through this book: seven people, a handful of offers, and an owner who can give the project a few hours a week. It does not assume a budget beyond the one the case spent, and it does not assume that the business stops working while the plan runs. Each block ends with a deliverable that somebody can check, which is the property that separates a plan from an intention.

Days 1 to 15 are the inventory and the platform, and they are the block most often skipped because they produce nothing that can be photographed. The inventory comes first. Every place a customer meets the business is listed, from the van to the invoice to the telephone greeting, and a sample of each is collected into one folder. The samples are not gathered to be judged; they are gathered so that the business can see what it is actually doing rather than what it believes it is doing. In the case followed through this book, that exercise produced 36 touchpoints, and the count itself was the first surprise, because the owner had guessed at a dozen.

The second half of the block is the platform workshop of Chapter 2: two half-days with the whole team, producing a purpose, a promise, three values with the decision each one changes, four personality traits and a tagline. The workshop is short on purpose. A platform written over three months is a platform written by the person who writes documents, and a platform written in two half-days by seven people is a platform those seven people can repeat from memory, which is the only test that matters in the first year.

The block ends with two deliverables. The first is the inventory, with each touchpoint marked consistent or not and ranked by how often customers see it. The second is a one-page platform, short enough to be read aloud in a meeting and specific enough that a designer can work from it without asking what the business means. Neither deliverable costs anything except time, and both of them change what the next forty-five days buy.

Figure 13.1 shows the sixty days as a single timeline. The figure is drawn as four blocks of fifteen days laid end to end, each block carrying its work above the line and its deliverable below it: the inventory and the platform in the first block, the names, the voice, the search and the brief in the second, the identity, the tokens, the templates and the architecture in the third, and the rollout of documents and digital materials with the first measures in the fourth. A reader listening rather than looking should hold one image: four blocks, four deliverables, and a business that can check each one before the next begins.

Figure 13.1 — A coherent brand in 60 days. Each block ends with a deliverable someone can check.

The following table sets out the four blocks with the work each contains, the deliverable it produces and the check that confirms the block is finished. Read the check column first: a block that cannot be checked is a block that will be declared finished on the day the calendar says so, whether or not the work was done.

Table 13.2 — The sixty-day plan

METHOD Running the day-one inventory

First, list every place a customer meets the business, and list them by kind rather than by department: the vehicle, the uniform, the telephone greeting, the proposal, the invoice, the report, the letter, the email signature, the website, the social profile, the depot sign, the business card, the trade-show table. Second, collect a sample of each into one folder, in the form the customer actually receives it, which means a photograph of the van and a saved copy of the invoice rather than a description of either. Third, note for each sample the version of the logo it carries, the colors it uses, the typefaces it is set in and the voice it speaks in, writing down what is there rather than what should be there. Fourth, mark each touchpoint consistent or not, using a single test: would a customer who saw this and one other touchpoint believe they came from the same business? Fifth, rank the touchpoints by how often customers see them, because the ranking is what decides the order of the rollout in days 46 to 60. Deliverable: one ranked inventory with a sample and a mark for every touchpoint. Pitfall: listing the touchpoints the business controls and forgetting the ones it does not, such as the review a customer writes or the way a technician answers a question at the door.

DO THE MATH What the sixty days cost Nordec

Illustrative figures; the case is fictional. Step 1 — The outside spending committed inside the sixty days is the design studio at $7,800, the trademark agent at $1,200 and the website update at $2,900, which is 7,800 + 1,200 + 2,900 = $11,900. Step 2 — The production that is scheduled after day 60, once the identity is settled, is the two vehicle wraps at 2 × $4,100 = $8,200, the uniforms at 7 people × 5 garments × $42 = $1,470, the depot sign at $1,900 and the printed items at $650, which is 8,200 + 1,470 + 1,900 + 650 = $12,220. Step 3 — The two together are 11,900 + 12,220 = $24,120, which is 24,120 ÷ 447,012 = 5.4% of the annual run-rate of $447,012, spent once. Step 4 — The in-house time is 64 hours of the administrator's time at $36, which is 64 × 36 = $2,304, and it is the part of the budget most often forgotten because it never appears on an invoice. Step 5 — The rhythm that keeps the result, described in Chapter 10, costs about $2,196 a year, which is the number to compare against the $24,120: the project is a one-time cost, and the maintenance is what decides whether it stays spent.

