Chapter 1 — Definition, Landscape and Stakes

You run a good business. Customers who find you tend to stay. The problem is the word find: most weeks, they find you by luck, by referral or by a search you never planned for — and when the luck runs dry, so does the pipeline.

What You'll Be Able to Do

• Define marketing rigorously, and separate it from selling, advertising, branding and public relations.
• Describe the marketing landscape your business operates in today: channels, platforms, artificial intelligence and the economics of attention.
• Name the challenges and constraints — time, budget, skills, technology, legal rules — that shape what you can realistically do.
• State what your business needs from marketing, in five outputs you can count.
• Read this book efficiently, knowing which chapter answers which question and how each chapter connects to La Machine.

Why It Matters

Marketing is the function that creates a customer — not the department that spends money on logos. When it is left to chance, revenue follows the mood of the market; when it is run as a system, revenue follows a plan you can inspect and correct. For the owner, that is the difference between hoping and steering. For the team, it is the difference between random tasks and a rhythm everyone understands.

In This Chapter

• 1.1 What Marketing Is — and Is Not
• 1.2 The Landscape in 2026
• 1.3 The Owner's Challenges
• 1.4 What a Business Needs from Marketing
• 1.5 Constraints
• 1.6 How This Book Is Organized

1.1 What Marketing Is — and Is Not

Ask ten business owners what marketing is and you will hear ten answers: "advertising", "social media", "our brand", "the website", "the thing the agency does". Each answer names a piece of marketing and mistakes it for the whole. The confusion is expensive, because a business that defines marketing as "posting on social media" will measure posts, budget for posts and hire for posts — and wonder why revenue does not move.

1.1.1 The definition

DEFINITION Marketing The American Marketing Association defines marketing as "the activity, set of institutions, and processes for creating, communicating, delivering, and exchanging offerings that have value for customers, clients, partners, and society at large" (American Marketing Association, n.d.). In plain terms: marketing is everything a business does to understand what a group of people value, to build an offer that delivers it, to make that group aware of the offer, and to make buying it easy and worthwhile — repeatedly, and at a cost the business can sustain.

Three words in that definition deserve a closer look, because each one corrects a common error.

Processes. Marketing is not an event (a launch, a campaign, a trade show) but a set of processes that run continuously: learning about customers, shaping the offer, communicating, converting, keeping. A process has inputs, steps, outputs and a cadence. That is why Chapter 11 of this book is a table of tasks and not a list of ideas.

Creating. Marketing starts before the product exists. The choice of whom to serve, which problem to solve and how to package the solution is a marketing decision — arguably the most important one. Peter Drucker made the point seventy years ago when he argued that the purpose of a business is to create a customer, and that a business therefore has two basic functions, marketing and innovation; everything else is a cost (Drucker, 1954). A company that treats marketing as "promotion after the fact" has skipped the part where the money is decided.

Exchanging. Marketing ends in an exchange: money for value, attention for content, a contact detail for a useful document. If nothing is exchanged, nothing was marketed — it was merely broadcast. This is why consent (Chapter 10) belongs to marketing fundamentals and not to a legal appendix: the exchange has to be wanted on both sides.

Philip Kotler and Kevin Lane Keller offer a shorter managerial definition: marketing management is the art and science of choosing target markets and getting, keeping and growing customers through creating, delivering and communicating superior customer value (Kotler & Keller, 2016). Notice the verbs — choosing, getting, keeping, growing. Notice, too, that the definition assumes you will not try to serve everyone. Chapter 3 is devoted to that choice.

KEY CONCEPT Marketing is a system, not a channel A channel (email, search, a trade show, a referral partner) is a pipe. Marketing is the plumbing: the decisions about whom to reach, with what offer, through which pipes, at what cost, measured how. A business can change channels every year and keep the same system. A business that has no system will buy pipes forever and never know where the water goes. Everything in this book — segments, personas, positioning, the mix, alignment with sales, compliance — is a component of that system. Figure 1.1 draws it.

Figure 1.1 — The marketing system. Inputs (data, content, offers, channels, budget and time) feed an engine made of strategy, execution and measurement, which produces countable outputs; a learning loop feeds what the market says back into the inputs. Original figure.

1.1.2 What marketing is not

The neighboring disciplines are real and valuable, but they are not synonyms. Table 1.1 separates them by the question each one answers, the time horizon it works on and the way its result is measured.

Table 1.1 — Marketing, sales, branding, advertising and public relations: what each one is

Two boundaries matter most for a small business.

Marketing versus selling. Selling persuades a specific person to buy now. Marketing makes a class of people want to buy, and makes selling easier, cheaper or in some cases unnecessary. Theodore Levitt drew the line sharply: selling focuses on the needs of the seller — converting product into cash — while marketing focuses on the needs of the buyer, on satisfying them through the product and the whole cluster of things associated with creating, delivering and consuming it (Levitt, 1960). In a business with one owner and no marketing function, the owner is usually selling very hard to compensate for the marketing that does not exist. Chapter 9 explains how the two functions hand work to each other.

Marketing versus advertising. Advertising is one paid channel among many. It is visible, which is why people confuse it with marketing, and it is optional, which is why a business can market for years without buying a single advertisement — through referrals, search, content, email and partners. Chapter 8 places advertising among the other channels; Collection D explains how to buy it.

DEFINITION Demand generation Demand generation is the set of marketing activities that create awareness and interest in a category and an offer among people who were not actively looking for it, and that convert that interest into identifiable leads. It differs from demand capture — activities such as search advertising or search engine optimization that intercept people who are already looking. A business that only captures demand depends on the size of the existing market; a business that also generates demand can grow the market it sells into. Both are covered in Chapter 2 and Chapter 8.

1.1.3 Marketing in a business that "does not do marketing"

Every business markets, whether or not anyone owns the function. The sign on the door, the way the phone is answered, the invoice layout, the delay before a quote is sent, the follow-up that never comes — each one communicates value, or its absence, to a customer. The question is never whether to market but whether marketing is designed or accidental.

CASE IN POINT A twelve-person distributor with no marketing function Fictional example; names and figures are invented. Northline Supply distributes industrial fasteners to about 240 active accounts. Revenue: $3.4 million CAD. Nobody holds a marketing title; the two owners "know everyone" and the three inside sales representatives handle inbound calls. Growth has been flat for four years. A quick audit shows: no written description of the ideal customer; a website last updated in 2019 that does not mention the two product lines that now make 40% of margin; 1,900 contact emails accumulated over a decade with no record of consent; no follow-up when a quote is not accepted (about 55% of quotes); no measurement of where new accounts come from. Northline is not "bad at marketing" — it has no marketing system, so results depend on the owners' memory and on customers calling first. Chapters 2 through 10 of this book are, in effect, the plan to build that system; Chapter 14 sequences it over ninety days.

1.2 The Landscape in 2026

Marketing's definition has been stable for decades. Its landscape — where attention lives, what it costs, who controls the channels — changes every year. This section describes the terrain as of September 2026 so that the rest of the book can concentrate on principles that outlast it.

1.2.1 Most businesses are small, and most small businesses are tiny

The starting point is a fact about the population of businesses. As of December 2024, Canada counted about 1.10 million employer businesses, of which 1.08 million (98.2%) were small businesses with 1 to 99 employees; medium-sized businesses (100 to 499 employees) numbered 16,953 (1.5%) and large businesses (500 or more) 3,380 (0.3%). Within that population, micro-enterprises with 1 to 4 employees represented 59.1% of all businesses. Small businesses employed 5.8 million people, 46.6% of the private labor force, and produced 33.2% of private-sector gross domestic product in 2022 (Innovation, Science and Economic Development Canada, 2025).

Why does this matter for marketing? Because almost everything written about marketing — textbooks, conference talks, software demos — is written for the 1.8% of businesses that have a marketing department. The other 98.2% have an owner with a marketing problem. Their constraints are different: no dedicated staff, no agency retainer, a budget measured in hundreds or a few thousand dollars per month, and a decision-maker whose time is the scarcest resource in the company. This book is written for that reader, and every method in it is checked against a simple question: can a solo owner, a small team or an established company each do this, at their scale?

1.2.2 The digital baseline

Digital tools are no longer a differentiator; they are the floor. Statistics Canada's Survey of Digital Technology and Internet Use found that in 2023, 48% of Canadian businesses with five or more employees used cloud computing, 32% received e-commerce orders, and 28% offered telework. Artificial intelligence was still rare: 7% of businesses used software or hardware with AI, but the gap by size was wide — 26% of large businesses against 6% of small ones — and about three in four AI users were already using generative AI (Statistics Canada, 2024).

Read those numbers as a map of opportunity rather than a scorecard. Two thirds of businesses did not take orders online; more than nine in ten small businesses had not adopted AI in their operations. A small business that runs a coherent marketing system — clean data, consent-based email, a website that converts, a measured pipeline — is not catching up with its competitors. In most local and B2B markets, it is ahead of them.

1.2.3 The channel map has fragmented — and consolidated

Twenty years ago a local business reached customers through a handful of channels: word of mouth, the Yellow Pages, local print, radio, direct mail, trade shows. Today the list is long: search engines, marketplaces, social platforms (each with its own formats, algorithms and advertising auctions), email and SMS, messaging applications, video platforms, podcasts, review sites, online communities, influencers, webinars, partner programs, and AI assistants that answer questions directly. Chapter 8 classifies them; Collections C and D explain how to work each one.

Two forces pull in opposite directions. Fragmentation means attention is spread across more places than any small business can cover; the only sane response is to choose — three channels done well beat ten done badly (Chapter 8). Consolidation means a few platforms control access to most of that attention and set the price of it through auctions. Access you do not own (a social following, a search ranking, a marketplace listing) can change overnight when a platform changes its rules. That is why this book insists on owned assets — your contact list, your website, your content, your customer data — as the foundation of the system (→ see Book C1, Direct Marketing; → see Book C2, Websites, Landing Pages and Conversion).

1.2.4 The economics of attention

Herbert Simon observed in 1971 that in an information-rich world, the wealth of information means a dearth of something else — the attention of its recipients — so that a wealth of information creates a poverty of attention (Simon, 1971). Fifty-five years later that observation is the operating condition of every marketer. Your prospective customer sees more commercial messages in a day than they can evaluate in a week. Attention is therefore rationed, and it is rationed by relevance: a message about a problem the reader has, arriving at the moment they have it, from a source they have reason to trust.

Seth Godin's answer, formulated in 1999, still stands: permission marketing — the privilege, not the right, of delivering anticipated, personal and relevant messages to people who actually want to get them (Godin, 1999). The alternative, which Godin called interruption marketing, gets more expensive every year as attention gets scarcer, and it is increasingly regulated (Section 1.5 and Chapter 10). Permission is therefore not only ethical and legal; it is the cheaper strategy over any horizon longer than a quarter.

1.2.5 Buying behavior has changed more than selling has

Buyers do their own research before they talk to a seller. McKinsey's work on the consumer decision journey showed that buyers do not narrow a set of options in a straight funnel; they add and remove brands during an active evaluation phase, and their decision is shaped by touchpoints the seller does not control — reviews, peers, search results (Court, Elzinga, Mulder, & Vetvik, 2009). In business-to-business markets the same pattern appears with more people involved: a buying group, not a buyer, that spends most of its process researching independently and consulting peers, and only a fraction of its time with suppliers' representatives (Gartner, 2023).

The consequence for a small business is direct: your marketing does most of the selling before the first conversation. If the website does not answer the buyer's questions, if there are no reviews, no case studies, no clear price signal, no proof — the sales conversation starts from a deficit, or never starts at all. Chapter 4 (personas and buying behavior) and Chapter 5 (positioning and proof) address this directly.

1.2.6 Artificial intelligence: a tool, an intermediary and a competitor for attention

AI enters marketing in three roles. As a tool, it drafts content, analyzes data, segments audiences, scores leads and automates workflows (Book K4, Artificial Intelligence for Business; Book N11, The AI Assistant, Agents and Orchestration). As an intermediary, AI assistants and machine-written answers in search engines summarize what businesses say about themselves and decide which ones to mention — a new channel with its own rules (Book C5, Answer Engine and Generative Engine Optimization). As a competitor for attention, AI floods every channel with cheap, generic content, which lowers the value of generic content to zero and raises the value of specific, proven, first-hand content. The strategic implication is the same as in Section 1.2.4: relevance and trust win; volume does not.

WHAT CHANGED Generic content stopped working As of 2026, most channels are saturated with machine-written articles and posts that say nothing a buyer could not get elsewhere. The businesses that keep earning attention are the ones that publish what only they know — their own numbers, cases, methods and opinions — and that treat consent-based email and direct relationships as their core distribution. This shift began around 2023 and accelerated in 2024–2025; it is the reason this library favors specific, tooled, actionable content over generic e-books.

1.3 The Owner's Challenges

The landscape explains what is possible. The owner's situation explains what is hard. Six challenges come up in almost every small business, regardless of sector.

1.3.1 Time

The owner is the bottleneck. In a solo or very small business, the person who should design the marketing system is also the person delivering the service, sending invoices and answering the phone. Marketing gets the leftover hours, and there are none. The result is episodic marketing: a burst of activity when sales dip, silence when the order book fills up — precisely the opposite of what a system requires, because leads generated during the burst arrive when there is no capacity to serve them, and the silence guarantees the next dip.

The remedy is not "find more time" but cadence and delegation: a small set of recurring tasks, scheduled and, where possible, automated, so that the system runs at a steady rate whether or not the owner is available that week. Chapter 11 turns this into a task table; Book J7 (Business Process Automation) and Book N3 (Workflows, Automation and Human Approval) show how La Machine executes it.

1.3.2 Budget

Small businesses cannot outspend anyone. What they can do is spend on the right things in the right order. Three mistakes waste most small-business marketing budgets: paying for reach before the offer and the message are clear (Chapter 5); paying for new leads while existing leads and past customers are ignored (Chapter 2, Chapter 9); and paying for tools before the process they should support exists (Chapter 13). A budget that follows the order of this book — segments and positioning first, owned channels second, paid channels third, tools as needed — spends less and returns more. Book B4 (The Marketing Plan, Budget and KPIs) builds the budget itself.

1.3.3 Noise and advice

The owner is also a target — of marketing. Every platform, agency and software vendor sells a channel as the answer. Advice arrives in contradictory bursts: "you need to be on video"; "email is dead"; "just run ads"; "SEO takes years, forget it". Without a framework to evaluate advice, the owner either freezes or follows the most recent voice. This book is that framework: a channel is evaluated on reach to your segments, cost, fit with your offer, control and time to results (Chapter 8) — never on how loudly it is recommended.

1.3.4 Measurement

Most small businesses cannot answer three questions: where do our customers come from?, what does a new customer cost us?, and how many prospects are in our pipeline right now? Without those answers, marketing decisions are guesses and every budget conversation is a matter of opinion. The cure is modest: define the stages of your funnel (Chapter 2), record the source of every lead (Chapter 9), and count the transitions between stages every week. Advanced analytics can come later (Book B7); counting cannot wait.

DO THE MATH What an unmeasured pipeline costs Illustrative figures. A consulting firm receives about 20 inquiries a month and signs 4 clients, at an average first-year value of $9,000 CAD. It does not track sources or follow up on the 16 inquiries that did not sign. Suppose that a simple follow-up sequence — three messages over three weeks — recovers just 1 additional client in 16 (a 6% recovery rate, conservative in most services markets). That is 12 additional clients a year × $9,000 = $108,000 of annual revenue from contacts the firm had already paid to attract. The cost of the sequence: about two hours to write it once, and the messaging tool the firm already pays for. The cost of not measuring was invisible — which is exactly why it persisted.

1.3.5 Skills

Modern marketing requires strategy, writing, design, data handling, channel know-how and technical configuration. No owner has all of them; few small businesses can hire all of them. The realistic path is to own the strategy and the customer knowledge (nobody knows your customers better than you), to standardize execution so it can be delegated or automated, and to buy the rest as needed — a freelancer, an agency, a fractional specialist. Chapter 12 maps which skills belong where in each situation (solo, small team, established) and how to brief external help so you keep control of your accounts and your data.

1.3.6 Compliance

Marketing is regulated. Sending a commercial email, collecting a contact form, publishing a claim, running a contest or placing a tracking pixel on a website each triggers rules — anti-spam law, privacy law, competition and consumer protection law, language law in Québec. Small businesses are not exempt, and the penalties are not symbolic (Section 1.5.1). Compliance is treated here as a design requirement of the marketing system — a filter applied before every send (Chapter 10) — not as a legal appendix. Designed in from the start, it costs very little; retrofitted after a complaint, it costs a great deal.

1.4 What a Business Needs from Marketing

If marketing is a system, its outputs should be nameable and countable. Every business — whatever it sells — needs five things from its marketing function. They are the outputs in Figure 1.1, and every chapter of this book contributes to at least one of them.

1.4.1 Awareness

People who could buy from you must know you exist and know what for. Awareness is not fame; it is the share of your target segments (Chapter 3) that would recognize your name and associate it with the problem you solve. For a regional B2B supplier that may mean 400 companies; for a neighborhood clinic, 6,000 households. Awareness is built by consistent presence in the channels those people actually use (Chapter 8), with a message that says one clear thing (Chapter 5). It is measured, at small scale, by direct traffic, branded searches, mentions and by simply asking new customers how they heard of you.

1.4.2 Demand

Awareness that never turns into inquiries is decoration. Demand is the flow of people who take a step toward you: a form submitted, a call, a quote requested, a document downloaded, a reply to an email. It is the first countable output of the system, and it should be counted by source and by stage (Chapter 2, Chapter 9). Demand has two origins — generated (people you made interested) and captured (people who were already looking) — and a healthy system does both (Section 1.1.2).

1.4.3 Trust

Between interest and purchase sits risk: will this work for me? can I trust these people? what if it goes wrong? Trust is manufactured deliberately — with proof (numbers, cases, testimonials, guarantees, credentials), with consistency (brand, Chapter 7), with responsiveness (speed to lead, Chapter 9) and with honesty (Chapter 10). Trust is the output most small businesses under-invest in, because it feels intangible; in practice it shows up as a higher conversion rate at every stage and a shorter sales cycle.

1.4.4 Retention and growth of existing customers

The cheapest customer to win is the one you already have. Marketing to existing customers — onboarding, education, cross-sell, renewal, reactivation, referral programs — usually costs a fraction of acquisition and compounds over time. It is a marketing responsibility, not only a customer-service one, and it needs the same system: segments (who is due for what), messages, cadence and measurement (Book B10, Customer Experience, Retention and Loyalty; Book L5, Subscription and Recurring Revenue Models).

1.4.5 Data and learning

The fifth output is knowledge: which segments respond, which messages convert, which channels produce customers rather than clicks, which objections recur. Marketing is the business's sensor toward its market. A system that does not capture what it learns will repeat its mistakes at full price. The learning loop in Figure 1.1 is not optional — it is where the compounding happens.

KEY CONCEPT Five outputs, one system Awareness, demand, trust, retention and learning are not five projects; they are five outputs of the same system, produced by the same components (segments, offer, message, channels, cadence, measurement). A business that pursues one — usually demand — while ignoring the others gets an expensive, leaky funnel. The order of this book follows the order in which the components must be built: Chapter 2 (the system and its stages), Chapters 3–5 (whom, what, how to say it), Chapter 6–8 (the mix, the brand, the channels), Chapter 9 (the hand-off to sales), Chapter 10 (the rules), Chapters 11–14 (how to run it).

1.5 Constraints

A constraint is a boundary you do not choose. Knowing yours before designing the system saves you from plans that cannot be executed. Five constraints frame every small-business marketing system.

1.5.1 Legal constraints

Marketing law in Canada rests on four pillars that every owner should be able to name. The body of this book states principles that apply in most jurisdictions; the specific rules, thresholds and penalties are stated in Québec & Canada Notes, dated and sourced, because they change.

QUÉBEC & CANADA NOTE The four rules every marketer must know (as of September 2026) 1. Anti-spam. Canada's anti-spam legislation — An Act to promote the efficiency and adaptability of the Canadian economy by regulating certain activities that discourage reliance on electronic means of carrying out commercial activities, S.C. 2010, c. 23 (CASL) — prohibits sending a commercial electronic message unless the recipient has consented (express or implied), the message identifies the sender and how to contact them, and it contains a working unsubscribe mechanism (CASL s. 6). The maximum administrative monetary penalty is $1,000,000 per violation for an individual and $10,000,000 for any other person (CASL s. 20(4)), enforced by the Canadian Radio-television and Telecommunications Commission (CRTC). → Chapter 10; → see Book H3. 2. Privacy. In Québec, the Act respecting the protection of personal information in the private sector, CQLR c P-39.1, as reformed by Law 25, governs how a business collects, uses and discloses personal information — including contact data used for marketing. The Commission d'accès à l'information (CAI) may impose administrative monetary penalties of up to $50,000 for a natural person and, in all other cases, $10,000,000 or 2% of worldwide turnover for the preceding fiscal year, whichever is greater (P-39.1 s. 90.12); penal provisions carry higher maximums (s. 91). Federally, PIPEDA applies to businesses in the other provinces and to interprovincial and international flows. → Chapter 10; → see Book H2. 3. Truthful marketing. The Competition Act, R.S.C. 1985, c. C-34 makes false or misleading representations to the public a reviewable practice (s. 74.01). For a corporation, the administrative monetary penalty may reach the greater of $10,000,000 ($15,000,000 for each subsequent order) or three times the value of the benefit derived from the deceptive conduct or, if that cannot be reasonably determined, 3% of annual worldwide gross revenues (s. 74.1(1)(c)), enforced by the Competition Bureau. Québec's Consumer Protection Act, CQLR c P-40.1 adds rules for consumer advertising. → Chapter 10; → see Book H4. 4. Language. In Québec, the Charter of the French language, CQLR c C-11 requires commercial publications — catalogs, brochures, websites — and public signage and commercial advertising to be in French, with specific rules on the presence and predominance of French; the Office québécois de la langue française (OQLF) administers it. → see Book H4. Verify the current text of each Act before acting; penalties and thresholds are as published in the consolidated statutes consulted on September 7, 2026.