13.4 Days 16 to 30

Days 16 to 30 produce the words. The block begins with names and ends with a chosen designer, and it contains the one step in the whole plan that is cheapest to perform and most expensive to skip: the knockout search.

The naming work of Chapter 3 starts from the platform rather than from a list of words. The business knows what it promises and how it sounds, so the naming criteria can be written before any candidate is proposed: the name must work in French and in English, it must be pronounceable on a telephone, it must not describe a service the business may stop offering, and it must not be confusable with a mark already registered in the classes where the business operates. In the case followed through this book, twelve candidates were proposed, seven passed the language and meaning screen, and four survived the knockout search. The search is not a clearance opinion and it is not a guarantee; it is a check that removes the candidates with an obvious problem before anybody becomes attached to them, and it is performed by a trademark agent when the business is serious about the name it has chosen.

The voice work belongs in the same block because it is the same material seen from another angle. Chapter 3 described a voice with four principles and a glossary of words to use and to avoid, and the glossary is what makes the voice repeatable by seven people who are not writers. A voice that exists as an adjective — "professional", "friendly" — is not a voice; it is a hope. A voice that exists as four rules and twelve words is something a dispatcher can apply to a message at seven in the morning, which is the only test a small business's verbal identity has to pass.

The block ends with the brief and the choice of a designer, and this is where Chapter 11 earns its place. The brief is a short document, and its length is its virtue: the platform, the inventory's ranked list, the naming decision, the voice rules, the deliverables expected, the fee, the schedule and the rights. The rights clause is not a formality. The Copyright Act provides that the author of a work is its first owner, that an assignment or grant of copyright is valid only if it is in writing and signed by the owner of the right, and that moral rights cannot be assigned but can be waived (Copyright Act, R.S.C. 1985, c. C-42, ss. 13 and 14.1). A business that pays for a logo without a signed assignment has paid for the use of a file it does not own, and the book explains these provisions rather than advising on them; a lawyer or a trademark agent is the person to consult about a particular contract.

The choice of a designer follows the brief rather than preceding it, and the criteria are the ones the brief already contains. The business is looking for a studio that has read the platform, that asks about the touchpoints rather than only about the logo, that can work in both languages, that will deliver the source files and the tokens, and that will sign the assignment. A portfolio is evidence of taste; the brief is evidence of fit. In the case, three routes were presented by an independent design studio on a fixed fee, and the fee covered the logo, the palette, the typography, eleven templates and the guide, which is the scope a small business should expect to define before it asks for a price.

QUÉBEC & CANADA NOTE French first, on every new material

In Québec, the language of a brand is governed before it is designed. The Charter of the French language requires French in commerce and business in Québec, including commercial documents, signs and product inscriptions, and this book states its rules through the explanations of the Office québécois de la langue française and of the Registraire des entreprises rather than by citing an article. Since June 1, 2025, a sign visible from outside a premises that carries a recognized trademark wholly or partly in another language, or a business name with a non-French specific element, must be accompanied by French terms so that French is markedly predominant — a much greater visual impact, at least twice the space, permanent, at least as visible and legible — and if a French version of the trademark is registered, it must be displayed. Since the same date, the generic or descriptive words inside a trademark on a product must appear in French, with a transition period to June 1, 2027 for some products made before June 1, 2025. A business name's generic term must be French and is mandatory when the specific element is in another language. Nordec's application for the word mark NORDEC in Nice class 37 was filed through a trademark agent and was still under examination six months later. This book explains these rules; it does not replace the advice of a lawyer or a trademark agent.

13.5 Days 31 to 45

Days 31 to 45 produce the identity and the system that carries it. The block is the one most owners look forward to and the one most likely to run over, because design work expands to fill the time available unless the brief fixes what is being delivered. The brief fixed it in the previous block, and this block is where that discipline is tested.

The identity work of Chapter 4 begins with the routes rather than with the mark. A studio that presents three routes is presenting three answers to the brief, and the business chooses between them on the criteria it wrote rather than on preference. In the case, the chosen route kept the letter N and drew a horizon line through it, reading as the building seen a year ahead, and it dropped the generic wrench that had stood in the first logo. The wrench is worth a sentence, because it is the most common mistake in a trade's first mark: a tool drawn from the category rather than from the business. A wrench says what the business does and nothing about how it does it, and every competitor can draw the same wrench.

The palette follows the same logic. Five colors were chosen — Nordec Blue #1F3A5F, Signal Amber #E8A33D, Slate #5B6770, Mist #F4F6F8 and white — and each one was measured before it was approved. White text on Nordec Blue measures 11.48:1, Nordec Blue text on Mist measures 10.60:1, Slate text on white measures 5.80:1, and Nordec Blue text on Signal Amber measures 5.32:1, all of them above the 4.5:1 threshold that the guidelines set for text at level AA. Signal Amber text on white measures 2.16:1, which is why amber never carries text on white and serves instead for accents, the van stripe and safety details. The measurement is the point: a palette is a set of contrast ratios before it is a set of colors, and the ratios are what make the palette usable by people who are not designers.

Typography is chosen in the same block, and it is chosen as a family rather than as a favorite. The case uses two open-license typefaces: a humanist sans-serif for everything and a monospaced face for building record codes. The license matters because a font is software. The SIL Open Font License permits using, studying, copying, embedding, modifying and redistributing the font software, with reserved font names protected, and it states that documents and other works made with the fonts are not covered by the license, so a logo set in such a font is not itself under it. A business that chooses a typeface without reading its license may discover, at the worst moment, that the terms do not cover the use it has made.

The system work of Chapter 5 closes the block. The case produced 21 design tokens — 7 color, 6 typography, 5 spacing and 3 logo sizes — and 11 templates, from the visit sheet to the social media post, with a guide of 24 pages and a one-page summary. The tokens are the part that makes the templates stay true: a color is defined once and referenced everywhere, so a change to the blue is a change in one place rather than in eleven documents. The architecture decision of Chapter 6 belongs in the same block, because it decides how many names the system has to carry. The case chose a branded house: one master brand, Nordec, with descriptive names for the tiers and for the Bilan Nordec, and it rejected a separate brand for the entry offer on the ground that a new brand would have to earn its own recognition from nothing.

THE RULE Assign the rights before the last payment

The Copyright Act provides at section 13(1) that the author of a work is its first owner, at section 13(3) that the employer is the first owner of a work made by an employee in the course of employment unless agreed otherwise, at section 13(4) that an assignment or grant of copyright is valid only if it is in writing and signed by the owner of the right, and at section 14.1 that moral rights — the integrity of the work and the right to be associated with it — cannot be assigned but can be waived, in whole or in part, and that an assignment of copyright does not by itself waive them (Copyright Act, R.S.C. 1985, c. C-42, ss. 13 and 14.1). For a brand project the rule is therefore simple to state and easy to postpone: the written assignment and the waiver are signed before the last payment, not after it. A studio that has been paid in full has no reason to sign anything, and a business that discovers it does not own its own logo has a problem that costs more to solve than the assignment would have cost to obtain. This book explains the law; it does not replace the advice of a lawyer or a trademark agent.

TOOLS OF THE TRADE The minimum kit for sixty days

Five functions carry the whole plan, and a business should look for the functions rather than for products. The first is an inventory spreadsheet with a row per touchpoint and columns for the sample, the logo version, the colors, the typefaces, the voice, the consistent-or-not mark and the ranking by how often customers see it. The second is a one-page platform, in a document the whole team can open, holding the purpose, the promise, the three values with their decisions, the four personality traits and the tagline in both languages. The third is a set of editable templates that read from the same defined values, so that a change to a color or a size is made once and appears everywhere it is used. The fourth is a shared asset folder with a naming convention and a single current version of each file, because the most common cause of drift is two folders that both look official. The fifth is a survey form capable of carrying a short set of questions, such as a likelihood-to-recommend question, with the answers kept together so that a later reading can be compared with an earlier one. No brand name belongs on this list, and no price; the requirement is the function.

13.6 Days 46 to 60

Days 46 to 60 put the system in front of customers, and they are deliberately the shortest block in terms of new decisions. Everything that could be decided has been decided; what remains is the order of the rollout, the guide that explains it, and the first measures that will tell the business whether the work landed.