The practical lesson is that consent, identification, honesty and language are design inputs of the system, not filters applied by a lawyer at the end. A contact record without a consent status is a liability; a claim without proof is a risk; a landing page without a French version is, in Québec, a violation. Chapter 10 shows how a small business documents all four with very little overhead.

1.5.2 Budget

The budget constraint is obvious; its shape is less so. Small-business marketing spending is lumpy — a website every five years, a trade show once a year — when the system needs it to be steady. A useful discipline is to separate three envelopes: foundations (one-time: positioning, website, brand assets, tooling), running costs (monthly: tools, content, email, a share of someone's time) and experiments (a fixed small amount to test a new channel with a stop rule). Book B4 turns this into a plan; here the point is that the budget's regularity matters as much as its size.

1.5.3 Skills and time

A system must be designed for the people who will actually run it. If the owner has four hours a week for marketing, the system must run on four hours a week plus automation — not on a plan that assumes a marketing manager. Chapter 11 sizes the recurring tasks; Chapter 12 decides who does what; Chapter 13 chooses tools that reduce the hours rather than add to them.

1.5.4 Technology

Technology constrains in two directions. Too little: no central contact database, no consent tracking, no way to send a segmented email — so the system cannot run. Too much: a dozen disconnected tools, each with its own list, so the data is inconsistent and the owner spends the four available hours reconciling spreadsheets. The rule in Chapter 13 is one source of truth for contacts, everything connected to it, free or already-paid tools first. In La Machine, that source of truth is the CRM and the marketing contact base (Section "In La Machine" below).

1.5.5 Brand and reputation

Finally, you are constrained by what people already believe about you. A business known for low prices cannot reposition as premium in a quarter; a business with poor reviews cannot advertise its way past them. Reputation is a slow variable: it takes years to build and is quick to lose (Chapter 7). Every marketing decision has to be consistent with the brand promise the business can actually keep.

WATCH OUT Constraints you discover after spending The costliest constraints are the ones found after the money is committed: discovering that your 4,000-contact list has no recorded consent after paying for an email platform; discovering that the agency owns the advertising account after paying it for a year; discovering that the new website cannot be edited without the developer. Each is avoided by one question asked before spending: what would I need to have in place for this to work, and who will own the result? Chapter 12 and Chapter 13 supply the checklists.

1.6 How This Book Is Organized

This is the first book of Collection B and the foundation of the 28 marketing books in Collections B, C and D. It defines the vocabulary they all use — funnel, journey, segment, ideal customer profile, persona, positioning, mix, channel, lead, consent — and it builds the system into which every later book plugs a component.

1.6.1 The spine

Every book in The Modern Business Encyclopedia follows the same spine, so that you always know where to look:

1. Chapter 1 — Definition, Landscape and Stakes (this chapter) says what the function is, where it stands today, what it must deliver and under which constraints.
2. Chapters 2 to 10 — the subject give the concepts, rules, strategies and methods. In this book they cover the marketing system (2), segments and the ideal customer profile (3), personas and buying behavior (4), positioning and the value proposition (5), the marketing mix (6), brand basics (7), the channels overview (8), marketing-and-sales alignment (9), and ethics, consent and compliance (10).
3. Chapter 11 — Typical Tasks, Cadence and Workflows gives the operating rhythm and the task table you can load into La Machine.
4. Chapter 12 — Expertise, Roles and When to Get Help says who does what, in a solo business, a small team and an established company, and when to hire, contract or automate.
5. Chapter 13 — Tools: Functions and Popular Tools lists the functions a marketing stack must perform, and the popular tools that perform them, with dated pricing models.
6. Chapter 14 — Practical Advice, Pitfalls and Your Action Plan gathers the advice that changes results, the costly mistakes, the 30-60-90-day plan and the master checklist.

1.6.2 The boxes

Twelve kinds of boxes punctuate the text. Each is labeled so you can skim for what you need: Key Concept (the idea to retain), Definition (a term, defined once, collected in the glossary), The Rule (an obligation, with its source and jurisdiction), Method (a step-by-step procedure), Do the Math (a fully worked calculation with illustrative figures), Case in Point (a fictional situation in a small business), Watch Out (a costly pitfall), Tools of the Trade (what a tool must do, and popular examples, dated), Who Does This (roles, in the three situations), In La Machine (the page or workflow that supports the point), What Changed (a recent change with its date) and Québec & Canada Note (a local rule, dated and sourced).

1.6.3 The closing sections

Every chapter ends the same way: Costly Mistakes, Key Takeaways, Your Action Plan (steps with deliverables and verification criteria), In La Machine (how the software supports the chapter, and honestly what it does not do yet), Self-Check (questions with answers) and Chapter References (APA 7). If you have twenty minutes for a chapter, read the opener and the closing sections; the body is there when you need the mechanism.

1.6.4 The three situations

Throughout, methods are checked against three situations: solo (you do everything, so automation and simplicity dominate), small team (you delegate and verify, so definitions and checklists dominate) and established (you steer managers and suppliers, so metrics and governance dominate). When the advice differs by situation, the difference is stated — often in a table with those three columns.

WHO DOES THIS Owning the marketing function In a solo business, the owner owns marketing strategy and delegates execution to automation and, occasionally, a freelancer; the minimum is one recurring block of time per week. In a small team, one person — often not full-time on marketing — owns the system, executes the recurring tasks and coordinates external help; the owner keeps the strategic decisions (segments, positioning, budget). In an established business, a marketing manager owns the system and its metrics, specialists or agencies execute channels, and the owner or general manager reviews results monthly against the plan. In all three cases, the owner never delegates the choice of whom to serve and what to promise — that choice is the business.

1.6.5 Where this book stops

This book explains what marketing is and how the system fits together. It deliberately does not operate any single channel in depth: the methods for email and direct marketing (C1), websites and conversion (C2), search (C3, C4, C5), content (B5), social media (C7), paid advertising (D7) and the others live in their own books, each of which assumes you have read this one. Likewise, market research (B2), branding (B3), the marketing plan and budget (B4), analytics (B7), automation and funnels (B8) and lead generation (B9) each get a full book. Where this chapter or the next ones touch those subjects, a cross-reference (→ see Book …) points the way.

Costly Mistakes

1. Equating marketing with one channel. The business that "does social media" measures posts and followers, spends its scarce hours feeding a platform it does not control, and never builds the list, the offer or the measurement that would turn attention into revenue. Consequence: years of activity, no compounding asset.
2. Marketing only when sales dip. Episodic bursts produce leads when there is no capacity to serve them and silence when the pipeline needs filling. Consequence: a revenue curve that oscillates with the owner's anxiety, and campaigns judged on the worst possible timing.
3. Buying reach before clarity. Paying for advertising or a new website before the segment, the offer and the message are defined multiplies the cost of every error by the size of the audience. Consequence: a typical small business can waste $5,000–$20,000 CAD in a quarter — the entire year's budget — learning what a two-day positioning exercise would have told it.
4. Ignoring consent until a complaint arrives. Accumulating thousands of contacts without recorded consent and then "sending a newsletter" exposes the business to CASL penalties of up to $10,000,000 per violation and to Law 25 penalties of up to 2% of turnover — and, more commonly, to being blocked by email providers and losing the list altogether. Consequence: the asset you built for ten years becomes unusable in a day.
5. Not measuring transitions. Counting visitors and followers but not inquiries, quotes, meetings and customers by source leaves every decision to opinion. Consequence: the follow-up worth six figures a year in the Do the Math box above never happens, because nobody sees the leak.
6. Delegating the strategy. Handing an agency the choice of whom to target and what to promise produces generic marketing for a generic business. Consequence: the money buys execution of a plan nobody in the company can defend, and the account, the data and the learning leave with the supplier.

Key Takeaways

• Marketing is the set of processes that creates a customer: understanding what a group of people value, building the offer, making them aware, making the exchange easy and worthwhile — repeatedly, at a sustainable cost.
• It is a system, not a channel. Channels change; the system — segments, offer, message, channels, cadence, measurement, compliance — is what you build and keep.
• Selling, advertising, branding and public relations are neighbors, not synonyms. Marketing makes selling easier; advertising is one optional paid channel; brand is the expectation people hold before they read a word.
• The landscape favors the small business that is systematic: 98.2% of Canadian employer businesses are small, most have not digitized their sales, attention is scarce and rationed by relevance, buyers research alone before they talk to anyone.
• Permission is the cheaper strategy. Consent-based, relevant marketing costs less over any horizon longer than a quarter — and it is what the law requires.
• The five outputs are countable: awareness, demand, trust, retention and learning. A system that pursues only demand leaks everywhere else.
• Constraints are design inputs: legal rules (CASL, Law 25 and PIPEDA, the Competition Act, the Charter of the French language), budget shape, skills and time, technology, reputation. Know them before you design.
• Never delegate the choice of whom to serve and what to promise. Everything else can be executed by a team member, a supplier or a machine.

Your Action Plan

The action plan below presents the five steps that turn this chapter into work, in the order in which each step consumes the output of the one before. For each step, the table gives the deliverable that must exist at the end, the timing on a small business's calendar, and the verification criterion by which anyone — a co-founder, a mentor, the owner's own future self — can tell that the step was done. The criterion column is the one to read first: a step whose criterion you cannot picture is a step you do not yet understand.

In La Machine

La Machine gathers the marketing system of a small business under Marketing. For this chapter, four pages matter.

• The marketing dashboard (business/marketing) is the home of the system: it shows the state of your contacts, lists and segments, the campaigns and sequences in flight, the email queue and the recent activity log — the "engine" and "outputs" boxes of Figure 1.1 on one screen. Use it for the weekly slot you booked in step 4 of the action plan.
• Sender identification — the legal name, mailing address and contact that every commercial message must carry under CASL — is a record of each line of business, and as read on September 15, 2026 the sending path refuses a message whose identification is incomplete; the marketing settings page (business/marketing/settings) is a placeholder, and sending channels are declared by the administrator. Confirming that the identification is complete is the first compliance task of the system (→ see Book N5).
• Contacts (business/marketing/contacts, with import at business/marketing/contacts/import) are the single source of truth for the people you market to; each contact carries five opt-in flags, one per channel, that only a documented consent can raise — the proof lives on the consent page (business/marketing/consent) — and a contact can be placed on the Do Not Reach list (business/marketing/do-not-reach), where the older address /exclusions now leads. The audit in step 2 of the action plan is largely a matter of importing your contacts and seeing how many carry a documented consent (→ see Book N4).
• The marketing log (business/marketing/log) records what left, to whom and when — and only what left, so refusals and results are counted elsewhere — the beginning of the learning loop.

What La Machine does not do: it does not write your marketing charter, choose your segments or define your positioning — those are decisions (Chapters 3 to 5) that the software then executes. It does not buy advertising: paid campaigns are run in the platforms' own tools and tracked here by campaign and source. And it does not replace the four legal rules with a checkbox — it enforces the ones it can (identification, unsubscribe, exclusions) and records the consent you are responsible for obtaining. Book N1 (Getting Started with La Machine) walks through the navigation; Book N5 (Marketing Automation in La Machine) covers the module page by page.

Self-Check

1. In one sentence, what distinguishes marketing from selling?
2. Name the five outputs a business needs from its marketing system.
3. Why is "permission marketing" described as the cheaper strategy, not only the ethical one?
4. A business has 4,000 email contacts collected over ten years with no record of consent. Which rule does sending them a newsletter engage, and what is the maximum administrative penalty for a business under that rule (as of September 2026)?
5. What is the one marketing decision the owner should never delegate, and why?

Answers. (1) Selling persuades a specific person to buy now; marketing makes a class of people want to buy and makes selling easier, cheaper or unnecessary (Section 1.1.2). (2) Awareness, demand, trust, retention and growth of existing customers, and data and learning (Section 1.4). (3) Because attention is scarce and rationed by relevance: messages people asked for convert at a higher rate and cost less per result than interruptions, whose cost rises every year and which are increasingly regulated (Sections 1.2.4 and 1.5.1). (4) CASL — sending a commercial electronic message requires consent, identification and an unsubscribe mechanism (CASL s. 6); the maximum administrative monetary penalty for a person other than an individual is $10,000,000 per violation (s. 20(4)) (Section 1.5.1). (5) The choice of whom to serve and what to promise — because that choice is the business; everything else is execution that can be delegated to a team member, a supplier or a machine (Section 1.6.4).

Chapter References

An Act to promote the efficiency and adaptability of the Canadian economy by regulating certain activities that discourage reliance on electronic means of carrying out commercial activities, S.C. 2010, c. 23 (Canada's Anti-Spam Legislation), ss. 6, 20. Justice Laws Website. Retrieved September 7, 2026, from https://laws-lois.justice.gc.ca/eng/acts/E-1.6/

Act respecting the protection of personal information in the private sector, CQLR c P-39.1, ss. 90.12, 91. LégisQuébec. Retrieved September 7, 2026, from https://www.legisquebec.gouv.qc.ca/en/document/cs/P-39.1

Charter of the French language, CQLR c C-11. LégisQuébec. Retrieved September 7, 2026, from https://www.legisquebec.gouv.qc.ca/en/document/cs/C-11

Competition Act, R.S.C. 1985, c. C-34, ss. 74.01, 74.1. Justice Laws Website. Retrieved September 7, 2026, from https://laws-lois.justice.gc.ca/eng/acts/C-34/

Consumer Protection Act, CQLR c P-40.1. LégisQuébec. Retrieved September 7, 2026, from https://www.legisquebec.gouv.qc.ca/en/document/cs/P-40.1

American Marketing Association. (n.d.). Definitions of marketing. Retrieved September 7, 2026, from https://www.ama.org/the-definition-of-marketing-what-is-marketing/

Court, D., Elzinga, D., Mulder, S., & Vetvik, O. J. (2009, June). The consumer decision journey. McKinsey Quarterly. McKinsey & Company.

Drucker, P. F. (1954). The practice of management. Harper & Row.

Gartner. (2023). The B2B buying journey. Gartner, Inc. Retrieved September 7, 2026, from https://www.gartner.com/en/sales/insights/b2b-buying-journey

Godin, S. (1999). Permission marketing: Turning strangers into friends, and friends into customers. Simon & Schuster.

Innovation, Science and Economic Development Canada. (2025). Key small business statistics 2025. Government of Canada. Retrieved September 7, 2026, from https://ised-isde.canada.ca/site/sme-research-statistics/en/key-small-business-statistics/key-small-business-statistics-2025

Kotler, P., & Keller, K. L. (2016). Marketing management (15th ed.). Pearson.

Levitt, T. (1960). Marketing myopia. Harvard Business Review, 38(4), 45–56.

Simon, H. A. (1971). Designing organizations for an information-rich world. In M. Greenberger (Ed.), Computers, communications, and the public interest (pp. 37–72). Johns Hopkins Press.

Statistics Canada. (2024, September 17). Survey of Digital Technology and Internet Use, 2023. The Daily. Retrieved September 7, 2026, from https://www150.statcan.gc.ca/n1/daily-quotidien/240917/dq240917c-eng.htm

Chapter 2 — The Marketing System

Last month your website had 1,000 visitors and you signed two customers. Was that good? You cannot say — and that is the problem this chapter solves.

What You'll Be Able to Do

• Draw your funnel with named stages and real counts, and read it as a measurement device rather than a picture.
• Explain the flywheel and use it to decide where existing customers add force to your system.
• Map a customer journey — stages, touchpoints, questions and opportunities — for one type of customer.
• Classify every channel you use as owned, earned or paid, and state what each class is for.
• Diagnose where your system leaks in sixty minutes, and know which chapter fixes each leak.

Why It Matters

A marketing system you can draw is a system you can improve. The funnel tells you where prospects are lost; the journey tells you why; the media classes tell you what you own and what you rent; the learning loop tells you whether last month's decision worked. Without these four views, marketing is a series of unrelated activities judged by feel. With them, it is a machine whose every part has a number, and every number has an owner.

In This Chapter

• 2.1 From Awareness to Advocacy: The Classic Funnel
• 2.2 The Flywheel and the Loop
• 2.3 The Customer Journey
• 2.4 Owned, Earned and Paid Media
• 2.5 Inputs and Outputs: Data, Content, Offers, Channels, Conversations
• 2.6 Feedback and Learning Loops
• 2.7 Diagnosing Your System: Where It Leaks

2.1 From Awareness to Advocacy: The Classic Funnel

2.1.1 A century-old model that still works

The funnel is the oldest model in marketing. Its ancestor is the AIDA sequence — attention, interest, desire, action — attributed to the American advertising pioneer E. St. Elmo Lewis at the end of the nineteenth century and popularized in the sales literature of the 1920s (Strong, 1925). The idea was simple: a buyer moves through mental states in order, and a message must accomplish each step before it can accomplish the next. Later generations added stages, renamed them and drew the sequence as a funnel, wide at the top and narrow at the bottom, because at every step some people drop out (Kotler & Keller, 2016).

DEFINITION Funnel A funnel is a model of the path from first contact to purchase (and beyond) as an ordered sequence of stages, in which the number of people at each stage is smaller than at the stage before. Its purpose is measurement: by counting how many people are at each stage and how many move to the next, you can see where the path leaks and what each customer costs. A funnel does not claim that real buyers move in a straight line — they do not (Section 2.3). It claims that, on average and over time, the counts at each stage are related, and that the relationships can be measured and improved.

A modern small-business funnel usually has five stages, sometimes six. The names vary; the logic does not.

1. Awareness is the stage of people who have been exposed to you: visitors to the website, viewers of a post, attendees at an event, recipients of a referral. It is counted as reach, visits or impressions.
2. Interest (or engagement) is the stage of people who did something that costs them a little: they read two pages, watched most of a video, opened three emails or followed you. It is counted as engaged visitors or subscribers.
3. Consideration (or lead) is the stage of people who identified themselves and signalled a possible need: they submitted a form, requested a quote, downloaded a guide, replied to a message or called. It is counted as leads.
4. Decision (or opportunity) is the stage of people in an active buying conversation: a meeting held, a proposal sent, a trial started. It is counted as opportunities or meetings.
5. Action (or customer) is the stage of people who bought. It is counted as customers and revenue.
6. Advocacy is the stage of customers who buy again, expand, review or refer. It is counted as repeat rate, referrals and reviews. Many small businesses omit this stage; Section 2.2 explains why they should not.

In business-to-business marketing, stage 3 is often split into lead, marketing-qualified lead (MQL) and sales-qualified lead (SQL), because the hand-off between marketing and sales happens there. Chapter 9 defines those terms precisely; here it is enough to see that a stage is a state a person is in, defined by an observable event, so that two people looking at the same contact would put it in the same stage.

2.1.2 Reading the funnel: stages, transitions and time

Three quantities describe a funnel.

Stage counts — how many people were at each stage in a period. They are the easiest to collect and the least useful on their own, because a large top means nothing if the bottom is empty.

Transition rates — the share of people who moved from one stage to the next: visitors who became engaged, engaged who became leads, leads who became opportunities, opportunities who became customers. Transition rates are the levers of the system. They tell you whether a problem is at the top (not enough people arrive), in the middle (people arrive but do not identify themselves), or at the bottom (leads arrive but do not buy).

Time between stages — how long a person typically spends between arriving and buying. In a local service business it may be days; in an industrial purchase, months. Time matters because a funnel that is measured monthly but takes a quarter to traverse will always look worse than it is.

KEY CONCEPT Measure the transitions, not the stages Most dashboards show stage counts: visitors, followers, leads. Those numbers grow with effort and say little about health. The health of a system is in its transition rates — what share of visitors become leads, what share of leads become customers — and in the time each transition takes. Two businesses with 1,000 visitors a month can have 2 or 20 customers depending on four ratios. When you report marketing, report the ratios; when you set goals, set them on the ratio you intend to move; when you test a change, judge it by the ratio it was supposed to change.

DO THE MATH Funnel arithmetic and the compounding of small improvements Illustrative figures. A firm attracts 1,000 website visitors a month. Of these, 12% engage (120), 25% of the engaged become leads (30), 20% of leads take a meeting (6) and 33% of meetings become customers (2). Monthly result: 2 customers, at an average first-year value of $12,000 CAD, so $24,000 of new annual revenue per month of marketing. Now improve each transition by one fifth — 12% → 14.4%, 25% → 30%, 20% → 24%, 33% → 40%. The arithmetic: 1,000 × 0.144 = 144 engaged; × 0.30 = 43.2 leads; × 0.24 = 10.4 meetings; × 0.40 = 4.1 customers — more than double, from the same 1,000 visitors. The improvements compound because they multiply (1.2⁴ ≈ 2.07). Compare the alternative of buying more traffic: to reach 4.1 customers by adding visitors alone, the firm would need 2,070 visitors — 1,070 more each month. At a typical paid-search cost of $3 to $8 per click for a professional service (illustrative), that is $3,200 to $8,600 a month, forever. Fixing the four ratios costs mostly attention: a clearer page, a better lead magnet, a faster reply, a stronger proposal. Fix the ratios before buying the traffic.