The order of the rollout is the one Chapter 8 sets out, and it is the reverse of the order that feels urgent. Documents and digital materials go first: the templates, the website, the email signatures and the business profile. They go first because they are cheap, fast to correct and easy to test against real content, and because they are the touchpoints a customer reads most closely. In the case, the website update was handled with the material of Book C2 (Websites, Landing Pages and Conversion), and the templates replaced the documents that had been produced one at a time for four years. Vehicles and uniforms come second, and the depot sign comes last, because a sign is a permit application as well as a design and its schedule is not entirely the business's to control.

The announcement belongs in this block and it is smaller than most businesses expect. In the case, a letter went to the 52 customers under contract and a line appeared in the next monthly report, saying what had changed and what had not. The second half of that sentence is the important one. A customer who receives a new-looking document wants to know whether the service, the price or the person they call has changed, and a business that answers that question in one line prevents a dozen telephone calls. The research on rebranding supports the emphasis: a change in marketing aesthetics affects brand equity less than factors such as the behavior of employees, and rebranding has to be managed as a whole, with the support of stakeholders (Muzellec & Lambkin, 2006). The letter is not the rebrand; the behavior of the seven people in uniform is.

The guide closes the block, and it is one page before it is twenty-four. The case produced a guide of 24 pages and a one-page summary, and the summary is the document that gets used. It holds the logo and its clear space, the palette with its contrast ratios, the two typefaces, the voice's four principles and the name of the person to ask. A guide that begins with twenty-four pages is a guide that is read once; a guide that begins with one page is a guide that is pinned beside a desk.

The first measures are taken in the same block, before the rollout is complete, because a measure taken only at the end has no baseline. The case took its first reading before the rollout: prompted awareness among owners of buildings not under contract in the two territories, by telephone, with 60 respondents, giving 13 of 60, or 21.7%, with a margin of error of about ±10.4 points at 95%. That figure is not a result; it is a starting line, and its value lies entirely in being taken before anything changed. Chapter 9 sets out the rest of the scorecard, and section 13.7 explains what each of its numbers can and cannot show.

WHO DOES THIS One owner, one brand keeper, one studio

Three roles carry a brand project in a small business, and the project fails when they are blurred. The owner holds the promise. In the case, Sophie Nadeau approves any change to the promise or to the identifiers, which means the tagline, the logo, the palette and the names of the offers do not change without her, and nothing else does. The brand keeper holds the system. In the case, the administrator holds the templates and the asset folder and checks every new material before it goes out, which is a task of perhaps an hour a week and the single most effective defense against drift. The studio holds the craft. An independent design studio produced the identity on a fixed fee and delivered the source files, the tokens and the templates, and its role ends when the system is handed over, which is why the handover is written into the brief. A fourth role appears only when it is needed: a trademark agent performs the knockout search and files the application, and a lawyer reads the assignment and the license terms. The distribution matters more than the titles. A business where the owner is also the brand keeper will find that the system drifts whenever the owner is busy, and a business where nobody owns the promise will find that it changes with every new document.

13.7 Measuring the Result

The scorecard of Chapter 9 is read twice in this plan: once at day 60, as a baseline, and once six months later, as a comparison. The two readings are the same instrument, and the instrument is deliberately small. It holds prompted awareness among a defined group, unprompted mentions, the share of touchpoints that are consistent, the number and average of online reviews, the share of prospects who knew the business before the first contact, and a customer survey on the likelihood to recommend. Six measures, each with a source and a date, and each with a stated limit.

The first limit is the sample. A telephone survey of 60 owners gives a margin of error of about ±10.4 points at 95% confidence, which means that a movement smaller than roughly twenty points between two readings cannot be distinguished from chance. This is not a technicality to be mentioned in a footnote; it is the single most important sentence in the section. A business that reads a rise from 21.7% to 31.7% as proof that its brand work succeeded has read a movement of ten points on an instrument whose error is ten points, and the honest reading is that the two figures are indistinguishable. The case states exactly that about its own numbers, and the discipline of stating it is what makes the rest of the scorecard credible.

The second limit is what a measure can show. Prompted awareness shows whether a name is recognized when it is offered; it does not show whether the business would be considered. Unprompted mentions show whether the name comes to mind without help, which is a harder and more valuable test, and they are usually too few in a small sample to support any conclusion at all. The share of consistent touchpoints is a measure of the business's own discipline rather than of the market's opinion, and it is the one figure in the scorecard that can move quickly and be trusted, because it is counted rather than sampled. Online reviews are a measure of what customers chose to say in public, which is neither a random sample nor a complete record.

The third limit is time. In this book's reading, brand building works over years and activation over weeks, and the research on the balance between the two warns against using very short-term metrics as the primary measures of performance (Binet & Field, 2013). A small business reading its scorecard six months after a rollout is reading an early signal, not a verdict, and the measures that will matter most are the ones that keep being taken. The value of a scorecard is not in any single reading; it is in the fourth and fifth reading, taken the same way, on the same questions, with the same sample size, so that a trend has something to stand on.

The fourth limit is attribution, and it is the one small businesses most often ignore. A brand project never runs alone. In the case, the six months after the rollout also contained the new tiers of Book A6 (Designing a Value-Added Offer), the Bilan Nordec entry offer and the referrals that followed from it, and none of those can be separated from the brand work in a business of seven people. The honest statement is that several things moved at once and the brand was one of them. A business that claims the whole movement for its brand has learned nothing it can use; a business that says so has learned where to look next.

The fifth limit is the customer survey itself. A likelihood-to-recommend question, answered by 41 of 52 customers, gives a figure that is useful as a repeated measure and misleading as a single number. The case recorded 27 promoters, 10 passives and 4 detractors, a net promoter score of 56.1, measured once with no earlier value. A number with no earlier value cannot show a change, and a number with a response rate of 79% is a good response rate and still not the whole customer base. The scorecard is a set of instruments, and an instrument is only as useful as the second time it is used.

WHAT CHANGED What changed at Nordec in six months

The case measured the same scorecard before the rollout and six months after it. Touchpoints consistent with one another went from 13 of 36 to 33 of 36. Prompted awareness among owners of buildings not under contract in the two territories, by telephone with 60 respondents each time, went from 13 of 60 (21.7%) to 19 of 60 (31.7%), with margins of error of about ±10.4 and ±11.8 points at 95%, so the rise cannot be distinguished from chance. Unprompted mentions went from 3 to 5. Online reviews went from 23 with an average of 4.7 to 31 with an average of 4.8. Prospects who knew Nordec before the first contact went from 5 of 15 to 8 of 14. The customer survey on the likelihood to recommend, answered by 41 of 52 customers, gave 27 promoters, 10 passives and 4 detractors, a net promoter score of 56.1, measured once with no earlier value. The trademark application was still under examination. The weekly check had caught two off-brand materials. The book does not attribute the growth that followed to the brand: the offer, the Bilan Nordec and referrals moved at the same time.

13.8 The Case, Six Months Later

Six months after the rollout, Nordec Entretien inc. held 56 buildings under contract, against 52 before: 6 Essential, 43 Preventive and 7 Portfolio. The book states that figure and then states what it cannot do with it. The offer of Book A6, the Bilan Nordec and the referrals that followed from it all moved in the same six months, and a business of seven people cannot separate the effect of a new palette from the effect of a new entry offer. The honest reading is that the brand work removed a set of frictions — three logos, five blues, four typefaces, documents that looked like several companies — and that removing friction makes everything else easier without being the thing that caused the growth.

The measures that can be read with confidence are the ones the business counts rather than samples. Consistency went from 13 of 36 touchpoints to 33 of 36, and the three that remain are known: a supplier's invoice that carries the old mark, a trade-show banner that will be replaced at its next use, and a piece of signage at a customer's building that belongs to the customer. The weekly check performed by the brand keeper has caught two off-brand materials in six months, which is a small number and the reason the number is small. The trademark application for the word mark NORDEC in Nice class 37 was still under examination at the six-month mark, and the book does not invent its outcome.

The measures that must be read with their limits are the market ones. Prompted awareness rose from 13 of 60 to 19 of 60, and the two margins of error overlap, so the rise cannot be distinguished from chance. Unprompted mentions rose from 3 to 5, which is a movement of two people in a sample of sixty and is not a trend. Online reviews rose from 23 to 31 with the average moving from 4.7 to 4.8, which is a real increase in volume and a stable level of satisfaction. Prospects who knew Nordec before the first contact rose from 5 of 15 to 8 of 14, which is the measure that most closely tracks the work, because it asks whether the name arrives before the first conversation. The customer survey gave a net promoter score of 56.1, measured once, with no earlier value to compare it against.