2.1.3 Where funnels leak, and why

Every funnel leaks; the question is whether the leaks are designed or accidental. A designed leak is a filter: people who are not your ideal customer (Chapter 3) should drop out early, at low cost to both sides. An accidental leak is a defect: a qualified person who wanted to continue and could not, or did not, because of something you controlled. The recurring accidental leaks in small businesses are:

• The unanswered question. The visitor could not find the price, the process, the delivery time or the proof, and left to find a competitor who states them. The cure is in Chapter 5 (proof) and in Book C2 (websites).
• The absent call to action. The page informs but never asks; the reader, willing, does nothing because nothing was proposed. The cure is in Chapter 5 and in Book B6 (copywriting).
• The slow reply. A lead arrived on Friday afternoon and was called on Tuesday. Speed to lead is one of the strongest predictors of conversion in most markets (Chapter 9).
• The single attempt. One call, no answer, no second attempt, no email — the lead is marked "not interested" when it was merely busy. The cure is in Chapters 9 and 11.
• The hand-off gap. Marketing produced a lead; nobody owned the next step; the lead expired in an inbox. The cure is in Chapter 9.
• The forgotten customer. The sale closed and the relationship ended; no onboarding, no follow-up, no ask for a review or a referral. The cure is in Section 2.2 and in Book B10.

None of these leaks is fixed by more traffic. All of them are visible in the transition rates — which is why the funnel, old as it is, remains the first drawing every owner should make.

2.2 The Flywheel and the Loop

2.2.1 Why the funnel is incomplete

The funnel has a structural flaw as a picture of the business: it ends. A customer falls out of the bottom and the model has nothing more to say about them. Yet in most businesses the customers already won are the largest source of the next customers — through repeat purchases, expansion, reviews and referrals. Treating them as the end of the process discards the cheapest growth available.

In 2018, Brian Halligan, co-founder of HubSpot, argued that companies should replace the funnel with a flywheel: a wheel that stores energy, in which every satisfied customer adds momentum, and in which friction — a poor hand-off, a broken promise, a slow response — slows the whole system down (Halligan, 2018). The flywheel is organized in three phases — attract, engage, delight — turning around the customer, and it makes two claims the funnel cannot: that customers are an input to marketing, not only its output, and that friction anywhere is a marketing problem, because it slows the wheel.

DEFINITION Flywheel A flywheel is a model of the marketing and customer system as a self-reinforcing loop rather than a one-way sequence: attracting prospects, engaging them into customers and delighting customers so that they buy again and bring others, which in turn attracts more prospects. Its two variables are force (everything that adds momentum: referrals, reviews, repeat business, word of mouth, useful content) and friction (everything that removes it: slow replies, unclear pricing, hand-off gaps, disappointing delivery). Figure 2.1 places the flywheel next to the funnel; the two are complementary — the funnel measures, the flywheel reminds you where the energy comes from.

Figure 2.1 — Funnel versus flywheel. Left, the funnel as a measurement device: stage counts and transition rates (illustrative figures). Right, the flywheel: three phases turning around the customer, accelerated by force and slowed by friction. Original figure; the flywheel adapts the concept described by Halligan (2018).

2.2.2 Force in a small business

For a business with fewer than fifty employees, force comes from four places, in roughly this order of leverage:

1. Repeat business. A customer who buys again costs almost nothing to acquire. A service firm whose customers renew at 80% needs to replace only one in five each year; at 50%, one in two. Retention is therefore the single largest determinant of how much acquisition the funnel must produce (→ see Book B10).
2. Referrals. A referred prospect arrives with trust already transferred; they convert at a higher rate and faster than a cold prospect in most markets, and they cost only the ask and the thank-you. Referrals rarely happen at scale without a system — a moment to ask, a way to make it easy, a reason to do it (→ see Book D1, Ambassador, Referral and Word-of-Mouth Programs).
3. Reviews and proof. Public reviews and documented cases do the work of a salesperson for every future prospect who reads them; they are force stored for later.
4. Content that customers share. A useful guide, a calculator, a checklist that a customer forwards to a peer extends your reach at no cost — and it only happens if the content is specific enough to be worth forwarding (Chapter 1, What Changed).

2.2.3 Friction, and why it is a marketing problem

Friction is the reason two businesses with identical marketing spend grow at different speeds. It hides in operations — the quote that takes four days, the invoice that is wrong, the technician who arrives late — and it is invisible in the marketing dashboard, which shows leads arriving as planned. Yet it destroys the flywheel's force: a customer who was disappointed does not refer, does not review, does not return, and often tells others. The marketing owner therefore has a legitimate interest in the parts of the business that touch customers after the sale; in a small company, that is most of them. Book J2 (Processes, SOPs and the Definition of Done) and Book B10 treat friction systematically; here, note only that the flywheel makes delivery quality a marketing input.

CASE IN POINT A landscaping company discovers its flywheel Fictional example; names and figures are invented. Verdant Grounds does residential landscaping and maintenance with a crew of nine. For years it spent about $18,000 CAD a season on advertising to win roughly 60 new maintenance contracts. When the owner finally counted, she found that 70% of new contracts came from referrals and neighbors who saw the crew at work — not from the advertising — and that 22% of maintenance customers did not renew, mostly citing "communication". She cut advertising to $6,000, spent $4,000 on a simple renewal-and-referral system (a spring call to every customer, a lawn sign with a QR code offering a neighbor discount, an automated review request two weeks after each project) and asked the crew leads to send a one-line message at the end of each visit. Two seasons later: renewals at 89%, 74 new contracts, 44% of them referred. Nothing in the service changed. The flywheel had always been there; nobody had been pushing it.

2.3 The Customer Journey

2.3.1 From stages to experience

The funnel counts people from the seller's point of view. The customer journey describes the same path from the buyer's — what they are trying to do, what they feel, where they look, what they ask, and where they meet you (Chaffey & Ellis-Chadwick, 2022). It is the model that explains why a transition rate is what it is.

DEFINITION Customer journey and touchpoint The customer journey is the sequence of experiences a person goes through from the moment a need is triggered to the moment the need is satisfied (and, ideally, the moment they recommend the solution to others). A touchpoint is any point of contact between the person and your business along that journey — a search result, an advertisement, a website page, a phone call, a proposal, an invoice, a technician at the door, a follow-up email, a review request. Some touchpoints are yours to design; others (a peer's opinion, a review site, an AI assistant's summary) are not, but they shape the journey all the same.

The journey is not linear. McKinsey's study of purchase decisions across several industries found that consumers do not steadily narrow a set of options; they start with an initial consideration set, then, during an active evaluation phase, add brands they discover and drop others — and the touchpoints that most influence them at that stage are largely consumer-driven (reviews, recommendations, in-store or online research) rather than company-driven (advertising). After purchase, the experience of using the product feeds a loyalty loop in which a satisfied customer skips the evaluation phase next time (Court, Elzinga, Mulder, & Vetvik, 2009). Two conclusions follow for a small business: you can enter a buyer's consideration late — if you are findable and credible during active evaluation — and the post-purchase experience is where the next sale is decided.

Two related ideas complete the picture. Jan Carlzon, then head of Scandinavian Airlines, called the brief interactions in which a customer forms an impression of a company its moments of truth — each one an opportunity to confirm or destroy the promise (Carlzon, 1987). Google later described a zero moment of truth: the research a buyer does online before ever contacting a seller, which now precedes most purchases of any consequence (Lecinski, 2011). Between the two, the lesson for an owner is that the journey has moments that count far more than others, and that several of them happen before you know the person exists.

2.3.2 The five stages of a small-business journey

A workable journey map for most small businesses has five stages (Figure 2.2):

1. Trigger and awareness. Something happens — a breakdown, a growth spurt, a regulation, a frustration — and the person becomes aware that they have a problem worth solving. They may not yet know that your category exists. The typical touchpoints are a peer's remark, a social post, an event or an advertisement, and the typical question is: is this normal, and who fixes it?
2. Research. The person learns about the problem and the options — searches, reads, watches, asks an AI assistant, asks peers. They form a first list. The typical question is: what does this cost, and who is credible?
3. Evaluation. They shortlist two to four providers, visit websites in detail, read reviews, request quotes or demonstrations, talk to salespeople. The typical question is: will this work for me, and what if it fails?
4. Purchase. They decide, sign, pay — and enter the fragile period between commitment and delivery. The typical question is: did I choose well, and what happens next?
5. Use and advocacy. They use the product or service, form a judgment, renew or leave, recommend or warn. The typical question is: was it worth it, and would I recommend it?

Figure 2.2 — Customer journey map template: five stages across, four rows down (actions, touchpoints, questions and emotions, opportunities). Fill one per persona (Chapter 4). Original figure.

2.3.3 Mapping the journey in practice

A journey map is only useful if it is specific. "Customers research online" is not a finding; "owners of 10-to-50-employee firms search for our category on their phone in the evening, read one comparison article and two websites, then ask their accountant" is. The evidence comes from three sources, in order of value: conversations with recent customers (five to ten, twenty minutes each, asking what was happening when you started looking? where did you look? what almost stopped you?), data you already have (how leads arrived, what pages they saw, how long they took) and observation of public places where your buyers talk (reviews of competitors, forums, association groups). Chapter 4 gives the interview method; Chapter 3 decides which customers to interview.

The map's fourth row — opportunities — is where marketing work comes from. Each cell answers one question: what could we do at this stage to make the next step easier, faster or more certain? Those answers become content (Book B5), pages (Book C2), sequences (Book B8), proof (Chapter 5), and operational fixes (Book J2). A map with a full opportunities row is, in effect, a marketing plan waiting to be prioritized (→ see Book B4).

WHO DOES THIS Owning the journey map Solo: the owner interviews five recent customers over two weeks and fills the map alone in an afternoon; the map lives in a shared document and is revisited each quarter. Small team: whoever owns marketing runs the interviews with one salesperson or service lead present, because the friction is usually in delivery; the owner validates the opportunities row. Established: the marketing manager owns one map per persona, updates it twice a year from interview notes and CRM data, and reviews the opportunities row with sales and operations — the journey map is the standing agenda of the alignment meeting (Chapter 9). In every case, the map is built from evidence, never from a brainstorm.

2.4 Owned, Earned and Paid Media

2.4.1 Three classes of channel

Channels differ in who controls them, what they cost and how fast they work. The most useful classification — proposed by Forrester Research analysts in 2009 and now standard — divides them into three classes: owned media, which you control (your website, your email list, your content, your customer data); earned media, which others give you (search rankings, reviews, mentions, referrals, shares); and paid media, which you rent (advertising in any form, sponsored content, paid placements) (Corcoran, 2009). Table 2.1 sets the three classes side by side: for each, examples, the degree of control you hold, the cost structure, the speed and durability of what it produces stated together in one column, and what it is for. Read it as the argument for building owned media first.

Table 2.1 — Owned, earned and paid media: what each class is for

Three rules follow from the table.

Build owned first. Paid channels without an owned destination waste money: a click that lands on a page that does not convert is a click bought twice, because you will have to buy it again. An email list you own can be reached a hundred times for the cost of one advertisement; a social following can be taken away by a policy change. The order of investment for a small business is almost always owned → earned → paid (→ see Book C1, Book C2).

Earn trust deliberately. Earned media looks free and is not: it is the return on work done to be worth recommending, and on the discipline of asking. Reviews are requested at the right moment; referrals are asked for and made easy; content is written to be worth sharing; the website is structured so search engines and AI assistants can find and cite it (→ see Book C3, Book C5, Book D1).

Rent paid media to accelerate, not to substitute. Advertising is the right tool to reach a segment you cannot yet reach organically, to test a message quickly, or to fill a pipeline during a known gap — always landing on owned assets that capture consent so the reach becomes an asset (→ see Book D7).

QUÉBEC & CANADA NOTE Owned media and the law of consent (as of September 2026) The most valuable owned channel — your email and SMS list — is regulated by CASL. A commercial electronic message may be sent only with express consent (the person actively agreed, after being told the purpose and who is asking — CASL s. 10(1)) or implied consent, which exists notably in an existing business relationship: a purchase or lease within the two years before the message, or an inquiry or application within the six months before it (CASL s. 10(9)–(10)). Implied consent therefore expires; express consent does not, until withdrawn. The practical consequence: record, for every contact, the type of consent, its date and its source, and let your system stop sending when an implied consent lapses. Chapter 10 details the classes; Book H3 covers the full regime; in La Machine, the consent register does the recording (business/marketing/consent).

2.4.2 Direct and indirect routes

A second axis cuts across the three classes: whether you reach the buyer directly or through an intermediary — a distributor, a reseller, a marketplace, a referral partner, an affiliate, a professional who recommends you. Indirect routes multiply reach and add a layer of trust (the partner's), at the price of margin and of control over the customer relationship. Chapter 8 places partners among the channels; Collection F (Distribution Networks and Partner Channels) builds the programs. For this chapter, note that a partner is also a segment to be marketed to (Chapter 3): they need awareness, trust and reasons to recommend you, exactly like a customer.

2.5 Inputs and Outputs: Data, Content, Offers, Channels, Conversations

Figure 1.1 showed the marketing system as inputs, an engine and outputs. This section opens the inputs box, because most small-business marketing failures are input failures dressed up as channel failures.

2.5.1 Data

The first input is knowledge of the people you serve, in a form the system can use: a contact record for each person or company, with the fields that let you segment (industry, size, role, location, product owned, stage, source), the consent status that lets you contact them legally, and the history of what they did and what you sent. In a solo business this can be 300 records; in an established one, 30,000. The rule is the same: one source of truth, kept current, from which every list and segment is derived — never a dozen spreadsheets (Chapter 13). Data quality is a marketing task, not an administrative one: a segment built on wrong fields sends the wrong message to the wrong people, and a list without consent status cannot be used at all.

2.5.2 Content

The second input is what you have to say — and, crucially, at which stage of the journey. Content for the trigger stage names the problem in the buyer's words; for research, it explains options and costs honestly; for evaluation, it proves (cases, comparisons, reviews, guarantees, credentials); for purchase, it reassures and onboards; for use, it educates and asks. A business with plenty of awareness content and no evaluation content will attract visitors who leave at the exact moment they were ready to choose. Book B5 (Content Marketing and the Content Factory) organizes production; here the point is that the inventory of content, mapped to stages, is an input you should audit before buying any reach.

2.5.3 Offers

The third input is what the person can do at each stage — the exchange proposed to them. An offer is not only the product for sale; it is also the guide in exchange for an email address (a lead magnet), the free assessment in exchange for a conversation, the trial, the demonstration, the quote, the package, the promotion, the renewal. A journey with an offer at every stage converts; a journey whose only offer is "buy now" loses everyone who is not ready today (→ see Book B9, Lead Generation; Book A6, Designing a Value-Added Offer; Book A7, Pricing Strategy and Revenue Models).

2.5.4 Channels

The fourth input is the set of channels you have chosen (Chapter 8) — chosen, not accumulated. Each channel in the system should have a role (which stage, which segment, which offer), a budget of money or hours, and a measure. A channel with none of the three is noise.

2.5.5 Conversations

The fifth input is the least automated and the most decisive: the human conversations — sales calls, service interactions, onboarding sessions, interviews — in which the business learns what the market actually thinks, and in which trust is won or lost. Conversations feed the data (what the customer told you), the content (the questions you keep hearing), the offers (what people ask for that you do not sell) and the journey map. A marketing system that has no channel for what salespeople and technicians hear is deaf; Chapter 9 installs that channel.

2.5.6 Outputs, again

The outputs — awareness, demand, trust, retention, learning (Chapter 1, Section 1.4) — are produced by these five inputs passing through the engine. When an output is short, the cause is almost always an input: demand is short because the offer is weak or the content stops at awareness; trust is short because proof is missing; retention is short because the post-purchase journey has no touchpoints. Diagnosing by input, not by channel, is the discipline of Section 2.7.

2.6 Feedback and Learning Loops

2.6.1 The weekly count

A learning loop starts with a habit: once a week, in the slot booked in Chapter 1, count the stages and compute the transitions for the period. The count takes fifteen minutes if the data is in one place and is impossible if it is not — which is a reason to fix the data before anything else. Track five numbers at first: visitors (or reach), leads, opportunities (meetings or quotes), customers and revenue; add source (where each lead came from) as soon as you can. Chapter 11 puts this in the task table.

2.6.2 Record the source of every lead

THE RULE Every lead carries its source This is a rule of practice, not of law, but it is non-negotiable in a functioning system: no lead enters the pipeline without a recorded source — the channel and, where possible, the specific campaign, page, event or referrer that produced it. Without source, you cannot compute cost per lead or cost per customer by channel, and every budget decision becomes an argument. The minimum: a source field on the contact record, filled at creation (automatically for web forms and campaign links; by the person who takes the call for phone and walk-in leads, by asking "how did you hear about us?"). Attribution can get sophisticated later (→ see Book B7); the field cannot wait.

2.6.3 Test and learn

The second habit is to change one thing at a time and judge it by the ratio it was meant to move. A test has four parts: a hypothesis ("stating the price range on the service page will raise the visitor-to-lead rate"), a metric (visitor-to-lead rate on that page), a sample large enough to mean something (at a 3% baseline, a few hundred visitors per variant at minimum — small businesses often need to run a test for several weeks), and a stop rule (a date or a sample size at which you decide). Most small-business tests will not reach statistical rigor; they can still reach practical rigor — a change that moves a ratio from 3% to 5% over six weeks and four hundred visitors is worth keeping, whatever a statistician says about confidence intervals. The point of the loop is not certainty; it is that decisions are made on counted results rather than on the last opinion heard.

2.6.4 The monthly review and the quarterly reset

Weekly counts feed a monthly review (one hour: what moved, what we tried, what we learned, what we stop and start) and a quarterly reset (half a day: are the segments still right? has the journey changed? which channel earns more budget, which loses it?). Book B4 supplies the plan and KPI structure these meetings use. What matters here is that the loop closes: a system whose measurements never change a decision is a dashboard, not a loop.

2.7 Diagnosing Your System: Where It Leaks

2.7.1 Symptoms, causes and chapters

Diagnosis begins with the funnel counts (Section 2.1) and continues with the journey (Section 2.3). Table 2.2 lists the most common symptoms, their usual causes and where in this book the cure lives.

Table 2.2 — Symptoms, likely causes and where to look

WATCH OUT Optimizing the top while the bottom leaks The most common small-business reflex when results are poor is to add traffic: more posts, more ads, a new channel. If the leak is downstream — a 4% lead-to-meeting rate because nobody calls back before Tuesday — more traffic produces more leads that nobody calls back before Tuesday. The money buys a bigger version of the same problem. Diagnose from the bottom up: is the close rate healthy? the meeting rate? the lead rate? Fix the deepest leak first; every dollar spent above it is multiplied by every ratio below it.

2.7.2 The sixty-minute system diagnostic

METHOD The 60-minute system diagnostic When to use it: before spending on any new channel or tool; at each quarterly reset; when results fall. Deliverable: one page — your funnel with counts and ratios, your top three leaks, and the chapter or book that addresses each. Minutes 0–15 — Count. For the last full quarter, write the number at each stage: reach or visitors, engaged, leads, meetings or quotes, customers, revenue; and, if you can, repeat customers and referrals. Write "unknown" where you have no number — each "unknown" is itself a finding (Section 2.6). Minutes 15–25 — Compute. Divide each stage by the one before to get the transition rates. Note the time from lead to customer for your last five customers. Minutes 25–40 — Locate the leak. Compare your ratios with what you would expect for your kind of business (your own history is the best benchmark; otherwise ask peers in your association or sector). Mark the deepest stage where the ratio is clearly weak — that is your first leak. Mark the next two. Minutes 40–55 — Explain with the journey. For each leak, ask what the buyer experiences at that stage: what question is unanswered, what offer is missing, how long they wait, who owns the next step. Use Figure 2.2's rows. If you do not know, write the name of the customer you will call to find out. Minutes 55–60 — Decide. For each leak, name one change, one owner, one metric and one date. Put the three changes in your task list (Chapter 11); do nothing else about marketing until they are done. Pitfall: diagnosing from opinions ("our website is old") instead of counts. The website may be old and irrelevant; the counts will tell you.

2.7.3 What a healthy system looks like

A healthy small-business marketing system is not one with impressive numbers; it is one where every stage has a count, every count has an owner, every transition is known, the weakest transition is being worked on, and last quarter's decision can be checked against this quarter's ratio. It draws on owned assets that grow each month (list, content, reviews), earns trust it can point to, rents reach only to accelerate, and captures what customers say so the next quarter's inputs are better than this one's. Everything else in this book — segments, personas, positioning, mix, brand, channels, alignment, compliance — is a component that makes one of those transitions higher or one of those inputs better.