What the case shows most clearly is not a result but a habit. The inventory that found 36 touchpoints is now a living document. The templates are in use, which is why consistency moved. The brand keeper checks new materials weekly, which is why the drift has not returned. The scorecard is read on a schedule, with its sample sizes and its margins of error written beside its figures. None of those things is a campaign, and all of them are what a brand is made of in a business of this size: a promise, a system that repeats it, and a person who notices when it stops.

CASE IN POINT Nordec Entretien inc.: what Sophie would do differently

Nordec Entretien inc. is a fictional business created for this encyclopedia; its owner, its employees, its customers, its competitors, its figures and the events described are invented, and any resemblance to a real business is coincidental. Asked what she would change, Sophie Nadeau names three things, and all three are about order rather than about money. The first is the inventory: she would count the touchpoints before calling any designer, because the count changed what she asked for. The first logo was drawn by a friend on a template site in 2022, and four years later the business was carrying three versions of it, five blues and four typefaces across 36 touchpoints, with only 13 of them consistent. The second is the search: she would run the knockout search before the team grew attached to a name, because the twelve candidates narrowed to four on facts rather than on preference, and the narrowing was easier before anybody had a favorite. The third is the vans: she would order them last. The two wraps cost $8,200 and they were the most expensive single item in the rollout, and she is glad they were applied after the palette was settled rather than before it. What she would not change is the order of the rollout itself: documents and digital first, then vehicles and uniforms, then the depot sign.

IN LA MACHINE Carrying the sixty-day plan

The application copy read for this book carries the plan in pieces rather than in one place (La Machine, 2026). Projects at productivity/projects, Tasks at productivity/tasks and the Calendar at productivity/calendar can hold the four blocks as a project with dated tasks, and in the Control center the Plan tasks page at /taches gives each task a verification criterion, which is the platform's version of the check column in Table 13.2. Forms at business/marketing/forms can carry a short survey, such as a likelihood-to-recommend question, with the answers kept with the form, which is where the first scorecard reading can live. My Website at business/website holds a Theme & Branding panel with a gallery of 28 themes, a brand name, a tagline, a logo, a favicon and a banner image, five color overrides and a heading and body font chosen from a catalog of 200, and it keeps that configuration in the browser. Business → Branding at business/branding offers a logo drop zone, a brand name, a tagline, a website address, a primary and a secondary color with ten preset swatches, a choice among five fonts, an email footer and three social links, and it computes the contrast of the primary color with a pass or fail badge at the 4.5:1 threshold. Its values are not saved: the save button shows a confirmation and no other page uses what was entered. It is a place to try a palette and see its contrast, not a brand register.

Costly Mistakes

• Starting with the logo. A mark chosen before the platform has no criteria to satisfy, so it is judged by preference, and it is replaced once the business discovers what it is actually promising.
• Letting the touchpoints drift. Three logo versions, five blues and four typefaces across 36 touchpoints leave a customer looking at what appears to be several companies, and the only thing that resolves the confusion is a price.
• Choosing a name before searching. A knockout search costs an afternoon and a rename costs every material the name has already reached, which is why the search happens before the team becomes attached.
• Paying for a design without a written assignment. The Copyright Act makes an assignment valid only if it is in writing and signed, and moral rights are waived rather than assigned, so the signature belongs before the last payment and not after it.
• Rolling out the vans first. The most expensive and least reversible touchpoint freezes a palette and a letterform that may still be moving, and a wrap cannot be corrected with a new template.
• Treating the guide as a document. A guide that is admired once and never opened has not changed anything; a system of tokens and templates changes what every new document looks like without anybody consulting it.
• Reading a small sample as a result. A rise of ten points on a survey of sixty, with a margin of error of about ten points, cannot be distinguished from chance, and a decision taken on it is a decision taken on noise.