Costly Mistakes

1. Treating the funnel as a picture instead of a measurement device. Drawing it once for a presentation and never counting produces a diagram, not a system. Consequence: decisions keep being made on opinion; the leak stays where it is, at full cost, for years.
2. Buying traffic to fix a downstream leak. Adding advertising when the lead-to-meeting rate is the problem multiplies the problem. Consequence: a business can spend $2,000–$5,000 CAD a month producing leads that expire in an inbox — and conclude that "advertising does not work".
3. Ending the model at the sale. Ignoring repeat business, referrals and reviews as marketing outputs leaves the cheapest growth untouched. Consequence: the funnel must be refilled from zero every period; the flywheel never turns.
4. Mapping the journey from the meeting room. A journey map built from what the team assumes, rather than from conversations with recent customers, encodes the company's blind spots. Consequence: content answers the wrong questions and the ratio it was supposed to move does not move.
5. Renting all your reach. Building a following on platforms you do not control, without converting it into consented contacts you own, exposes the whole system to a policy change. Consequence: businesses have lost the majority of their reach overnight when a platform changed its algorithm or its rules; the ones with an email list did not.
6. No source on the lead. Leads entering the pipeline without a recorded source make cost per customer by channel unknowable. Consequence: budget follows the loudest channel, not the best one; the learning loop never closes.

Key Takeaways

• The funnel is a measurement device: stages defined by observable events, counted, with transition rates and time between stages. Measure the transitions, not the stages.
• Small improvements compound. Improving four transition rates by one fifth each roughly doubles the result from the same traffic — usually at a fraction of the cost of buying more traffic.
• The flywheel completes the funnel: existing customers are an input (repeat, referrals, reviews, shared content) and friction anywhere in delivery is a marketing problem.
• The journey explains the ratios: five stages, each with actions, touchpoints, questions and opportunities; buyers research alone and may add you late, so findability and proof during active evaluation matter, and the post-purchase experience decides the next sale.
• Owned, earned and paid media have different jobs: owned is the foundation you build first; earned is the trust you earn deliberately; paid is the acceleration you rent, landing on owned assets that capture consent.
• Inputs, not channels, cause most failures: data (one source of truth, with consent), content by stage, offers at every stage, chosen channels, and the conversations that feed all four.
• Run a loop, not a dashboard: count weekly, put a source on every lead, run one test at a time and judge it by its ratio, review monthly, reset quarterly.
• Diagnose from the bottom up with the sixty-minute method, and fix the deepest leak first.

Your Action Plan

The action plan below presents the six steps that turn this chapter into work, in the order in which each step consumes the output of the one before. For each step, the table gives the deliverable that must exist at the end, the timing on a small business's calendar, and the verification criterion by which anyone — a co-founder, a mentor, the owner's own future self — can tell that the step was done. The criterion column is the one to read first: a step whose criterion you cannot picture is a step you do not yet understand.

In La Machine

The marketing system of this chapter maps directly onto the Marketing module.

• Funnels (business/marketing/funnels, editor at business/marketing/funnels/:id/edit) group an audience with the sequences and campaigns that work it; as read on September 15, 2026, a funnel is not a pipeline and keeps no stage counts, so the counts of Figure 2.1 are read from responses, tags, segments, the log and sales records, and the funnel's description is the place to name where each stage is counted (→ see Book N5). Use those sources for the weekly count of Section 2.6.1.
• Contacts, lists and segments (business/marketing/contacts, business/marketing/lists, business/marketing/segments with live preview) are the data input of Section 2.5.1: one contact base, from which lists and dynamic segments are derived. Each contact carries five opt-in flags, one per channel, which the product raises only where a documented consent supports them; the consent itself, with its type, its date and its proof, lives on the consent page (business/marketing/consent) — the register the Québec & Canada Note of Section 2.4 asks you to keep — and each contact carries its source, the field the rule of Section 2.6.2 requires (→ see Book N4).
• Campaigns and sequences (business/marketing/campaigns, business/marketing/sequences) are the offers and content by stage of Section 2.5: a sequence attached to a stage of the funnel moves people to the next one (the follow-up that recovers the unfollowed quote, the onboarding that starts the flywheel). Templates and CTAs (business/marketing/templates, business/marketing/ctas) hold the messages and the calls to action.
• The log (business/marketing/log) shows what left, and the email queue page (business/marketing/email-queue) is a manual fallback that the sending path does not use — the learning loop in raw form starts from the log. Forms, landing pages and lead magnets (business/marketing/forms, business/marketing/landing-pages, business/marketing/lead-magnets, each with a /responses page) are the owned-media capture points of Section 2.4, and every response arrives with its source.
• CRM intake and tasks (crm/intake, crm/tasks) serve the hand-off of Chapter 9 differently from what their names suggest: crm/intake is a three-step client profile form — identity, consent and subscription — rather than a queue of incoming leads, and crm/tasks is no longer a page of its own, since it redirects to Productivity → Tasks (productivity/tasks), where the activities attached to contacts and companies are read and edited (→ see Book N4).

What La Machine does not do here: it does not run the customer interviews or fill the journey map for you (the map is a document you build from evidence; store it with the book's checklist and revisit it quarterly); it does not decide your stage definitions — you write them, and the funnel records them; and it reports the counts and transitions of your funnel, not benchmarks for your industry, which you obtain from your own history or from peers. Book N4 (CRM, Contacts and Prospect Data) and Book N5 (Marketing Automation in La Machine) walk through each page.

Self-Check

1. What is the difference between a stage count and a transition rate, and which one tells you where the system leaks?
2. If each of four transition rates improves by 20%, by roughly what factor does the number of customers from the same traffic increase?
3. Name the two things the flywheel model says that the funnel model cannot.
4. Give one example each of owned, earned and paid media, and state the rule about the order in which a small business should invest in them.
5. Under CASL, for how long does implied consent from an existing business relationship based on a purchase last, and from an inquiry?

Answers. (1) A stage count is the number of people at a stage; a transition rate is the share who moved from one stage to the next. The transition rates (and the time between stages) locate the leaks; counts alone do not (Section 2.1.2). (2) By about 2.07 — because 1.2 × 1.2 × 1.2 × 1.2 ≈ 2.07 (Section 2.1.2, Do the Math). (3) That customers are an input to marketing (repeat, referrals, reviews), not only its output; and that friction anywhere in delivery is a marketing problem because it slows the whole system (Section 2.2). (4) Owned: your website or email list; earned: a customer review or a search ranking; paid: a search advertisement. Invest owned → earned → paid: build the destination, earn trust, then rent reach that lands on owned assets (Section 2.4.1). (5) Two years from the purchase or lease; six months from the inquiry or application (CASL s. 10(10)) — after which implied consent lapses and sending must stop unless express consent was obtained (Section 2.4.1).

Chapter References

An Act to promote the efficiency and adaptability of the Canadian economy by regulating certain activities that discourage reliance on electronic means of carrying out commercial activities, S.C. 2010, c. 23 (Canada's Anti-Spam Legislation), s. 10. Justice Laws Website. Retrieved September 7, 2026, from https://laws-lois.justice.gc.ca/eng/acts/E-1.6/section-10.html

Carlzon, J. (1987). Moments of truth. Ballinger Publishing.

Chaffey, D., & Ellis-Chadwick, F. (2022). Digital marketing (8th ed.). Pearson.

Corcoran, S. (2009, December 16). Defining earned, owned and paid media. Forrester Research.

Court, D., Elzinga, D., Mulder, S., & Vetvik, O. J. (2009, June). The consumer decision journey. McKinsey Quarterly. McKinsey & Company.

Halligan, B. (2018, November 20). Replacing the sales funnel with the sales flywheel. Harvard Business Review. https://hbr.org/2018/11/replacing-the-sales-funnel-with-the-sales-flywheel

Kotler, P., & Keller, K. L. (2016). Marketing management (15th ed.). Pearson.

Lecinski, J. (2011). ZMOT: Winning the zero moment of truth. Google.

Strong, E. K. (1925). The psychology of selling and advertising. McGraw-Hill.

Chapter 3 — Markets, Segments and the Ideal Customer Profile

Ask an owner who their customers are and the honest answer is often "anyone who calls". That answer is why the marketing budget produces so little: a message for everyone is a message for no one, and a business that serves everyone serves each of them worse than a competitor who chose.

What You'll Be Able to Do

• Define your market and divide it into segments using criteria that predict how people buy — not merely how they look.
• Tell a B2B decision unit from a B2C one, and know what each implies for whom you must convince.
• Write a one-page ideal customer profile (ICP) with four parts: fit, value, reachability and exclusions.
• Score and rank segments on attractiveness and on your ability to win, and decide which one comes first.
• Set your territory and reach constraints — geographic, legal, linguistic, contractual — and write your do-not-prospect rules down.
• Keep the ICP alive with a quarterly review that uses your own win, loss and margin data.

Why It Matters

Every later decision in marketing — the message, the channels, the offers, the budget — depends on whom you have chosen to serve. Choose well and the rest gets easier: the message writes itself because you know the reader; the channels are obvious because you know where those people gather; the sales cycle shortens because the prospects fit. Choose badly, or not at all, and every downstream activity is diluted across people who will never buy, or who will buy once and cost more to serve than they pay. For the owner, the ICP is the most valuable page in the company; for the team, it is the document that lets them say no with confidence.

In This Chapter

• 3.1 Markets and Segmentation Criteria
• 3.2 B2B and B2C Decision Units
• 3.3 The Ideal Customer Profile
• 3.4 Scoring and Prioritizing Segments
• 3.5 Territories and Reach Constraints
• 3.6 Writing the ICP Document
• 3.7 Keeping the ICP Alive

3.1 Markets and Segmentation Criteria

3.1.1 What a market is

A market, in marketing, is not a place; it is a set of actual and potential buyers of an offer (Kotler & Keller, 2016). Defining your market is the first act of strategy, because it decides who your competitors are, what "share" means and where growth can come from. Derek Abell showed that a business is best defined along three dimensions — which customer groups it serves, which customer needs it satisfies and which technologies or methods it uses to satisfy them (Abell, 1980). A bookkeeping firm that defines itself as "bookkeeping for anyone" competes with every accountant and every piece of software; the same firm defined as "monthly bookkeeping and cash-flow reporting for owner-operated trades businesses with 5 to 30 employees" has a market it can count, reach and win.

DEFINITION Market, segment and segmentation A market is the set of people or organizations that have, or could have, a need your offer satisfies, the ability to pay for it and the possibility of being reached. A segment is a subset of that market whose members share characteristics that make them respond similarly to a marketing approach — the same needs, the same buying process, the same channels, the same objections. Segmentation is the act of dividing the market into such subsets so that a business can choose which to serve and how. Wendell Smith introduced the concept in 1956 as an alternative to treating a heterogeneous market as one (Smith, 1956); it remains the foundation of every targeting decision.

Three nested numbers are useful when sizing a market, even approximately. The total addressable market is everyone who could conceivably buy the category. The serviceable available market is the part you can actually serve given your offer, geography and capacity. The obtainable market is the share of that you can realistically win in a given period, given competition and your resources. Small businesses almost never need precise figures for the first two; they need an honest estimate of the third, because that is what the funnel of Chapter 2 must produce (Section 3.1.4).

3.1.2 Criteria: what to segment by

Segmentation criteria fall into five families. The first three describe who the buyer is; the last two describe how and why they buy — and those are the ones that predict behavior.

1. Geographic criteria are country, province, region, city, urban or rural, distance from you and climate. They are essential for any business that delivers physically or sells in person, and relevant for language and law (Section 3.5).
2. Demographic criteria (consumers) are age, household composition, income, occupation, education, life stage and home ownership. Firmographic criteria (organizations) are the industry (usually coded by the North American Industry Classification System, NAICS), the number of employees, revenue, the age of the business, its legal form, its ownership (independent, franchise, subsidiary) and the number of locations.
3. Technographic and operational criteria are the tools, systems and methods the buyer already uses (which accounting software, which equipment, which platform) and, for organizations, their operating model (manufacturer, distributor, service provider), their capabilities and whether they buy centrally or locally.
4. Behavioural criteria are purchase history, frequency, volume, loyalty, channel preference, how they search, whether they have bought your category before, and what triggered the last purchase (a breakdown, an audit, a hire, a move, a regulation).
5. Needs-based and attitudinal criteria are the job the buyer is trying to get done (Chapter 4), the outcome they value most (speed, price, risk reduction, prestige, simplicity), their attitude to risk and to change, and their level of expertise.

For industrial and business markets, Bonoma and Shapiro proposed a nested approach: start with the outer, easily observed layers — demographics of the firm, its operating variables — and work inward through purchasing approaches, situational factors and finally the personal characteristics of the buyers, stopping when the segment is precise enough to act on (Bonoma & Shapiro, 1983). The nesting is useful because the outer layers are cheap to obtain (a business register gives you industry and size) while the inner layers are what actually differentiate buyers — and you learn them only by talking to customers.

KEY CONCEPT Segment by what predicts behavior, not by what is easy to see Industry and size are visible; they are where most small businesses stop. But two 30-employee manufacturers in the same city can behave completely differently as buyers if one has just failed an audit and the other has not, if one buys centrally and the other lets each site decide, if one values speed and the other values price. The observable criteria (geography, firmographics) tell you where to look; the behavioral and needs-based criteria tell you who will buy and what to say. A good segment combines both: an observable outer shell you can build a list from, and an inner definition — need, trigger, attitude — that you confirm in the first conversation and that the message speaks to.

3.1.3 What makes a segment usable

Not every subset of a market is a segment worth pursuing. Kotler and Keller list five requirements: a segment must be measurable (you can estimate its size and purchasing power), substantial (large and profitable enough to serve), accessible (you can reach and serve it), differentiable (it responds differently from other segments to different marketing) and actionable (you can design a program to attract and serve it) (Kotler & Keller, 2016). A sixth requirement matters especially for small businesses: the segment must be winnable — you have, or can build, a reason to be chosen over the alternatives already serving it (Section 3.4).

Figure 3.1 draws the movement those six requirements produce, from the whole market at the top to the ideal customer profile at the bottom. It reads downward in three levels: the market as a whole, the segments carved out of it by the variables of Section 3.2, and — inside the segment chosen — the profile of the customer the business is designed to serve well. The point of the drawing is that each level is narrower and better described than the one above it, and that the profile at the bottom is a description of a customer rather than a description of a market.

Figure 3.1 — Segmentation tree. The market splits into segments by criteria that predict behaviour; one segment becomes the ideal customer profile; others are written down as exclusions. Original figure.

3.1.4 Sizing a segment with public data

Sizing does not require a research budget. For business markets, Statistics Canada publishes Canadian Business Counts, a semi-annual count of employer businesses by NAICS industry, employment-size range and geography (Canada and provinces; also census metropolitan areas and subdivisions); the December 2024 release is Table 33-10-0764-01 (Statistics Canada, 2025). For consumer markets, census profiles give population, households, age and income by geography. Trade associations publish member counts; the Registraire des entreprises du Québec offers a public register of every enterprise doing business in Québec (Registraire des entreprises du Québec, n.d.). These sources give the outer shell of a segment; your own customer data gives the rates.

DO THE MATH Sizing an obtainable market Illustrative figures. A firm sells a managed IT service to professional firms (law, accounting, engineering, architecture) with 10 to 99 employees in one province. From the business counts it finds 3,400 such firms in the province. It cannot serve them all: about 60% are within reasonable travel distance or already served remotely, so the serviceable market is ≈ 2,040 firms. In its experience, one firm in three that it reaches with a fitting message takes a meeting, and one meeting in four becomes a customer — but it can only reach about 500 firms a year with its channels and budget. Obtainable in year one: 500 × 1/3 × 1/4 ≈ 42 new customers. At $14,000 CAD of annual value each, that is ≈ $588,000 of new annual revenue — a number the owner can plan capacity, hiring and budget around. Over three years, if reach and rates hold, the firm would serve about 125 of 2,040 firms, or 6% of the serviceable market: substantial for the firm, small enough that competition will not react violently. If the same arithmetic had produced 8 customers, the segment would fail the substantial test and the firm should widen it or move on.

QUÉBEC & CANADA NOTE Where the numbers come from (as of September 2026) Three public sources cover most small-business sizing needs. Statistics Canada — Canadian Business Counts (Table 33-10-0764-01, December 2024, and its June and December updates) counts employer businesses by NAICS code, employment-size range and geography, free of charge (Statistics Canada, 2025). The Registraire des entreprises du Québec (REQ) lets anyone consult the public register of enterprises — legal name, form, address, sector — one enterprise at a time, and offers bulk data for those who need lists (Registraire des entreprises du Québec, n.d.). Statistics Canada's Census Profile gives population, household, income and language data down to the neighborhood. Note that counting is one thing and contacting another: a firm found in a register may be reached by mail or by telephone subject to the CRTC's Unsolicited Telecommunications Rules and the National Do Not Call List, and by email only with consent under CASL (→ Chapter 10; → see Book H3).

3.2 B2B and B2C Decision Units

3.2.1 Who decides

A segment is a set of buyers; a decision unit is the set of people, inside one buyer, who take part in a purchase. In consumer markets the unit is an individual or a household; in business markets it is a group. Webster and Wind's model of organizational buying identified the roles that recur in that group: users, who will work with the purchase; influencers, who shape the criteria (often technical or financial staff); buyers, who hold formal authority to select and negotiate; deciders, who make the final choice, formally or informally; and gatekeepers, who control the flow of information to the others (Webster & Wind, 1972). In a 15-person company, several roles sit in one person — the owner; in a 200-person company they may be five people who disagree.

Understanding the decision unit changes three things in your marketing: whom the message addresses (the decider's outcome, the user's daily life, the influencer's criteria — each needs a version), what proof is required (a decider wants risk reduction and return; a user wants ease; a buyer wants terms) and how long the cycle takes (more people, more time). Chapter 4 builds the personas for each role; here, the point is that a B2B segment must be defined at two levels — the account (the firm: industry, size, situation) and the people in the decision unit — while a B2C segment is usually defined at one.

3.2.2 What differs, and what does not

Table 3.1 compares business and consumer markets from the marketer's point of view. Each row is a dimension of the purchase — who decides, how many buyers there are, the value per customer, the length of the cycle, the part played by rationality, the nature of the relationship, and the regulation of contact — with a column for business-to-business markets, a column for business-to-consumer markets, and a last column that says what the difference implies for the ideal customer profile. The point to retain is that the mechanics differ more than the psychology: in both markets a person decides, but in a business market several people take part, the cycle is longer and the value per customer is higher, so the profile must name the account and the roles.

Table 3.1 — Business and consumer markets compared, from the marketer's point of view

What does not differ is the underlying method: you still segment by criteria that predict behavior, you still choose, and you still define whom you will not pursue. Many small businesses serve both kinds of buyer — a landscaper with residential and commercial customers, a software firm with individuals and companies. The choice is not to abandon one but to write two profiles, name a primary one that gets the strategic attention and the first budget, and treat the other as secondary or opportunistic (Section 3.4).

3.3 The Ideal Customer Profile

3.3.1 Definition and purpose

DEFINITION Ideal customer profile (ICP) The ideal customer profile is a written description of the type of customer — organization or person — that your business can serve best, win most easily and profitably, and reach at a reasonable cost, together with the types you will not pursue. It has four parts: fit (who they are, in observable and situational terms), value (what they are worth to you and what they cost to serve), reachability (how you can find and contact them) and exclusions (whom you refuse and why). It is not a persona: a persona (Chapter 4) describes a person and their behavior; the ICP describes the customer — the account or household — that the business chooses to serve. In B2B the ICP is the account definition inside which personas live.

The ICP serves four purposes. It focuses the message and the channels. It qualifies: any lead can be checked against it in two minutes, so that sales time goes to prospects who fit. It protects margin, because the customers who cost the most to serve are almost always the ones outside the profile. And it aligns marketing and sales (Chapter 9): both functions agree in advance on what a good lead looks like, instead of arguing about it lead by lead.

KEY CONCEPT An ICP is also a list of who you will NOT pursue The most useful half of an ICP is the half that says no. Every business has customers it should not have taken: too small to pay for the attention they need, too large to serve without capabilities you lack, in a sector whose rules you do not know, in a geography you cannot reach profitably, or with a buying behavior — endless quotes, late payment, scope creep — that destroys margin. Writing those exclusions down converts them from painful memories into policy. It also frees the team: a salesperson who can say "we do not serve that segment, here is who does" spends the saved hours on prospects who fit, and the referred prospect remembers the honesty.

3.3.2 Fit: who they are

Fit combines the observable shell and the inner situation. For a business customer: industry (NAICS codes or plain names), size (employees, revenue, locations), geography, ownership and structure, technologies in use, and — decisively — situation: the triggers and conditions that make them need you now (a growth spurt, a compliance deadline, a departure, an aging system, a new location). For a consumer: life stage, household, location, income band, and again the situation (a move, a birth, a renovation, a retirement, a diagnosis). Fit should be written so that a stranger could sort a list of prospects with it.

3.3.3 Value: what they are worth

Value has four components, each of which your accounting can approximate: revenue (first purchase and annual), margin (after direct costs — some segments pay well and cost more), lifetime (how long they stay, how often they buy again, whether they expand), and cost to serve (support, travel, customization, payment delays). The best customers are rarely the largest; they are the ones with the best margin over lifetime relative to cost to serve. A useful habit is to compute, once a year, revenue and margin per customer and to rank customers by it — the top fifth usually shares traits that belong in the ICP, and the bottom fifth shares traits that belong in the exclusions.

3.3.4 Reachability: how you find them

A perfect segment you cannot reach is a fantasy. Reachability asks: can we list them (a register, an association directory, a purchased list, a partner's customers)? where do they gather (events, publications, online groups, the professionals they all use)? which channels reach them at acceptable cost (Chapter 8)? who could introduce us (partners, referrers, suppliers)? and — in Canada — how can we lawfully contact them (consent path: inquiry, existing relationship, express opt-in, permitted mail or telephone contact)? A segment that scores high on fit and value but low on reachability may still be right, if you are prepared to invest in building the reach (content, partnerships, a presence at their events) over a longer horizon.

3.3.5 Exclusions: whom you refuse

Exclusions come in four kinds: structural (sizes, sectors, geographies you cannot serve well), regulatory (sectors requiring licenses or clearances you do not hold), economic (customers whose expected margin over lifetime is below the cost to acquire and serve them) and behavioral (patterns that predict a bad relationship — a history of switching suppliers yearly, a procurement process built on the lowest price, a refusal to give access to the people who will use the product). Exclusions must be reasons, not prejudices: they are written because the data or the experience supports them, and they are reviewed (Section 3.7).

CASE IN POINT A consulting firm writes its ICP Fictional example; names and figures are invented. Meridian Advisory, eight consultants, helps companies design their operations and adopt software. For five years it took every engagement. When the partners ranked their 61 clients by margin over lifetime, the top twelve had four things in common: 10 to 100 employees, in four sectors (specialized manufacturing, distribution, construction trades, professional services), owner-led, and engaged after a trigger — a new system, a failed audit, a growth spurt — rather than "to see what we could do". The bottom twelve shared two traits: solo practitioners and micro-firms under 10 employees, whose engagements were small and whose support demands were not; and a public-sector agency and a hospital foundation, whose procurement rules stretched a six-week project to nine months. Meridian's ICP became one page: fit — owner-led firms of 10–100 employees in the four sectors, in the province, with an active trigger; value — engagements of $18,000 to $60,000 CAD, 70% returning within two years, cost to serve moderate; reachability — two sector associations, the firms' accountants as referrers, search for the trigger terms, a quarterly workshop; exclusions — under 10 employees (referred to a partner who serves them), public-sector procurement, and any prospect who wants a fixed price before a diagnostic. In the following year, revenue per consultant rose 23% with fewer projects; the partners' explanation was not that they worked harder but that they stopped saying yes.

WATCH OUT An ICP built only from the customers you happen to have Ranking your existing customers is the right start, but it carries a bias: it can only reveal segments you have already reached. If your best customers all came from one referral source, the ICP will describe that source's network, not your best possible market. Test the draft against the outside: the segment sizes from public data (Section 3.1.4), the deals you lost and why, the customers of competitors you admire, and a few conversations with prospects who fit the draft but have never heard of you. If a large, reachable segment fits the inner definition (need, trigger, value) but is absent from your customer list, it may be your best opportunity — not an exclusion.

3.4 Scoring and Prioritizing Segments

3.4.1 Two questions, one grid

When several segments survive the tests of Section 3.1.3, choose among them with two questions: how attractive is the segment? and how able are we to win it? The first is about the market; the second about you. Each is scored on a handful of criteria, weighted, and the two scores place the segment in a grid — the logic of the portfolio matrices used in corporate strategy since the 1970s, reduced to what a small business needs.

Attractiveness criteria (score 1 to 5 each): size of the segment in your serviceable market; growth; value per customer (margin over lifetime); intensity of the need (how painful, how urgent, how often); competitive intensity (how many capable alternatives serve it, how satisfied their customers are); and regulatory or structural barriers.

Ability-to-win criteria (1 to 5): fit of your offer with the segment's need; references and proof you already hold in it; reachability at acceptable cost; cost to serve relative to price; and your team's knowledge of the sector's vocabulary and rules.

3.4.2 A worked scoring

Table 3.2 shows the grid of Section 3.4.1 filled for a fictional firm and three candidate segments: owner-led manufacturers with 20 to 99 employees, national distributors with 10 to 49 employees, and solo practitioners. Each row is a criterion with its weight — first the four criteria of attractiveness (size in the serviceable market, value per customer, intensity of need, and competitive intensity, inverted), then the four criteria of ability to win (fit of the offer, references held, reachability at an acceptable cost, and cost to serve against price, inverted). Each column is a segment with its score from 1 to 5; two bold rows give the weighted totals, out of 40 and out of 45, and the last row states the decision.

Table 3.2 — Scoring three candidate segments (illustrative figures)

Segment B is the most attractive on paper and the least winnable today; Segment A is where the firm should concentrate; Segment C is large and reachable but does not pay. The grid does not make the decision — the owner does — but it makes the decision explainable, which is what the team and the budget need.

3.4.3 Primary, secondary, opportunistic, excluded

The output of scoring is a priority order, not a single winner. The primary segment receives the strategic work: the positioning (Chapter 5), the content, the channel investment, the first proof. A secondary segment is served when it comes and prepared for — references built, a partner found — without carrying the plan. Opportunistic segments are accepted when the margin is right and never pursued. Excluded segments are refused and, ideally, referred elsewhere. A small business can usually carry one primary and one secondary segment; three primaries is the same as none.

WHO DOES THIS Segmentation and the ICP in the three situations Solo: the owner does it alone in an afternoon, using the customer list, one public data source and the last ten lost deals; the risk is the survivorship bias of the Watch Out box, so the owner calls two prospects who fit the draft but are not customers. Small team: the owner leads, with the person who sells and the person who delivers in the room — the deliverer knows cost to serve better than anyone; a half-day workshop (Section 3.6) produces the page. Established: the marketing manager prepares the data (revenue, margin, tenure, cost to serve by customer; segment sizes; win/loss), sales and operations score the criteria, the owner or general manager decides the priority order and signs the page; an external facilitator is useful when the sales team and the operations team disagree about which customers are "good". In every situation, the owner signs the exclusions — nobody else can commit the company to saying no.

3.5 Territories and Reach Constraints

3.5.1 The constraints that shape the map

A segment lives somewhere, and where limits whom you can serve. Five kinds of territory constraint recur.

Geographic and logistical. Travel time and cost for on-site services; delivery radius and shipping economics for physical goods; time zones for services delivered live. A landscaper's market ends where the crew's travel time makes the job unprofitable; a bakery's ends where the product cannot arrive fresh.

Legal and regulatory. Licences and permits are often provincial or municipal; professional titles are regulated by provincial orders; some sectors (financial services, health, construction, transport) require registrations your firm may not hold in every jurisdiction. Selling across the border adds tax registration, product-compliance and data-protection obligations (→ see Book A4, Legal Structures, Registration, Permits and Licences; Book H2 for cross-border data).

Linguistic. In Québec, the Charter of the French language, CQLR c C-11 governs commercial publications, websites, signage and the language of service; a business that markets in Québec markets in French, whatever else it adds. Elsewhere in Canada, serving francophone or allophone communities is a segmentation choice, not an obligation — and often an under-served opportunity (→ see Book H4).

Contractual. Exclusive distribution agreements, partner territories, non-compete clauses from a former employer or from the sale of a business, franchise territories — each defines zones where you may not sell, or may sell only through someone else (→ see Collection F).

Capacity. The most common constraint of all: how many customers you can serve well at once. A segment that would bring forty customers a year is the wrong segment if you can onboard twelve. Reach should be sized to capacity, and grown with it (→ see Book L1, Growth Strategy and Scaling).

3.5.2 Do-not-prospect rules

Beyond territories, every business accumulates people and organizations it must not contact for commercial purposes: individuals who have unsubscribed or withdrawn consent; numbers registered on the National Do Not Call List, which telemarketers must check and honor under the CRTC's Unsolicited Telecommunications Rules (Canadian Radio-television and Telecommunications Commission, n.d.); customers of a partner under an agreement; accounts in dispute or litigation; competitors' staff; a former client who asked never to be contacted; employees of a customer who asked that only one person be approached. These rules are part of the segment definition — a negative list that the outbound system must check before every contact.

THE RULE Excluded zones and do-not-prospect rules must be written, not remembered A rule of practice, and in part of law: every territory limit and every do-not-prospect rule is recorded in the system that sends messages and assigns calls, never only in someone's memory. The legal part is strict — unsubscribes under CASL s. 11 must be honored without delay and no later than 10 business days, and telemarketing calls must respect the National Do Not Call List and the internal do-not-call list under the CRTC's rules — and the consequences of forgetting are penalties, complaints and lost trust (→ Chapter 10; → see Book H3). The practice part protects relationships: the partner whose customer you approached, the client in dispute whom you emailed a promotion, the person who asked to be left alone. In La Machine, the do-not-reach list and the exclusion filters exist for this purpose (see In La Machine below); the rule is that they are filled — every exclusion, the day it is known.

3.6 Writing the ICP Document

3.6.1 The canvas

The ICP fits on one page. Table 3.3 is the canvas; every cell is written in plain language a new employee could apply.

Table 3.3 — The ICP canvas

3.6.2 Rules for writing

Four rules make an ICP usable. Specific: numbers and names, not adjectives ("10–100 employees", not "small to medium"). Evidence-based: each claim traceable to data or conversations, and the evidence line filled. Testable: a stranger with the page and a prospect's details can say fit or no fit in two minutes; if a field cannot be tested, rewrite it. Short: one page; the persona documents (Chapter 4) carry the detail about people.

METHOD The ICP workshop (90 minutes) When to use it: when no written ICP exists; at the annual reset; when a new offer or region is considered. Participants: the owner; whoever sells; whoever delivers; a facilitator if there are more than three. Deliverable: the one-page canvas of Table 3.3, dated and signed. Before the workshop (2 hours of preparation): export the customer list with revenue, margin (or a proxy), start date, repeat purchases, referrals given and a rough cost-to-serve rating; pull the last ten lost deals with the reason; pull segment sizes from public data for two or three candidate segments. 0–15 min — Rank. Sort customers by margin over lifetime; mark the top fifth and the bottom fifth. 15–35 min — Find the shared traits. For each group, list what they have in common — sector, size, situation, trigger, how they bought, how they behave. Write traits as facts, not opinions. 35–50 min — Test against the outside. Compare with the lost deals (were they in the top-fifth profile? why lost?) and with the segment sizes: is the emerging profile substantial and reachable? Name any large fitting segment absent from the customer list. 50–70 min — Draft the four parts. Fit, value, reachability, exclusions — one line per field of the canvas, specific and testable. 70–85 min — Say no. Read the exclusions aloud; for each, decide whether the company is prepared to refuse and, where possible, whom to refer to. Remove any exclusion nobody will enforce. 85–90 min — Sign and schedule. Owner signs; next review date set; the page is stored where sales and marketing will read it — and the segments are created in the CRM the same day. Pitfalls: ranking by revenue instead of margin over lifetime (the biggest customers are often the least profitable); letting the loudest anecdote override the data; writing exclusions the company will not enforce.

3.7 Keeping the ICP Alive

3.7.1 The quarterly review

An ICP decays: markets move, your offer changes, you learn. Review it quarterly with four numbers, all of which the funnel of Chapter 2 already produces if the source and segment fields are filled: the fit rate of new leads (share matching the ICP — if it falls, your channels or message are attracting the wrong people); the win rate by segment (if you win more outside the profile than inside, the profile is wrong or your positioning has shifted); margin by segment (the ICP's value part, confirmed or refuted by accounting); and cost to acquire by segment (the reachability part in dollars). Add the qualitative signal: what did sales and delivery hear this quarter that the page does not say?

3.7.2 Triggers for a rewrite

Some events justify revisiting the ICP before the quarter ends: a new product or service that serves a different buyer; a new region or a new language; three consecutive losses in the primary segment for the same reason; a competitor entering or leaving; a regulatory change that alters who needs you; a change in capacity (a hire, a departure) that alters who you can serve. Version the page — date and a line on what changed — so the team can see the reasoning, and so the exclusions are not quietly forgotten.

3.7.3 From page to system

An ICP that lives only in a document is advice; an ICP encoded in the systems is policy. Encode it three ways: as segments in the contact base (the fit criteria as filters, so the fitting contacts can be listed, counted and messaged); as qualification questions in the intake process, so that every lead is checked against fit in the first conversation (Chapter 9); and as exclusion rules in the sending and calling systems, so the negative list is enforced automatically (Section 3.5.2). Chapter 11 puts the quarterly review in the task table; Chapter 13 chooses tools that can hold the fields.

Costly Mistakes

1. Serving everyone. A message for everyone converts no one; a business that accepts every customer serves each worse than a competitor who chose. Consequence: diluted marketing, longer sales cycles, margin eroded by customers who cost more than they pay.
2. Segmenting only by what is visible. Industry and size without need, trigger and buying behavior produce lists of firms that look alike and buy differently. Consequence: outbound that hits the right companies at the wrong moment with the wrong message; low reply rates blamed on the channel.
3. Ranking customers by revenue instead of margin over lifetime. The largest accounts are frequently the least profitable once cost to serve and payment delays are counted. Consequence: an ICP that describes the customers who consume the company, and a plan to find more of them.
4. Exclusions that live in memory. Territory limits, partner accounts, do-not-call rules and withdrawn consents kept in people's heads are forgotten under pressure. Consequence: CASL and telemarketing violations, a partner betrayed, a client in dispute emailed a promotion — each costing more than the sale it hoped for.
5. Three primary segments. Spreading the strategic work over three profiles delivers a third of a positioning to each. Consequence: none of the three sees the company as for them; the budget produces awareness without preference.
6. An ICP written once. Markets, offers and capacity change; a two-year-old ICP describes a company that no longer exists. Consequence: leads that fit yesterday's profile, and a team that has stopped reading the page.

Key Takeaways

• A market is a set of buyers, not a place; define it along customer groups, needs and methods, and size it in three nested numbers — total, serviceable, obtainable.
• Segment by criteria that predict behavior. Geography and firmographics tell you where to look; needs, triggers and buying behavior tell you who will buy and what to say. Usable segments are measurable, substantial, accessible, differentiable, actionable — and winnable.
• B2B buyers are decision units: users, influencers, buyers, deciders, gatekeepers; define the account and the roles. B2C buyers are individuals or households; define the person and the situation.
• The ICP has four parts — fit, value, reachability, exclusions — and the exclusions are its most useful half.
• Rank customers by margin over lifetime, then test the emerging profile against the outside: lost deals, segment sizes, prospects who fit but never heard of you.
• Score segments on attractiveness and ability to win; carry one primary and at most one secondary; refuse and refer the excluded.
• Territories and do-not-prospect rules are written into the systems that send and call — CASL unsubscribes, the National Do Not Call List, partner accounts, disputes, withdrawn consents.
• Review quarterly with fit rate, win rate, margin and cost to acquire by segment; encode the ICP as segments, qualification questions and exclusion rules.

Your Action Plan

The action plan below presents the six steps that turn this chapter into work, in the order in which each step consumes the output of the one before. For each step, the table gives the deliverable that must exist at the end, the timing on a small business's calendar, and the verification criterion by which anyone — a co-founder, a mentor, the owner's own future self — can tell that the step was done. The criterion column is the one to read first: a step whose criterion you cannot picture is a step you do not yet understand.

In La Machine

The ICP becomes operational in the Marketing module through five pages.

• Segments (business/marketing/segments, editor at business/marketing/segments/:id/edit) hold the fit criteria as filters on contact and company fields — sector, size, region, tags for situation and trigger — with a live preview of the contacts that match, so the count of Section 3.1.4 and the fit rate of Section 3.7.1 are read rather than estimated. Create one segment per profile (primary, secondary) and one per exclusion you want to monitor.
• Companies (business/marketing/companies) and contacts (business/marketing/contacts) carry the fields the segments filter on; the import page (business/marketing/contacts/import) accepts lists built from public registers or association directories; it does not ask for a consent status, so the consent of imported contacts is documented afterwards on the Consent page before any message (Chapter 10; → see Book N4).
• Business Explorer (business/marketing/business-explorer) helps find organizations matching an ICP's outer shell — sector, size, location — as a starting point for account lists; the page access rules open it to the administrator and the reseller roles, and no others, although the page's own banner names administrators, resellers and celebrities (→ see Book B2, Market Research and Competitive Intelligence; → see Book N4).
• Do-not-reach (business/marketing/do-not-reach) is the negative list of Section 3.5.2: contacts and organizations that must never receive outbound, whatever segment they fall in; the sending and dialing functions check it before every message and call. Exclusion filters on lists and segments complete it.
• Lists (business/marketing/lists) are static snapshots — a workshop's attendees, an association's members — that segments can include or exclude.

What La Machine does not do: it does not compute margin over lifetime — that comes from your accounting (→ see Book G1) and is entered as a field or a tag; it holds territories in the prospect warehouse rather than in the marketing pages as of September 2026 — a territory is a named group of municipalities with a priority and a rank, written by the warehouse, and the marketing side reads it as a territory tag and as a filter on the Contacts page rather than as an object a marketer edits; and it does not choose your primary segment. Book N4 (CRM, Contacts and Prospect Data) covers the fields and imports page by page.

Self-Check

1. What is the difference between a segment and a list?
2. Name the five requirements of a usable segment and the sixth this chapter adds for small businesses.
3. Why is the exclusions part of the ICP described as its most useful half?
4. A firm ranks its customers by annual revenue and builds its ICP from the top ten. What two errors is it at risk of?
5. Where must territory limits and do-not-prospect rules be recorded, and why?

Answers. (1) A segment is a subset of the market defined by shared characteristics that predict similar buying behavior; a list is a set of contacts, which may or may not correspond to a segment. Segments are definitions; lists are snapshots (Sections 3.1.1 and In La Machine). (2) Measurable, substantial, accessible, differentiable, actionable (Kotler & Keller, 2016) — and winnable (Section 3.1.3). (3) Because writing down whom you refuse converts painful experience into policy, protects margin from customers who cost more than they pay, and lets the team say no with confidence and refer elsewhere (Section 3.3.1). (4) Ranking by revenue rather than margin over lifetime (the largest customers are often the least profitable), and survivorship bias — describing only the customers already reached rather than testing the profile against lost deals, segment sizes and prospects who fit but never heard of the firm (Sections 3.3.3 and 3.3.5). (5) In the systems that send messages and assign calls — never only in memory — because unsubscribes and do-not-call rules are legal obligations (CASL s. 11; CRTC Unsolicited Telecommunications Rules) and because partner accounts, disputes and withdrawn consents are forgotten under pressure (Section 3.5.2).

Chapter References

An Act to promote the efficiency and adaptability of the Canadian economy by regulating certain activities that discourage reliance on electronic means of carrying out commercial activities, S.C. 2010, c. 23 (Canada's Anti-Spam Legislation), ss. 6, 10, 11. Justice Laws Website. Retrieved September 7, 2026, from https://laws-lois.justice.gc.ca/eng/acts/E-1.6/

Charter of the French language, CQLR c C-11. LégisQuébec. Retrieved September 7, 2026, from https://www.legisquebec.gouv.qc.ca/en/document/cs/C-11

Abell, D. F. (1980). Defining the business: The starting point of strategic planning. Prentice-Hall.

Bonoma, T. V., & Shapiro, B. P. (1983). Segmenting the industrial market. Lexington Books.

Canadian Radio-television and Telecommunications Commission. (n.d.). National Do Not Call List. Retrieved September 7, 2026, from https://lnnte-dncl.gc.ca/

Kotler, P., & Keller, K. L. (2016). Marketing management (15th ed.). Pearson.

Registraire des entreprises du Québec. (n.d.). Registre des entreprises. Gouvernement du Québec. Retrieved September 7, 2026, from https://www.registreentreprises.gouv.qc.ca/

Smith, W. R. (1956). Product differentiation and market segmentation as alternative marketing strategies. Journal of Marketing, 21(1), 3–8.

Statistics Canada. (2025). Canadian Business Counts, with employees, December 2024 (Table 33-10-0764-01). Retrieved September 7, 2026, from https://www150.statcan.gc.ca/t1/tbl1/en/tv.action?pid=3310076401

Webster, F. E., Jr., & Wind, Y. (1972). A general model for understanding organizational buying behavior. Journal of Marketing, 36(2), 12–19.

Chapter 4 — Personas and Buying Behavior

The ideal customer profile tells you which companies or households to serve. It does not tell you who, inside them, will read your email at 6:40 in the morning, what was happening the day they searched for you, or what they are afraid of. Personas do.

What You'll Be Able to Do

• Explain what a persona is — and what it is not — and avoid the caricature that makes most personas useless.
• Describe the job a buyer is trying to get done, the triggers that start a purchase and the forces that stop it.
• Map a decision unit — user, influencer, buyer, decider, gatekeeper, economic buyer — and write the question each role is asking.
• List a persona's objections, risks and the proof that answers each one.
• Identify where each persona looks and listens, so channel choices follow the buyer rather than fashion.
• Build two or three personas from evidence — interviews, data and public reviews — using a repeatable method.

Why It Matters

Marketing messages fail for one reason more than any other: they describe the seller. A persona forces the reverse — it puts the reader's situation, words, fears and criteria on one page, so that every message, page, offer and sales conversation can be checked against it. For the owner, personas turn "our customers" into a small number of recognizable people whose behavior can be predicted. For the team, they end the argument about what to write and where to publish: the persona already answered.

In This Chapter

• 4.1 What a Persona Is — and the Caricature Trap
• 4.2 Jobs-to-Be-Done and Triggers
• 4.3 Decision Units: Buyer, User, Influencer, Gatekeeper, Economic Buyer
• 4.4 Objections, Risks and the Proof Needed
• 4.5 Information Sources and Channels by Persona
• 4.6 The VITO Rule: Talk to the Person Who Decides
• 4.7 Building Personas from Evidence

4.1 What a Persona Is — and the Caricature Trap

4.1.1 Origin and definition

The persona comes from software design. Alan Cooper introduced it in the late 1990s as a way to stop designing "for everyone": a small number of fictional but precise archetypes, each built from research on real users, against whom every design decision could be tested (Cooper, 1999). Marketing adopted the tool because it faced the same problem — a message written for "small businesses" pleases nobody — and Adele Revella gave it its most rigorous marketing form: a buyer persona describes not who the buyer is but how they decide, and it is built from interviews with people who recently made the decision (Revella, 2015).