Key Takeaways

• Seven principles carry the book: a brand is a promise kept at every touchpoint, the platform comes before the logo, names are checked before they are loved, accessibility is part of the identity, a system beats a set of files, a business should evolve before it revolts, and measurement must use honest samples.
• The ten most costly mistakes are all mistakes of order rather than of taste, and each one has a remedy that costs less than the mistake: the platform before the mark, the search before the attachment, the documents before the vans, the assignment before the last payment.
• The sixty-day plan runs in four blocks of fifteen days, each ending with a deliverable somebody can check: the inventory and the platform, the names and the brief, the identity and the tokens, the rollout and the first measures.
• The plan cost the case $11,900 of outside spending inside the sixty days and $12,220 of production scheduled after day 60, a total of $24,120, or 5.4% of the $447,012 run-rate, plus 64 hours of in-house time at $36, or $2,304.
• The scorecard is read at day 60 and six months later, and its limits are stated with its figures: a sample of 60 carries a margin of error of about ±10.4 points, consistency is counted rather than sampled, and a brand project never runs alone.
• Six months after the rollout the case held 56 buildings, consistency had moved from 13 of 36 touchpoints to 33 of 36, and the book does not attribute the growth to the brand, because the offer, the Bilan Nordec and referrals moved at the same time.

Your Action Plan

1. Count the touchpoints before calling anyone. List every place a customer meets the business, collect a sample of each, note the logo version, the colors, the typefaces and the voice, and rank them by how often customers see them. Deliverable: one ranked inventory with a sample and a consistent-or-not mark for every touchpoint. Verification: the count is higher than the owner's first guess, and every touchpoint has a sample in the folder.
2. Write the platform in two half-days and keep it to one page. Hold the workshop with the whole team, produce a purpose, a promise, three values with the decision each one changes, four personality traits and a tagline in both languages. Deliverable: a one-page platform. Verification: it can be read aloud in a meeting without anyone asking what a sentence means.
3. Search the name before the team falls in love with it. Write the naming criteria from the platform, propose candidates, screen them for language and meaning, and have a trademark agent run a knockout search on the survivors. Deliverable: a documented search and one chosen name. Verification: the search is written down, and the name works in French and in English on a telephone.
4. Sign the brief and the assignment before the last payment. Put the platform, the ranked inventory, the deliverables, the fee, the schedule and the rights into one short brief, and require a written assignment of copyright and a waiver of moral rights. Deliverable: a signed brief and a signed assignment. Verification: the assignment is in writing and signed by the owner of the right, and the source files are delivered.
5. Roll out documents and digital first, and take the first reading before you finish. Replace the templates, the website and the email signatures, publish the one-page guide, write one line to the customers saying what has not changed, and take the scorecard's first reading with its sample size and margin of error recorded. Deliverable: live templates, a published guide and a dated baseline. Verification: the materials in use are the templates, and the baseline states how many people were asked.

In La Machine

The application copy read for this book carries the sixty-day plan in pieces rather than in one place (La Machine, 2026). Projects at productivity/projects, Tasks at productivity/tasks and the Calendar at productivity/calendar can hold the four blocks as a project with dated tasks, and in the Control center the Plan tasks page at /taches gives each task a verification criterion, which is the platform's version of the check column in Table 13.2. Forms at business/marketing/forms can carry a short survey, such as a likelihood-to-recommend question, with the answers kept with the form, which is where a first scorecard reading can live. My Website at business/website holds a Theme & Branding panel with a gallery of 28 themes, a brand name, a tagline, a logo, a favicon and a banner image, five color overrides and a heading and body font chosen from a catalog of 200, and it keeps that configuration in the browser. Business → Branding at business/branding offers a logo drop zone, a brand name, a tagline, a website address, a primary and a secondary color with ten preset swatches, a choice among five fonts, an email footer and three social links, and it computes the contrast of the primary color with a pass or fail badge at the 4.5:1 threshold, with a live preview of a client invitation card and a client email. Its values are not saved: the save button shows a confirmation and no other page uses what was entered. It is a place to try a palette and see its contrast, not a brand register.

What the platform does not do belongs in the same breath, and the list is long enough to be worth stating plainly. La Machine has no saved brand kit that other pages read, so the palette and the logo entered on the Branding page do not follow the business into its documents. It has no brand portal and no guidelines publisher, so the one-page guide lives in a document rather than in a page. It has no digital asset management with versions, usage rights or expiry dates, so the shared asset folder is kept outside it. It has no brand consistency audit, which is why the weekly check is a person's task rather than a report. It has no brand health tracking, no awareness survey and no net promoter tool beyond the general form builder, so the scorecard is assembled by hand. It has no trademark search and no naming tool, so the knockout search belongs to a trademark agent. And it has no image generation in the product. The Companies page at business/marketing/companies and the Contacts page at business/marketing/contacts hold one record per customer and the people attached to it, with notes, which is where the customers who receive the announcement letter are listed; Book N1 maps what the platform covers, and Book K1 (The SME Technology Stack) treats the rest of the stack.