DEFINITION Persona A persona is an evidence-based description of a type of person involved in buying or using your offer: the role they hold, the job they are trying to get done, the triggers that start their search, the questions and objections they raise, the proof they need, the sources they consult and the way they prefer to be approached. It is written as one recognizable person — with a name and a situation — so that the team can ask, before any decision, "what would this person think?" A persona is not a demographic summary, a market segment (Chapter 3) or a character in a story; it is a compression of what real buyers told you and did, arranged so it can be used.

Three objects are often confused. The segment is a group in the market defined by criteria that predict behavior (Chapter 3). The ICP is the account or household your business chooses to serve, with its exclusions. The persona is a person — inside the ICP account, or the household itself — with a role in the decision. In a B2B business a single ICP typically contains two to four personas (the owner who decides, the manager who will use, the accountant who influences); in a B2C business, the persona and the ICP often coincide, with a second persona for the partner or family member who influences the purchase.

4.1.2 The caricature trap

Most personas in circulation are useless, and they are useless in the same way: a stock photo, a first name, an age, a list of hobbies ("Sarah, 42, likes yoga and podcasts") and a paragraph of adjectives ("busy, tech-savvy, values quality"). None of that predicts a purchase. Whether Sarah likes yoga has no bearing on how she chooses an accountant; whether she was just told by the bank that her financial statements are unacceptable has every bearing. The caricature is a demographic sketch dressed as a person; a persona is a decision described as a person.

WATCH OUT Personas invented in a meeting room The fastest way to produce a persona is to gather the team and imagine the customer. The result is a portrait of the team's assumptions, with the company's favorite features projected onto the buyer as "needs". Such personas are worse than none: they lend false authority to guesses, and they are quoted for years. The test of a real persona is simple — every line can be traced to something a buyer said or did (an interview note, a support ticket, a lost-deal review, a search query, a review of a competitor). If a line cannot be traced, it is a hypothesis; write it in italics and go and check it. Section 4.7 gives the method; ten interviews are usually enough to build the first version of one persona.

4.1.3 How many personas, and which first

A small business should carry two to four personas in total, never ten. Start with the persona who decides in your primary segment (Section 4.6), add the persona who uses or lives with the purchase if their objections are different, and add an influencer persona only if that role routinely blocks or accelerates deals (an accountant, an IT provider, a spouse). Each additional persona multiplies the content, pages and sequences you must maintain; if two candidate personas share the job, the triggers and the objections, they are one persona with two job titles.

4.2 Jobs-to-Be-Done and Triggers

4.2.1 People hire products to make progress

The most useful lens for understanding why anyone buys is the theory of jobs to be done: customers do not buy products, they "hire" them to make progress in a specific circumstance — to get a job done — and they "fire" them when something else does the job better (Christensen, Hall, Dillon, & Duncan, 2016). A job has three dimensions: functional (the practical task — get the books closed monthly), emotional (how the person wants to feel — in control, not embarrassed in front of the bank) and social (how they want to be seen — as a competent owner, a serious partner). Most small-business marketing addresses only the functional dimension, in the seller's vocabulary; the buyer decided on all three, in their own.

DEFINITION Job to be done A job to be done is the progress a person is trying to make in a particular circumstance — the outcome they want, in the situation they are in — expressed independently of any product. "Close the month in two days without a full-time bookkeeper" is a job; "buy accounting software" is one possible hire. Jobs are stable over time (the need to close the month does not change), while solutions and channels change constantly; a persona anchored on the job stays valid when the tools around it move.

The practical technique is to write the job as a sentence with three parts — when (the circumstance), I want to (the progress), so I can (the outcome that matters) — and to fill it from interviews, not imagination. "When a big customer threatens to leave over late deliveries, I want to see every job's status without walking the floor, so I can promise dates I will keep and stop losing evenings to firefighting." Every clause of that sentence is a marketing asset: the circumstance is the trigger to speak to, the progress is the promise, the outcome is the headline.

4.2.2 Triggers: why now

A buyer who has lived with a problem for three years does not wake up one morning and decide to solve it; something happens. The trigger is the event that converts a tolerated problem into an active search — and it is the single most valuable fact a persona can contain, because it tells you when to be present and what to name. Triggers come in recognizable families: a failure (a breakdown, a lost customer, a failed audit, a penalty), a change (a new hire, a departure, a new location, a new regulation, a growth spurt), a comparison (a competitor's visible advantage, a peer's remark, an advisor's recommendation), a deadline (year end, a renewal, a tender) and a threshold (the tenth late job this month, the third weekend lost).

The researchers who developed jobs theory for sales describe four forces acting on every switch: the push of the current situation (what is wrong now), the pull of the new solution (what it promises), the anxiety about the new (will it work, will it disrupt, will I look foolish) and the habit of the present (the comfort of the known, the effort of change) (Moesta & Christensen, as described in Christensen et al., 2016; Moesta, 2020). A purchase happens when push and pull together exceed anxiety and habit. Marketing can strengthen push (name the cost of the problem), strengthen pull (show the outcome, with proof), reduce anxiety (guarantees, references, a clear process) and reduce habit (make switching easy — migration, onboarding, a pilot). A persona records where each force is strongest for that person; the message and the offer then work on the right one.

KEY CONCEPT People buy progress in a circumstance, not products in general The buyer's question is never "what does this product do?" but "will this get my job done, here, now, without making things worse?" A persona that states the job, the trigger and the four forces lets you write to that question. The seller's vocabulary (features, modules, methodology) becomes secondary; the buyer's circumstance leads. This is also why demand generation (Chapter 1) works: naming a circumstance the buyer recognizes — before they are searching — is how a business enters the consideration set early.

4.3 Decision Units: Buyer, User, Influencer, Gatekeeper, Economic Buyer

4.3.1 Roles, not titles

Chapter 3 introduced the decision unit: the set of people who take part in a purchase. Personas are built per role, not per job title, because the same title plays different roles in different companies and the same person plays several roles in a small one. Two classic frameworks name the roles. Webster and Wind identified users, influencers, buyers, deciders and gatekeepers (Webster & Wind, 1972). The sales methodology of Miller and Heiman distinguished the economic buyer — the one person who can release the money and say yes when everyone else says no — from user buyers, technical buyers (who screen against specifications and can say no but not yes) and the coach, an insider who wants you to win and tells you how the decision really works (Miller & Heiman, 1985). Combining them gives the six roles of Table 4.1.

Table 4.1 — Decision-unit roles, their question, and what marketing owes each

4.3.2 Small business: many roles, few people

In a company of twelve, the owner is economic buyer, decider and often user; the office manager is gatekeeper and influencer; the accountant, outside the company, is the influencer whose opinion carries most weight on anything financial. The roles do not disappear because the people are few — they are stacked, and the marketing must address them in order: the owner's outcome first (Section 4.6), then the objections the owner will hear from the manager and the accountant. In a household, the roles are stacked too: one person searches, another decides jointly, a third — a child, a parent — uses.

Stacking has a practical consequence for the timing of messages. When the owner is also the user, the objections of both roles arrive in the same conversation, and the proof for both must be ready at once: the outcome and the return for the decider, the ease and the training for the user. When the roles are held by different people, the objections arrive in sequence — the owner's first, then, days later, "I talked to my team and they are worried about the switch" — and the follow-up material must anticipate the second conversation before it happens: a one-page note the owner can hand to the manager, a short video for the crew, an answer to the accountant's question about payment terms. A persona set that names the stacked roles lets the sequence of Chapter 9 send the right piece to the right person at the right moment, instead of leaving the owner to defend a purchase alone.

4.3.3 Map the unit for your primary ICP

For the primary segment of Chapter 3, list the roles that appear in at least half of your deals and mark, for each, whether it typically decides, influences, uses, screens or blocks, and what it most often asks. That map decides how many personas you need (Section 4.1.3) and, later, how many versions of each message (Chapter 5) and which follow-up goes to whom (Chapter 9). Sales conversations are the cheapest source for the map: every won and lost deal of the last year names the roles that mattered.

4.4 Objections, Risks and the Proof Needed

4.4.1 Objections are information

An objection is not resistance to be overcome; it is a statement of a risk the buyer perceives and, therefore, an instruction about what proof to supply. Objections cluster in six families, and each calls for a different kind of evidence:

1. Need sounds like "we are fine as we are." The buyer has not felt the push. The evidence that answers it is the cost of the problem, quantified in their terms, and a peer's account of the same situation.
2. Fit sounds like "this is not for businesses like ours." The evidence that answers it is cases and references from the same segment, language and examples from their world, and an ICP-fit statement that says who you serve and who you do not.
3. Price and value sounds like "it costs too much." It is rarely about the number; it is usually about the relationship between price and the outcome, or about the absence of a visible price. The evidence that answers it is a price stated in a range before the meeting, the arithmetic of return, and a comparison with the cost of the problem and of the alternatives (including doing nothing).
4. Risk sounds like "what if it fails, what if it disrupts us, what if we pick wrong?" The evidence that answers it is guarantees, pilots, phased plans, a named process, references that can be called, and a statement of what happens if it goes wrong.
5. Timing sounds like "not now." It is often a real capacity constraint, sometimes a soft no. The evidence that answers it is the cost of waiting, a smaller first step, an agreed date to resume — and a nurture sequence that keeps the conversation alive (Chapter 9).
6. Trust sounds like "who are you?" It is the objection nobody states aloud. The evidence that answers it is consistency (Chapter 7), reviews, credentials, a physical address and a real person to talk to, responsiveness, and honesty about limits.

A persona lists, for its role, the two or three objections that recur and the proof that answered them in won deals. The proof then goes where the objection arises: on the page the persona reads at the evaluation stage, in the proposal, in the sales conversation script (→ see Book E1; Book E5).

4.4.2 Risk reversal and the buyer's fear of looking foolish

Behind most objections in business purchases is a personal risk: the decider fears looking foolish to their partners, their staff or their spouse. This is why references from peers outweigh feature lists, why a guarantee that names adoption ("if your team is not using it in 60 days, we refund the setup") reassures more than one that names features, and why a small first step — a diagnostic, a pilot, a paid workshop — converts buyers a large commitment repels. Design the offer ladder (Chapter 6) with this fear in mind.

DO THE MATH What one unanswered objection costs Illustrative figures. A firm holds 8 sales meetings a month and wins 2 (25%). Its lost-deal reviews show that 3 of the 6 losses each month cite the same objection — "we could not tell what it would cost until the proposal" — a price-visibility problem, not a price problem. The firm adds a price-range page to its website, states the range in the first call and adds a one-page "how we price" document to the pre-meeting email. Suppose the change converts just one of those three monthly losses into a win: 3 customers instead of 2 — a 50% increase in new business — at an average value of $9,500 CAD, or $114,000 a year, for a page and a paragraph. Meanwhile, some prospects who would never have paid the range stop booking meetings: the calendar has fewer meetings and more customers, which is the intended effect. The objection was information; answering it early cost a day of writing.

4.5 Information Sources and Channels by Persona

4.5.1 The channel is a property of the persona

Owners choose channels by habit or fashion — "we should be on the platform everyone talks about." The persona reverses the logic: a channel is worth using if, and only if, the persona is there at a moment that matters. Where the persona looks depends on the role, the stage of the journey (Chapter 2) and the sector, and it is one of the most reliable findings of buyer interviews — people remember where they looked.

Business buyers, in most studies, rely heavily on peers and advisors (other owners, their accountant, their industry association), on search (often on a phone, outside working hours, with the trigger in the query), increasingly on AI assistants that summarize options, and on suppliers' own content once a shortlist exists. Gartner's research on B2B buying has found that buying groups spend most of their purchase time researching independently — online and offline — and only a small fraction, around a sixth, meeting with potential suppliers (Gartner, 2023). Consumers rely more on reviews, social recommendations from friends and family, comparison content and video, with search as the entry point and the retailer's or provider's page as the decision point. Neither list is universal: the persona's own answers are.

KEY CONCEPT Follow the persona, not the fashion The question is never "which channels are popular?" but "where is this person, at this stage, with this question?" An owner who reads the association newsletter and asks their accountant is reached through the association and the accountants — not through a channel that is popular with a different persona. Chapter 8 scores channels on reach, cost, fit, control and time; the reach score is nothing more than the persona's answers to "where did you look?"

4.5.2 Record the sources by stage

For each persona, the card records where they look at each stage of the journey (Figure 2.2): at the trigger, the peer's remark, the advisor, the association; in research, search terms in their words, the sites and videos they trust, the AI question they typed; in evaluation, the proof they seek — reviews, cases, references — and whom they call; at purchase, the reassurance they want; in use, how they prefer to be supported and asked. Table 4.2 gives two contrasting examples.

Table 4.2 — Where two personas look and listen (composite, illustrative)

4.5.3 Sources change; the method does not

The specific sources in Table 4.2 will be out of date within a few years — a platform will fade, an AI assistant will replace a search habit, an association will move its newsletter to a messaging application. What does not change is the method: ask recent buyers where they looked, at each stage, and record the answers with a date. Two additions keep the panel current between interview rounds. First, ask every new lead — on the form, on the first call — not only "how did you hear about us?" but "where did you look before contacting us?"; the second answer maps the research stage. Second, watch the AI assistants: type the persona's trigger question into two or three assistants every quarter and note which businesses and sources they cite; if you are absent, the fix is content that answers the question specifically and can be cited (→ see Book C5). The sources panel is therefore dated like every other volatile fact in this book, and the persona card records when it was last checked.

4.6 The VITO Rule: Talk to the Person Who Decides

4.6.1 The rule

Anthony Parinello's sales method gave the decision-maker a name — VITO, the Very Important Top Officer — and a rule: address the person who has the authority to say yes, in the language of the outcomes they are accountable for, rather than spending the sales effort on people who can only say no (Parinello, 1994). In a small business, VITO is the owner; in a household, the person whose money and risk are at stake. The rule has two halves that both matter: reach the decider (design the journey so that the owner, not only the gatekeeper, encounters you) and speak the decider's language (outcomes, risk, return, time — never features first).

4.6.2 What it means for marketing

Three practical consequences follow. Headlines and first lines address the decider's outcome: "Promise delivery dates you will keep" reaches Marc; "Advanced scheduling module" does not. Content exists for the influencer and the user too, because the decider will consult them — but it is written to help the decider decide, not to replace them. The path to the decider is designed: a referral partner who talks to owners, a workshop owners attend, an association newsletter owners read, a direct approach to the owner with a reason worth their two minutes. Gatekeepers are respected — they protect the decider's time — and given a reason the decider will thank them for.

4.6.3 When the rule misleads

VITO is a rule about where the yes lives, not a license to ignore everyone else. A decider who says yes over the objections of the users will watch the purchase fail in adoption and will not buy again; an influencer bypassed will be an enemy at renewal. In larger organizations, going over the head of a technical buyer who has been engaged is a well-known way to lose the deal (Miller & Heiman, 1985). The persona set should therefore include the roles who can block, and the message to the decider should show that the users and influencers have been considered — which is exactly what a good case study does.

4.7 Building Personas from Evidence

4.7.1 Three sources of evidence

Personas are built from three sources, in order of value. Interviews with recent buyers — customers who chose you in the last six months, and prospects who chose someone else or nothing — are the richest: they remember the trigger, the search, the shortlist, the fears and the deciding factor, and they use the words your marketing should use. Data you already have — CRM notes, the source of each lead, support tickets, the questions asked in sales calls, the pages read before a form was submitted, search queries that bring visitors — confirms and quantifies what interviews suggest. Public evidence — reviews of competitors (especially negative ones, which are objections in the wild), forum and association discussions, job postings in the segment (which reveal the problems companies are hiring to solve) — fills the gaps and checks the bias of your own customer base.

METHOD Ten interviews, one persona When to use it: to build the first version of a persona for the primary segment; to rebuild one when the market or the offer changes. Deliverable: the persona card of Figure 4.1, every line traceable to evidence. Time: about 12 hours over three weeks. 1. Recruit (2 hours). Choose 6 recent customers and 4 lost prospects in the primary segment who played the role you are profiling (usually the decider). Ask by email or phone for a 25-minute conversation "to understand how you made your decision"; offer nothing but the courtesy of a summary. Most say yes; the lost prospects are the most instructive. 2. Interview (10 × 25 minutes). Use one script, open questions, no selling: What was happening in your business when you first started looking? What did you try first? Where did you look, and what did you read or watch? Who else was involved and what did they want? What almost stopped you? What made the difference in the end? What would you tell another owner in your situation? Record with permission or take verbatim notes; the words matter as much as the facts. 3. Synthesize (3 hours). Put every answer on a card; group the cards by the six panels of the persona (job and outcome, triggers, questions and objections, proof, sources, a good first conversation). Look for what recurs in at least half the interviews; that becomes the card. What appears once is a hypothesis, marked in italics. 4. Check against data (1 hour). Compare with lead sources, sales notes and support tickets. Where the data contradicts the interviews, trust the data for what people did and the interviews for why. 5. Write and validate (2 hours). Fill the card with the buyers' own phrases where possible. Show it to whoever sells and whoever delivers: does this match the person they meet? Adjust, date, name an owner. Pitfalls: interviewing only happy customers (recruit the lost deals); asking leading questions ("did our reputation matter?"); letting one vivid interview dominate; skipping the lost prospects because they are uncomfortable to call.

4.7.2 The persona card

Figure 4.1 shows the card itself, which is the deliverable of this section. It has four zones: the identity and situation at the top, the trigger and the job to be done beside it, the objections and the sources the person trusts in the middle, and the words they use in their own language at the bottom. Each zone is filled from the interviews rather than from imagination, and a zone that cannot be filled from evidence is left visibly empty rather than invented.

Figure 4.1 — The persona card: header (name, role, decision-unit role, segment, evidence, version) and six panels — job to be done and outcome; triggers; questions and objections; proof needed; where they look and listen; what a good first conversation sounds like. Original figure; the example is fictional.

The card is one page on purpose. The header names the persona, their role in the decision unit, the segment they belong to and the evidence behind the card (how many interviews, which data). The six panels answer the six questions marketing must answer for that person. The last panel — a good first conversation — is the bridge to sales (Chapter 9) and to the sequences of Book B8: it says how this person wants to be approached, what to lead with and what to leave behind.

CASE IN POINT Two personas for one product Fictional example; names and figures are invented. Clearwater Software sells a scheduling and job-tracking tool to trades companies with 8 to 60 field employees. Its marketing spoke of "real-time visibility" and "mobile workflows" and converted poorly. Ten interviews revealed two personas with different jobs. "Owner Dominique" (economic buyer) wanted to stop being the only person who knows where everyone is, was triggered by a missed job that cost a commercial contract, feared that "the guys will never use an app", and trusted only other owners. "Dispatcher Sam" (user, and the person who usually blocked adoption) wanted to get through the morning without twenty phone calls, was triggered by nothing — the tool was imposed — and feared looking slow in front of the crews during the switch. Clearwater rewrote its site with two paths: an owners' page led by "know where every job stands without calling anyone", proven with three named owners and an adoption guarantee; a dispatchers' page led by "your morning, without the phone", with a two-week training plan and a video by a dispatcher. The sales sequence approached Dominique first (the VITO rule) and, on the first call, asked to meet Sam. Within two quarters the meeting-to-customer rate rose from 22% to 37%, and cancellations in the first 90 days fell by half — the users had been sold, not only the buyers.

4.7.3 Keeping personas alive

A persona is reviewed with the ICP (Chapter 3, Section 3.7), twice a year at most and whenever the offer, the segment or the market changes. Add every lost-deal reason and every recurring sales question to the card's evidence file; the card itself changes only when a pattern recurs. Version the card (a date and a line about what changed), and retire personas that no longer appear in the pipeline — a business that carries five personas because it never deletes one has none.

WHO DOES THIS Owning the personas Solo: the owner conducts the interviews personally — nobody learns more from ten conversations with buyers — and writes the cards in an afternoon; the cards live beside the ICP. Small team: whoever owns marketing conducts the interviews with the salesperson present for two or three of them (the salesperson hears the buyer's words without a deal at stake), synthesizes and writes; the owner validates. Established: the marketing manager owns the method and the cards; interviews are shared between marketing and sales; an external researcher is worth the cost when the company needs many interviews quickly or when internal biases are strong; the cards are the reference for every agency and freelancer brief (Chapter 12). In every case, the people who write the messages read the interview notes, not only the card.

QUÉBEC & CANADA NOTE Interviews collect personal information (as of September 2026) Interview notes and recordings are personal information about identifiable people. In Québec, the Act respecting the protection of personal information in the private sector, CQLR c P-39.1 requires that a business determine the purposes before collecting personal information, collect only what is necessary for those purposes and inform the person of the purposes at the time of collection; federally, PIPEDA sets comparable obligations. In plain terms: tell the interviewee why you are asking, what you will keep and for how long, ask before recording, store notes and recordings where only the people who need them can read them, and delete the recordings once the synthesis is done. Publishing a quotation with a name requires the person's agreement. → Chapter 10; → see Book H2.

Costly Mistakes

1. Demographic sketches called personas. Age, hobbies and adjectives predict nothing about a purchase; they give false authority to guesses and are quoted for years. Consequence: messages that describe the seller, content that answers no real question, channels chosen by fashion.
2. Personas built in a meeting room. Without interviews and data, the persona is a portrait of the team's assumptions. Consequence: the company's favorite features projected onto the buyer as needs; an expensive website that converts no better than the old one.
3. Selling the product instead of the progress. Features and methodology lead; the buyer's circumstance and outcome follow, or never appear. Consequence: the buyer cannot see themselves in the message and does not enter the consideration set until a competitor names their situation.
4. Ignoring the trigger. Without knowing what starts the search, marketing is present at random moments rather than at the one that matters. Consequence: budget spent on people not currently buying, and absence when they are.
5. Addressing the wrong role. Writing for the user when the owner decides — or winning the owner while ignoring the user who will block adoption. Consequence: deals that stall at "we need to check with the team", or purchases that fail in the first ninety days and never renew.
6. Ten personas. Each persona multiplies the content, pages and sequences to maintain. Consequence: none is maintained; the team stops reading them; the cards decay into decoration.

Key Takeaways

• A persona describes a decision as a person: role, job to be done, triggers, objections, proof, sources, preferred approach — every line traceable to evidence. It is not a demographic sketch, a segment or an ICP.
• People hire products to make progress in a circumstance. Write the job as when — I want to — so I can; the circumstance is your trigger to name, the progress your promise, the outcome your headline.
• Triggers tell you when to be present; the four forces — push, pull, anxiety, habit — tell you which lever the message and the offer must work on.
• Decision units are roles, not titles: economic buyer or decider, user, influencer, technical buyer, gatekeeper, coach. In a small business the roles are stacked on few people; in a household, too.
• Objections are information about risk; each family — need, fit, price, risk, timing, trust — calls for a specific proof, placed where the objection arises.
• The channel is a property of the persona: reach is where this person looks at this stage with this question. Follow the persona, not the fashion.
• Talk to the person who decides, in the language of their outcomes — while making sure the users and influencers who can block have been considered.
• Ten interviews make one persona: recruit customers and lost prospects, use one script, synthesize by panel, check against data, validate with sales; carry two to four personas, review them with the ICP, retire the ones that disappear.

Your Action Plan

The action plan below presents the six steps that turn this chapter into work, in the order in which each step consumes the output of the one before. For each step, the table gives the deliverable that must exist at the end, the timing on a small business's calendar, and the verification criterion by which anyone — a co-founder, a mentor, the owner's own future self — can tell that the step was done. The criterion column is the one to read first: a step whose criterion you cannot picture is a step you do not yet understand.

In La Machine

Personas live in documents; what La Machine holds is the data that builds and applies them.

• Contacts (business/marketing/contacts, detail at business/marketing/contacts/:id, edit at business/marketing/contacts/:id/edit) carry the fields that place a person in a persona — role, decision level, company, tags for the triggers and objections observed — so that segments (business/marketing/segments) can target the deciders of the primary ICP separately from the users, and sequences (business/marketing/sequences) can send each role its own version of a message. Record the persona as a tag or field on every contact you qualify.
• Companies (business/marketing/companies) hold the account; the contacts attached to a company show the decision unit as it actually appears in your data — who was involved in each deal.
• CRM client profiles and the intake form (crm/client-profile, crm/client-details, crm/intake — a client profile form in three steps, not a lead queue) are where sales conversations leave their notes — the verbatim answers to "what was happening when you started looking?" that feed the persona's evidence file — and where the qualification questions of Chapter 3 are asked.
• Templates and CTAs (business/marketing/templates, business/marketing/ctas) hold one version of each message per persona, so the decider's outcome leads for the decider and the user's day leads for the user.
• Forms and lead-magnet responses (business/marketing/forms/responses, business/marketing/lead-magnets/responses) record which content each contact requested — the persona's sources panel, measured.

What La Machine does not do: it does not conduct or transcribe your interviews, and it does not write the persona card — the card is a document built from evidence, kept beside the ICP and the book's checklist. It stores the roles, tags and notes that make the personas operational, and it keeps the personal information you collect in interviews under the consent and retention rules you set (→ Chapter 10; Book N4, CRM, Contacts and Prospect Data).

Self-Check

1. What distinguishes a persona from a segment and from an ICP?
2. Write the three-part sentence that expresses a job to be done, and say what each part becomes in your marketing.
3. Name the four forces acting on a switch, and give one marketing action that works on each.
4. Who is the economic buyer, and why does the VITO rule say to address them first — and what is the risk of addressing only them?
5. What is the test that separates a real persona from a caricature?

Answers. (1) A segment is a group in the market defined by criteria that predict behavior; the ICP is the account or household the business chooses to serve, with its exclusions; a persona is a person with a role in the decision, described by how they decide (Section 4.1.1). (2) When (circumstance) — I want to (progress) — so I can (outcome): the circumstance is the trigger to name, the progress is the promise, the outcome is the headline (Section 4.2.1). (3) Push of the situation — name the cost of the problem; pull of the new solution — show the outcome with proof; anxiety — guarantees, references, a clear process; habit — make switching easy with migration, onboarding or a pilot (Section 4.2.2). (4) The one person who can release the money and say yes when others say no — the owner in a small business; address them first because that is where the yes lives, in the language of their outcomes; but a decider who overrides the users and influencers watches adoption fail and does not renew, so the roles who can block must also be considered (Sections 4.3.1 and 4.6). (5) Every line can be traced to something a buyer said or did — an interview, a data point, a review; untraceable lines are hypotheses, marked as such (Section 4.1.2).

Chapter References

Act respecting the protection of personal information in the private sector, CQLR c P-39.1. LégisQuébec. Retrieved September 7, 2026, from https://www.legisquebec.gouv.qc.ca/en/document/cs/P-39.1

Personal Information Protection and Electronic Documents Act, S.C. 2000, c. 5. Justice Laws Website. Retrieved September 7, 2026, from https://laws-lois.justice.gc.ca/eng/acts/P-8.6/

Christensen, C. M., Hall, T., Dillon, K., & Duncan, D. S. (2016). Competing against luck: The story of innovation and customer choice. HarperBusiness.

Cooper, A. (1999). The inmates are running the asylum: Why high-tech products drive us crazy and how to restore the sanity. Sams.

Gartner. (2023). The B2B buying journey. Gartner, Inc. Retrieved September 7, 2026, from https://www.gartner.com/en/sales/insights/b2b-buying-journey

Miller, R. B., & Heiman, S. E. (1985). Strategic selling: The unique sales system proven successful by America's best companies. William Morrow.

Moesta, B. (2020). Demand-side sales 101: Stop selling and help your customers make progress. Lioncrest Publishing.

Parinello, A. (1994). Selling to VITO, the very important top officer. Adams Media.

Revella, A. (2015). Buyer personas: How to gain insight into your customer's expectations, align your marketing strategies, and win more business. Wiley.

Webster, F. E., Jr., & Wind, Y. (1972). A general model for understanding organizational buying behavior. Journal of Marketing, 36(2), 12–19.

Chapter 5 — Positioning and the Value Proposition

Two firms sell the same service at the same price in the same city. One is "a full-service agency for all your needs"; the other is "the firm that gets trades companies their first fifty online reviews in ninety days". Guess which one the trades company calls.

What You'll Be Able to Do

• Define positioning as the place you occupy in the buyer's mind relative to the alternatives — and explain why it is relative.
• Identify your real competitive alternatives and the frame of reference in which your value becomes obvious.
• Name the differences customers actually value, and separate them from the ones only you care about.
• Write a positioning statement and a value proposition that a stranger could repeat.
• Build a messaging hierarchy — pillars, proofs, calls to action — from which every page, email and conversation is drawn.
• Assemble proof that is credible, specific and lawful, and test positioning before you spend to broadcast it.

Why It Matters

Positioning is the decision that makes every other marketing dollar work harder or not at all. With a clear position, the message writes itself, the channels are obvious, the sales conversation starts from strength and price objections shrink because the buyer can see what they are paying for. Without one, the business competes as an interchangeable option and wins on price or on the salesperson's charm — neither of which compounds. For the owner, positioning is the strategic choice that follows directly from the ICP and the personas; for the team, it is the single source from which all messaging flows, so that the website, the proposals and the follow-up emails finally say the same thing.

In This Chapter

• 5.1 Positioning: The Place You Occupy in the Customer's Mind
• 5.2 Alternatives and Frames of Reference
• 5.3 Differentiation That Customers Value
• 5.4 The Value Proposition Statement
• 5.5 The Messaging Hierarchy: Pillars, Proofs, Calls to Action
• 5.6 Proof: Numbers, Cases, Reviews, Guarantees
• 5.7 Testing Positioning

5.1 Positioning: The Place You Occupy in the Customer's Mind

5.1.1 A concept about minds, not products

Al Ries and Jack Trout, who popularized the term in the 1970s, insisted on a point that is still misunderstood: positioning is not something you do to a product; it is something you do to the mind of the prospect — you position the product in the prospect's mind, relative to what is already there (Ries & Trout, 2001). The buyer's mind is crowded and defensive; it simplifies by category and by ladder ("the expensive one", "the local one", "the one my accountant uses"). A business either chooses the rung it occupies or is assigned one by default — usually "one of the many".

DEFINITION Positioning Positioning is the deliberate choice of the place your offer occupies in the mind of a defined buyer, relative to the alternatives they would otherwise consider: which category you belong to, whom you are for, what you do better or differently, and why they should believe it. It is expressed in a short statement (Section 5.4), executed through a messaging hierarchy (Section 5.5) and made credible by proof (Section 5.6). Positioning is relative by nature — it exists only against alternatives — and external: it is what the buyer concludes, not what the company declares.

Positioning is decided after the ICP (Chapter 3) and the personas (Chapter 4) and before the mix, the brand and the channels (Chapters 6 to 8). The order matters: you cannot position for everyone, and you cannot position against alternatives you have not identified.

5.1.2 Why small businesses struggle with it

Three habits keep small businesses unpositioned. Fear of exclusion — "if we say we are for trades companies, we will lose the dentist who might call." In practice the dentist rarely called, and the trades companies now call more. Confusing positioning with slogans — a clever line is not a position; a position is a choice that could be wrong, that costs something, that makes some prospects walk away. Positioning on the seller's virtues — "quality, service, experience, passion" — which every competitor also claims and which no buyer can verify before purchase. The cure for all three is the same discipline: start from the buyer's alternatives, not from your own qualities.

KEY CONCEPT Positioning is relative to alternatives A position is a comparison the buyer makes. "We are fast" means nothing; "we deliver in five days when the industry standard is three weeks" is a position, because the buyer can compare. Before writing a single line about yourself, write down what the buyer would do instead of buying from you — including doing nothing — and what is wrong with each alternative from the buyer's point of view. Your position lives in the gap between their alternatives and their desired progress (Chapter 4). If there is no gap, there is no position, and the honest move is to change the offer or the segment.

5.2 Alternatives and Frames of Reference

5.2.1 The real alternatives

Your competitors are not the businesses that look like you; they are the options the buyer actually considers for getting the job done. April Dunford's method starts precisely there — with the competitive alternatives as the customer sees them — because everything that follows (your unique attributes, the value they enable, the customers who care, the market category) can only be defined against them (Dunford, 2019). For most small businesses, the alternatives fall into five kinds:

1. Doing nothing means living with the problem. It is often the strongest competitor, and the one most marketing ignores.
2. Doing it in-house means a spreadsheet, an employee's spare time or the owner's evenings.
3. A generalist is the agency, accountant or contractor who does "everything", usually already known to the buyer.
4. A larger or branded provider is the national firm or the well-known software, chosen for safety.
5. A cheaper or informal provider is the freelancer, the relative or the platform marketplace.

For each, write what the buyer gets from it (familiarity, low price, safety) and what it costs them (time, risk, results they never see). That table is the raw material of your differentiation (Section 5.3) and of your objection handling (Chapter 4): the buyer's reasons for the alternative are the objections you will hear.

5.2.2 Choosing the frame of reference

The frame of reference — the category the buyer files you under — decides which alternatives you are compared with and which attributes count. A firm that calls itself "management consulting" is compared with large consultancies on breadth and prestige; the same firm positioned as "operations coaching for owner-led manufacturers" is compared with doing nothing and with generic software, and wins on specificity and hands-on help. Dunford distinguishes three frames a business can choose: head-to-head (compete in an existing category as the best choice for some buyers), a subsegment of an existing category (the category's solution for a specific kind of customer), and a new category (rare, expensive, and only for those who can afford to educate a market) (Dunford, 2019). For nearly every small business, the second frame — the [category] for [specific customer in a specific situation] — is the right one: it borrows the category's familiarity and adds the specificity that makes you the obvious choice for someone.

CASE IN POINT Repositioning a generalist agency Fictional example; names and figures are invented. Atelier Nord, a seven-person marketing agency, described itself as "a creative agency for ambitious brands" and competed with forty similar agencies in its region, winning about one pitch in six, mostly on price. When the partners listed their best clients by margin (Chapter 3), nine of the top twelve were trades and construction companies with 15 to 80 employees whose main problem was reputation and lead flow — reviews, a website that converted, a referral system — not creativity. Their alternatives were doing nothing, a nephew, and a national website-builder platform. Atelier Nord repositioned as "the marketing partner for trades companies that want a steady flow of qualified calls", made a single offer ("fifty reviews and a converting site in ninety days" as the entry point), rewrote the site around three pillars (reviews, calls, follow-up) with named clients and their numbers, and stopped pitching to everyone else. Twelve months later: pitches down 40%, win rate up to one in two, average contract value up 35%, and — the partners' favorite result — clients who arrived already convinced, because a peer had sent them.

5.3 Differentiation That Customers Value

5.3.1 Attributes, value, and the test of caring

A unique attribute is something you have or do that the alternatives lack: a method, a guarantee, a specialization, a speed, a location, a piece of technology, a way of pricing. An attribute only matters if it enables value the buyer cares about — an outcome in the buyer's terms (Chapter 4). The sequence is attribute → value → who cares most. "We use method X" is an attribute; "you get a plan you can execute with your current team in a week" is value; "owner-led firms without a marketing manager" are the people who care most. Most small businesses list attributes and stop; the buyer never hears the value and never recognizes themselves.

Three tests separate differentiation from decoration. The alternative test: would the buyer get this from any of the alternatives listed in Section 5.2? If yes, it is a requirement, not a differentiator. The caring test: did buyers mention this, in interviews or in won deals, as a reason they chose you? If no, it is a virtue you value and they do not. The proof test: can you demonstrate it before purchase — with a number, a case, a guarantee, a visible process? If no, it is a claim, and claims are discounted.

5.3.2 Where small businesses can genuinely differ

Small businesses rarely win on price, scale or breadth. They can win on specificity (knowing one kind of customer's world in detail — their vocabulary, rules, seasons, tools), speed and access (the owner answers; decisions are taken in a day), method (a named, visible process that reduces the buyer's risk), guarantees (which large firms are slower to offer), proof from peers (cases the buyer recognizes as their own situation), bundling and simplicity (one accountable provider for a set of adjacent needs) and place (being local where local matters). Any of these, chosen deliberately and proven, is a position; all of them claimed at once is noise.

WATCH OUT Claims you cannot prove Every differentiation claim is also a legal representation. Under the Competition Act, R.S.C. 1985, c. C-34, making a representation to the public that is false or misleading in a material respect is reviewable conduct (s. 74.01(1)(a)), and a representation about the performance, efficacy or length of life of a product must be based on an adequate and proper test — with the burden of proving the test on the person making the claim (s. 74.01(1)(b)). Penalties for corporations reach the greater of $10,000,000 or three times the benefit or 3% of worldwide gross revenues (s. 74.1). In plain terms: "fastest in the industry", "guaranteed results", "#1 rated" and specific numbers need evidence you hold before you publish, and testimonials must be genuine and typical. In Québec, the Consumer Protection Act, CQLR c P-40.1 adds rules for consumer advertising, and the Charter requires French. Chapter 10 details the rules; → see Book H4. The practical discipline is simple: for every claim, a file — the test, the data, the signed testimonial, the date.

5.4 The Value Proposition Statement

5.4.1 From position to promise

The positioning statement is the internal sentence that records the choice; the value proposition is the external promise derived from it — what the buyer gets, why it is better than the alternatives, and why they should believe it. Osterwalder and his colleagues describe the fit between a value map (the products, pain relievers and gain creators you offer) and a customer profile (the jobs, pains and gains of the segment) as the test of a value proposition: it works when the relievers and creators address the pains and gains that matter most to that customer (Osterwalder, Pigneur, Bernarda, & Smith, 2014). Chapter 4 supplied the customer side; Sections 5.2 and 5.3 supplied yours.

DEFINITION Value proposition A value proposition is a clear statement of the value a defined customer receives from your offer — the progress it enables and the pains it removes — expressed in the customer's terms, distinguished from the alternatives and supported by proof. It is not a slogan, a mission statement or a list of features; it is the answer to the buyer's question "why should I buy this, from you, rather than do something else?" A business has one value proposition per ICP (and sometimes per persona), and it precedes every message: headline, first paragraph, proposal opening, sales pitch.

5.4.2 The statement template

Geoffrey Moore's template, written for technology positioning and used ever since, gives the statement its discipline: For [target customer] who [statement of need or opportunity], [product name] is a [product category] that [statement of key benefit]. Unlike [primary competitive alternative], our product [statement of primary differentiation] (Moore, 1991). Adapted to a small business and to the work of the previous sections, the statement becomes:

For [best-fit customers, from the ICP] who [situation or trigger, from the persona], [name] is the [category / frame of reference] that [key value in the buyer's terms], because [unique attributes] — unlike [competitive alternatives], which [their limitation from the buyer's point of view].

Every bracket is filled from work already done: the customers from Chapter 3, the situation from Chapter 4, the category and alternatives from Section 5.2, the attributes and value from Section 5.3. Figure 5.1 draws the sequence as a canvas.

Figure 5.1 — Positioning canvas: competitive alternatives → unique attributes → value for the buyer → best-fit customers → market category, then the statement. Original figure; the canvas adapts the sequence described by Dunford (2019) and the statement form popularized by Moore (1991). Example entries are fictional.

5.4.3 Rules for a statement that works

One sentence — if it needs two, the position is not chosen. The buyer's words — the value clause uses phrases from the interviews, not from your brochure. A real alternative — "unlike other providers" is not an alternative; "unlike the national platforms, which leave you to configure it yourself" is. Something you would lose by — a statement that excludes no one positions no one. Provable — each clause of value and differentiation points to a proof you hold (Section 5.6). Stable — the statement changes when the ICP or the offer changes, not with the season; the messaging (Section 5.5) can vary, the position should not.

METHOD The positioning exercise (half a day) When to use it: after the ICP and the first persona exist; when win rates fall or price objections rise; before a website or brand project. Participants: owner, whoever sells, whoever delivers; a facilitator helps. Deliverable: the canvas of Figure 5.1 filled, the statement in one sentence, and the three pillars of Section 5.5. Step 1 — Alternatives (30 min). List what your best-fit customers would do without you, from lost-deal reviews and interviews: doing nothing, in-house, generalist, big brand, cheap or informal. For each: what they get, what it costs them. Step 2 — Attributes (30 min). List everything you have or do that those alternatives lack. Be concrete: methods, guarantees, speed, specialization, pricing, access. Strike anything an alternative also offers. Step 3 — Value (30 min). For each surviving attribute, write what it enables for the buyer, in the buyer's words from interviews. Strike attributes whose value nobody mentioned. Step 4 — Who cares most (20 min). Which customers, in which situation, value that most? Check against the ICP; tighten the ICP if the answer is narrower than the profile. Step 5 — Category (20 min). Choose the frame of reference — usually "the [category] for [those customers]" — in which the value is obvious and the right alternatives are compared. Step 6 — Statement (30 min). Fill the template; read it aloud; test it against the six rules above. Then derive three pillars (Section 5.5), each with one proof you already hold. Step 7 — Decide what you will stop saying (20 min). List the claims, services and audiences the new position excludes from the website and the pitch. If the list is empty, repeat step 4. Pitfalls: listing competitors instead of the buyer's alternatives; attributes without value; value without a buyer's quote; a statement that pleases the team and excludes no one; skipping step 7.

5.5 The Messaging Hierarchy: Pillars, Proofs, Calls to Action

5.5.1 One source for every message

The position is a sentence; a business needs hundreds of messages — pages, emails, posts, proposals, scripts. Without a structure between the two, every message is written from scratch and the company says something slightly different each time, which the buyer registers as inconsistency. The messaging hierarchy is that structure: the value proposition at the top; three pillars beneath it — the three reasons the value is true, each a theme the buyer cares about; under each pillar, proofs — the specific evidence that makes it credible; and for each stage of the journey, a call to action — the next step you propose. Every message is assembled from these parts; none introduces a claim that is not in the hierarchy. Table 5.1 shows the hierarchy for the fictional Atelier Nord, one level per row: the content of each level and the rule that governs it.

Table 5.1 — The messaging hierarchy (Atelier Nord example, fictional)

5.5.2 Pillars: three, not seven

Three pillars is a discipline, not a rule of nature: buyers remember three reasons, and a business can prove three things well. Pillars come from the value column of the canvas (Section 5.3), grouped into themes; each is phrased as an outcome or a promise the buyer cares about, not as a department or a service line. A useful check: could a competitor honestly put the same pillar on their site? If yes, sharpen it with a number, a method or a guarantee until they could not.