Self-Check

1. Why does the order of the work matter more than the budget in a brand project, and which principle does that order express?
2. What are the ten most costly branding mistakes, and what do the first three of them have in common?
3. What does each of the four blocks of the sixty-day plan deliver, and what is the check that confirms a block is finished?
4. What does the Copyright Act require for a business to own the work it has paid for, and when should the signature happen?
5. What can the scorecard show six months after a rollout, and what can it not show?

Answers. 1. The order matters because each step supplies the criteria for the next: a platform gives the designer a specification, a specification gives the templates their values, and the values give the rollout its consistency, so a business that starts with the logo has no criteria and judges the mark by preference; the principle is that the platform comes before the logo. 2. The ten are starting with the logo, letting the touchpoints drift, choosing a name without searching, writing a promise the business cannot keep, treating the guide as a document, giving every offer its own brand, rolling out the vans first, leaving the rights with the supplier, measuring with a sample too small to read, and rebranding as a revolution; the first three are all mistakes of sequence, made by doing something visible before doing something that would have decided it. 3. Days 1 to 15 deliver a ranked inventory and a one-page platform, checked by confirming that every touchpoint has a sample and a mark and that the platform can be read aloud; days 16 to 30 deliver a named offer, a voice, a documented search, a signed brief and a chosen studio, checked by the search record and the brief's terms; days 31 to 45 deliver the identity, the tokens, the templates and the architecture, checked by the measured contrast ratios; days 46 to 60 deliver live templates, a published guide and a first scorecard, checked by confirming that the materials in use are the templates and that the scorecard states its sample. 4. The Copyright Act provides that the author of a work is its first owner, that an assignment or grant of copyright is valid only if it is in writing and signed by the owner of the right, and that moral rights cannot be assigned but can be waived, so the written assignment and the waiver are signed before the last payment rather than after it (Copyright Act, R.S.C. 1985, c. C-42, ss. 13 and 14.1). 5. It can show what the business counts rather than samples, such as the share of touchpoints that are consistent, and it can show the direction of the market measures with their limits stated; it cannot show that the brand caused a change in the business, because the offer, the entry product and referrals move at the same time, and a movement smaller than the margin of error cannot be distinguished from chance.

Chapter References

Binet, L., & Field, P. (2013). The long and the short of it: Balancing short and long-term marketing strategies. Institute of Practitioners in Advertising.

Charter of the French language, CQLR c C-11 (Québec). LégisQuébec, updated to June 10, 2026 — as read for Book C2 (Websites, Landing Pages and Conversion); its rules on trademarks, signs, products and business names are stated in this book through the explanations of the Office québécois de la langue française and of the Registraire des entreprises.

Copyright Act, R.S.C. 1985, c. C-42, ss. 13 and 14.1 (Canada). Justice Laws Website, current to September 3, 2026 (last amended November 7, 2024) — read September 25, 2026.

La Machine. (2026). Documentation — route reference inventory. Internal product documentation, read September 25, 2026, together with the code of the pages this book describes: productivity/projects, productivity/tasks, business/marketing/forms, business/website and business/branding.

Muzellec, L., & Lambkin, M. (2006). Corporate rebranding: Destroying, transferring or creating brand equity? European Journal of Marketing, 40(7/8), 803–824. https://doi.org/10.1108/03090560610670007

Walsh, M. F., Winterich, K. P., & Mittal, V. (2010). Do logo redesigns help or hurt your brand? The role of brand commitment. Journal of Product & Brand Management, 19(2), 76–84. https://doi.org/10.1108/10610421011033421

World Wide Web Consortium. (2024). Web Content Accessibility Guidelines (WCAG) 2.2. W3C Recommendation, 12 December 2024 (first published as a Recommendation on 5 October 2023). Retrieved September 25, 2026, from https://www.w3.org/TR/WCAG22/