5.5.3 Calls to action: one per stage

A call to action is the next step proposed to the reader. It must match the stage of the journey (Chapter 2): a person at the trigger stage will not "request a quote", but will download a guide that names their problem; a person in evaluation will not download a guide, but will book an audit. The hierarchy therefore holds one call to action per stage, ordered from the smallest commitment to the largest — the offer ladder of Chapter 6 — and each page or message uses the one that fits where its reader is. Book B6 (Copywriting and Persuasive Communication) refines the wording; Book C2 (Websites, Landing Pages and Conversion) places them on the page.

5.5.4 One hierarchy, several audiences: adapting the message by persona and stage

The hierarchy is single; its expression varies with the reader. The position and the pillars do not change from one persona to another — that would be several positions, which is none — but the order in which the pillars are presented, the proof chosen under each and the call to action offered all change with the role of the reader (Chapter 4) and the stage they are at (Chapter 2). A decider reads the pillar about outcomes and risk first and wants a peer's case and a guarantee; a user reads the pillar about daily ease first and wants a training plan and a colleague's testimonial; an influencer — the accountant, the IT provider — reads the pillar about method and compatibility and wants documentation and transparency on price. Table 5.2 shows the same hierarchy expressed three ways for the Atelier Nord example.

Table 5.2 — The same hierarchy, three readers (Atelier Nord, fictional)

Two rules keep the variations honest. Nothing in a variation contradicts another — the decider's page and the user's page describe the same offer, the same price and the same process, or the buyers will compare notes and find the seams. Every variation is drawn from the hierarchy — if a page needs a claim that is not under a pillar, either the hierarchy is incomplete or the claim does not belong. In practice, a small business maintains one hierarchy document with a column per persona for the order, the proof and the call to action; the templates and pages are generated from those columns, and updated when the hierarchy is.

The stage adds a second axis. At the trigger stage, the message names the circumstance and offers something free that helps; at the research stage, it explains the options and states the price range; at evaluation, it proves; at purchase, it reassures and shows what happens next; in use, it educates and asks for the review and the referral. A message that proves to someone who has not yet recognized their problem is wasted; a message that names the problem to someone who is already comparing quotes is patronizing. The messaging hierarchy, read against Figure 2.2's journey map, produces a small matrix — persona by stage — that tells the team, for any message they are about to write, which pillar leads, which proof appears and which next step is proposed. Book B8 (Marketing Automation and Funnels) turns that matrix into sequences.

5.6 Proof: Numbers, Cases, Reviews, Guarantees

5.6.1 The six kinds of proof

Proof is what turns a claim into a belief. Six kinds recur, and a small business should hold at least three.

1. Numbers are results with their base and period ("median 3.1× inquiries across 11 sites in 2024–2025"), operational facts ("answered within one hour, 96% of the time in 2025") and scale where relevant ("14 trades clients"). Numbers need a source you can show and a date; averages need a base; "up to" is a warning sign.
2. Cases tell a named customer's story: their situation and trigger, what was done, what changed, in their words where possible. A case fits on one page; it is the persona's own story told back to them. Cases convert because they let the buyer recognize themselves (Chapter 4).
3. Reviews and testimonials are third-party voices on platforms the buyer trusts, and quoted testimonials on your own pages. They must be genuine, current and representative of typical results, with the source and date shown.
4. Guarantees and commitments are a promise with a consequence: money back, work until done, a service level in the contract. Guarantees reverse the buyer's risk (Chapter 4) and signal confidence; they must be written precisely and honored without argument.
5. Credentials and affiliations — licenses, certifications, professional orders, association memberships, awards — are proof of competence and standing, especially where the buyer cannot judge the work directly.
6. Visible process is the method shown before purchase: the diagnostic, the ninety-day plan, the weekly report template. A visible process proves how, which reduces the anxiety force of Chapter 4.

5.6.2 Collecting proof as a routine

Proof does not accumulate by itself. Three routines make it a by-product of the work: ask for the review at the moment of satisfaction (the delivery that went well, the milestone reached — a task in Chapter 11's table); write the case at the end of every engagement that fits the ICP, with the customer's agreement, while the numbers are fresh; and record the operational numbers (response times, results, counts) monthly, so that next year's proof exists. Book B10 covers testimonials and reviews systematically; here the point is that proof is planned, not hoped for.

QUÉBEC & CANADA NOTE Testimonials, reviews and the law (as of September 2026) The Competition Act treats a testimonial as a representation: it must be genuine, reflect the honest opinion and typical experience of a real customer, and — when it makes a performance claim — rest on the same adequate and proper test as any other claim (s. 74.01). Fake reviews, paid reviews presented as spontaneous, and cherry-picked results presented as typical are deceptive marketing; the Competition Bureau has published guidance on online reviews and influencer disclosure, and the Bureau enforces against fake or misleading reviews. Quoting a customer by name is also a use of personal information: obtain and record their agreement (P-39.1, PIPEDA). In Québec, testimonials and claims in consumer advertising must comply with the Consumer Protection Act and appear in French under the Charter of the French language. → Chapter 10; → see Book H4. Practical rule: a signed authorization and a dated file for every testimonial and every number you publish.

5.6.3 Anatomy of a case study, and the arithmetic of a number

The case study is the proof that does most of the work for a small business, because it lets the buyer recognize themselves. It is also the proof most often done badly — a page of praise with no numbers, or a project description with no outcome. A case that converts has six parts, in this order, on one page: the customer (who they are, in the terms of the ICP — sector, size, situation — so the reader can place themselves); the trigger (what was happening when they started looking, in their words — the persona's trigger panel made concrete); the alternatives they considered (including doing nothing) and why those were unsatisfactory; what was done, briefly — the method, the timeline, the people involved, without becoming a project report; what changed, with numbers that have a base and a period ("inquiries rose from 11 to 34 a month over the second quarter"); and the customer's own assessment, a short quotation about the experience, with their name, role and company, authorized in writing. A photograph of the customer's premises or team, taken with permission, adds recognition; a stock image subtracts it. Book B6 covers the writing; here, the point is that a case is structured evidence, and the structure is what makes it credible.

Numbers deserve their own discipline, because a number is the proof most easily inflated and most easily checked. Four rules keep numbers honest and persuasive at once. State the base: "3.1× more inquiries" means nothing without "across 11 rebuilt sites"; a percentage means nothing without the count it is a percentage of. State the period: results from 2019 are not results from 2026; "as of March 2026" dates the claim and signals that you will update it. Prefer medians to averages when results vary widely: an average lifted by one exceptional client misleads; the median is what a typical client saw — and the Competition Act's concern with the general impression of a representation (Section 5.3.2) makes the typical result the one to show. Show the range when it is wide: "between 1.8× and 6.2×, median 3.1×" is more credible than the median alone, because it admits variation. Each published number then gets a line in the proof file: the data it came from, the calculation, the date and who approved it. A number a competitor could not attack is a number you can defend line by line; that is what makes it worth putting on the page.

5.7 Testing Positioning

5.7.1 Test before you broadcast

A position is a hypothesis about the buyer's mind. Test it cheaply before rebuilding a website or printing a brochure around it. Four tests, in increasing cost:

The repeat test. Read the statement to five people who match the persona but do not know you. Ask them to say back, in their words, what you do and for whom. If they cannot, or they say something different, the statement is not clear. Free; one afternoon.

The sales-conversation test. For one month, open every first conversation with the new position and the three pillars, and record the reaction: does the prospect say "that's us", ask a question that shows they understood, or look puzzled? Track meeting-to-proposal rate against the previous month. Free; one month.

The page test. Publish the new value proposition and pillars on one page — a landing page or the home page — and compare visitor-to-lead rate with the previous version over enough visitors (Chapter 2, Section 2.6.3). Low cost; four to eight weeks.

The message test. Send two versions of the same email to two halves of a fitting segment — one with the old framing, one with the new — and compare reply and meeting rates. Low cost; two weeks; needs a list with consent (Chapter 10).

DO THE MATH What a sharper position is worth Illustrative figures. A firm's home page converts visitors to leads at 1.8% (18 leads from 1,000 monthly visitors), and leads to customers at 10% (1.8 customers a month) at $8,000 CAD of first-year value. It rewrites the page around a specific position and three proven pillars. Two effects are typical: fewer but better-fitting visitors act — say conversion rises to 2.6% (26 leads) — and leads that fit convert better — say 14% (3.6 customers). Result: from 1.8 to 3.6 customers a month, +$14,400 of monthly first-year value, or about $173,000 a year, from the same traffic. The cost: a half-day exercise, two days of writing, and the courage to stop saying "we do everything". Even if the effects were half as large, the return on the exercise would still exceed most channel spending — which is why positioning comes before channels.

5.7.2 Signs the position is working — and when to revisit

Working: prospects describe you the way you describe yourself; referrals arrive pre-qualified; price objections fall while win rates rise; competitors begin to copy your language. Revisit when the ICP changes (Chapter 3), when a new alternative appears that makes your differentiation ordinary, when win rates fall in the primary segment for a reason interviews trace to the message, or when the offer changes materially. Do not revisit because the team is bored with the statement — buyers hear it far less often than you do.

WHO DOES THIS Owning the position and the messaging Solo: the owner runs the half-day exercise alone (with the canvas and the interview notes), writes the statement and pillars, and asks five fitting strangers the repeat test; a copywriter can polish the words but must not choose the position. Small team: the owner leads the exercise with the seller and the deliverer; whoever owns marketing maintains the hierarchy document and checks every new page, email and proposal against it — the hierarchy is the brief for every freelancer. Established: the marketing manager owns the hierarchy and the proof files; sales confirms the pillars in the field; the owner or general manager approves any change to the statement, because it commits the company. In all three, the position is the owner's decision — it is the choice of whom to be for, and it cannot be outsourced.

Costly Mistakes

1. Positioning on the seller's virtues. "Quality, service, experience, passion" — claimed by every competitor, verifiable by no buyer. Consequence: the buyer defaults to price, and the salesperson's charm becomes the only differentiator.
2. Listing competitors instead of the buyer's alternatives. Ignoring "do nothing", "do it in-house" and "the nephew" leaves the strongest competitors unaddressed. Consequence: messaging that beats firms the buyer never considered and loses to inertia.
3. A statement that excludes no one. A position everyone can agree with positions no one. Consequence: awareness without preference; leads that do not fit; a website that describes an interchangeable option.
4. Claims without files. Publishing "fastest", "guaranteed", "#1" or a number without the test, data or authorization behind it. Consequence: exposure under the Competition Act and the Consumer Protection Act, a testimonial withdrawn at the worst moment — and, more often, buyers who discount every claim because one was hollow.
5. Seven pillars. Trying to prove everything proves nothing memorable. Consequence: pages that read like a catalog; buyers who cannot repeat what you do.
6. Rebuilding the website before testing the position. Committing $10,000–$40,000 CAD to a site around an untested statement. Consequence: an expensive way to learn what a five-person repeat test and a one-month sales test would have shown.

Key Takeaways

• Positioning happens in the buyer's mind, relative to alternatives — including doing nothing and doing it in-house. It is a choice that excludes; a statement no one could disagree with positions no one.
• Start from the alternatives, then the attributes they lack, then the value those attributes enable in the buyer's words, then who cares most, then the frame of reference in which the value is obvious — usually "the [category] for [specific customer in a specific situation]".
• Differentiation must pass three tests: the alternatives do not offer it; buyers said it mattered; you can prove it before purchase.
• The statement is one sentence — For … who …, [name] is the … that …, because … — unlike …, which … — filled from the ICP, the persona and the canvas, and stable until the ICP or the offer changes.
• The messaging hierarchy — value proposition, three pillars, proofs under each, one call to action per stage — is the single source of every page, email, proposal and script.
• Proof comes in six kinds — numbers, cases, reviews, guarantees, credentials, visible process — and is collected as a routine, with a signed authorization and a dated file behind every claim.
• Claims are legal representations: false or misleading representations and untested performance claims are reviewable under the Competition Act; testimonials must be genuine and typical; in Québec, consumer advertising follows the Consumer Protection Act and appears in French.
• Test before you broadcast: the repeat test, the sales-conversation test, the page test, the message test — in that order of cost.

Your Action Plan

The action plan below presents the six steps that turn this chapter into work, in the order in which each step consumes the output of the one before. For each step, the table gives the deliverable that must exist at the end, the timing on a small business's calendar, and the verification criterion by which anyone — a co-founder, a mentor, the owner's own future self — can tell that the step was done. The criterion column is the one to read first: a step whose criterion you cannot picture is a step you do not yet understand.

In La Machine

The position lives in a document; La Machine carries the messaging that executes it, so that every message the platform sends says the same thing.

• Templates (business/marketing/templates, editor at business/marketing/templates/:id/edit) hold the email and message templates derived from the hierarchy — one set per persona, each opening with the value proposition and one pillar, with the proof attached. When the hierarchy changes, the templates change in one place.
• CTAs (business/marketing/ctas, editor at business/marketing/ctas/:id/edit) hold the calls to action of Section 5.5.3 — one per stage — so that forms, landing pages and emails reuse the same next steps instead of inventing new ones.
• Landing pages (business/marketing/landing-pages, editor and /responses) are where the page test of Section 5.7.1 runs: publish the new value proposition and pillars on one page and read the response counts by source; forms (business/marketing/forms) capture the lead with its source, as Chapter 2 requires.
• Campaigns and sequences (business/marketing/campaigns, business/marketing/sequences) carry the message test: two versions of the same message to two halves of a consented segment, with the sends read from the log (business/marketing/log) and the replies and meetings counted by the team, since the log records only what left.
• Branding (business/branding) holds the visual identity within which the messaging appears — the subject of Chapter 7 and Book B3.

What La Machine does not do: it does not choose your position or write your pillars — those are the owner's decisions and the team's writing; it does not hold the proof files themselves (tests, data, signed authorizations), which belong in your document system with a date on each; and it does not check your claims against the Competition Act — the file-behind-every-claim discipline is yours. Book N5 (Marketing Automation in La Machine) shows templates, CTAs, landing pages and sequences page by page; Book N7 covers the website and lead generation pages.

Self-Check

1. Why is positioning described as relative, and what does that imply about the first step of the exercise?
2. Name the five kinds of competitive alternative a small business usually faces. Which one is most often ignored?
3. State the three tests that separate genuine differentiation from decoration.
4. What are the levels of the messaging hierarchy, and why does it hold one call to action per stage?
5. Under the Competition Act, what must a performance claim rest on, and who bears the burden of proof?

Answers. (1) Because a position exists only against the alternatives in the buyer's mind; the first step is therefore to list what the buyer would do instead of buying from you, including doing nothing (Sections 5.1 and 5.2.1). (2) Doing nothing; doing it in-house; a generalist; a larger or branded provider; a cheaper or informal provider. Doing nothing is most often ignored — and often the strongest (Section 5.2.1). (3) The alternatives do not offer it; buyers said it mattered (in interviews or won deals); you can prove it before purchase (Section 5.3.1). (4) Value proposition → three pillars → proofs under each → one call to action per journey stage; because a reader at the trigger stage will not request a quote and a reader in evaluation will not download a beginner's guide — the next step must match where the reader is (Section 5.5). (5) An adequate and proper test, made before the claim; the person making the representation bears the burden of proving it (Competition Act s. 74.01(1)(b)) (Section 5.3.2).

Chapter References

Act respecting the protection of personal information in the private sector, CQLR c P-39.1. LégisQuébec. Retrieved September 7, 2026, from https://www.legisquebec.gouv.qc.ca/en/document/cs/P-39.1

Charter of the French language, CQLR c C-11. LégisQuébec. Retrieved September 7, 2026, from https://www.legisquebec.gouv.qc.ca/en/document/cs/C-11

Competition Act, R.S.C. 1985, c. C-34, ss. 74.01, 74.1. Justice Laws Website. Retrieved September 7, 2026, from https://laws-lois.justice.gc.ca/eng/acts/C-34/

Consumer Protection Act, CQLR c P-40.1. LégisQuébec. Retrieved September 7, 2026, from https://www.legisquebec.gouv.qc.ca/en/document/cs/P-40.1

Personal Information Protection and Electronic Documents Act, S.C. 2000, c. 5. Justice Laws Website. Retrieved September 7, 2026, from https://laws-lois.justice.gc.ca/eng/acts/P-8.6/

Dunford, A. (2019). Obviously awesome: How to nail product positioning so customers get it, buy it, love it. Ambient Press.

Moore, G. A. (1991). Crossing the chasm: Marketing and selling high-tech products to mainstream customers. HarperBusiness.

Osterwalder, A., Pigneur, Y., Bernarda, G., & Smith, A. (2014). Value proposition design: How to create products and services customers want. Wiley.

Ries, A., & Trout, J. (2001). Positioning: The battle for your mind (20th anniversary ed.). McGraw-Hill.

Chapter 6 — The Marketing Mix

A premium consulting firm with a website that looks like a discount flyer. A craft product sold through a channel that piles it next to the cheapest alternative. A service promised as "effortless" whose onboarding takes six forms and three weeks. Each business is contradicting itself — and the buyer notices before the owner does.

What You'll Be Able to Do

• Name the elements of the marketing mix — the four Ps and the seven Ps — and explain what decision each one covers.
• Read your offer as a set of decisions about product, price, place and promotion, plus people, process and physical evidence for services.
• Recognize price as a marketing signal, not only a number, and connect it to positioning.
• Audit your mix for coherence with a grid, and spot where one P contradicts another.
• Build an offer ladder from a free first step to the full purchase.
• Decide the two Ps to fix first, with a deliverable and a date.

Why It Matters

Positioning (Chapter 5) is a promise; the mix is where the promise is kept or broken. Every element of the offer — what it includes, what it costs, where it can be bought, how it is described, who delivers it, how the process feels, what the customer sees and touches — either confirms the position or contradicts it. Buyers do not read the position; they infer it from the mix. For the owner, the mix is the checklist that turns a strategy into an offer people can buy; for the team, it is the shared vocabulary for the decisions that shape the customer's experience before, during and after the sale.

In This Chapter

• 6.1 The Four Ps and the Seven Ps
• 6.2 Product and Offer Decisions
• 6.3 Price as a Marketing Signal
• 6.4 Place: Channels and Access
• 6.5 Promotion: The Communication Mix
• 6.6 People, Process and Physical Evidence in Services
• 6.7 Coherence: When One P Contradicts Another

6.1 The Four Ps and the Seven Ps

6.1.1 A framework that has lasted since 1960

E. Jerome McCarthy organized the decisions a marketer controls into four groups — product, price, place, promotion — in a textbook first published in 1960 (McCarthy, 1960). The framework survived because it is complete for physical goods and simple enough to remember: what you sell, what you charge, where the buyer can get it, how the buyer learns about it. Two decades later, Bernard Booms and Mary Jo Bitner observed that services — which are produced and consumed at the same time, often by people, and cannot be inspected before purchase — needed three more groups: people (who delivers and who serves), process (how the service is delivered, step by step) and physical evidence (everything tangible the customer sees, from the premises to the invoice) (Booms & Bitner, 1981). Together, the seven Ps describe the controllable elements of a business's offer; Kotler and Keller treat them as the toolkit that executes a positioning (Kotler & Keller, 2016).

DEFINITION Marketing mix The marketing mix is the set of controllable decisions a business makes about its offer in order to produce the response it wants from its target segment: product (what is offered and what it includes), price (what is asked in exchange and how), place (where and how the buyer can obtain it), promotion (how the buyer learns about it and is persuaded) — and, for services, people (who delivers and serves), process (how delivery happens) and physical evidence (the tangible signals of quality). The mix is the execution of the positioning: each element should confirm the place the business has chosen in the buyer's mind (Chapter 5). In this library, product is developed in Book A6, price in Book A7, place in Collections C, D and F, promotion in Collections B, C and D; this chapter shows how they fit together.

6.1.2 The buyer's side of each P

The four Ps describe the seller's decisions. A useful complement, proposed by Robert Lauterborn, restates each from the buyer's side: not product but the customer's wants and needs; not price but the cost to the customer (money, time, effort, risk); not place but convenience of access; not promotion but communication, a two-way exchange (Lauterborn, 1990). The restatement is not a replacement — the seller still decides — but a test: for each P, ask what it costs and gives the buyer, and whether the answer matches what the persona (Chapter 4) said matters.

KEY CONCEPT The mix is a coherence test The seven Ps are not seven independent optimizations. They are seven expressions of one position, and the buyer reads them together. A single element that contradicts the others — a premium price on a bargain-looking website, an "expert" positioning delivered by a junior on the first visit, a "simple" offer with a complicated order process — makes the whole offer less credible than its weakest element. The first use of the mix is therefore not to improve each P but to check that they agree; Section 6.7 gives the grid.

6.2 Product and Offer Decisions

6.2.1 Product, service, and the offer around them

In marketing, the product is everything the buyer receives — the core good or service, the features and options, the quality level, the packaging, the warranty, the support, the brand — and the offer is the product as it is proposed at a given moment: bundled, priced, with terms and a call to action. The same service can be several offers: a diagnostic, a project, a monthly retainer, a workshop. Book A6 (Designing a Value-Added Offer) covers how to design what you sell; this section covers the offer decisions that marketing owns.

Four decisions recur. Scope — what is included and, more importantly, what is not; a clearly bounded offer is easier to buy than an open-ended one. Levels — good, better, best; two or three tiers let different segments self-select and anchor the middle option. Packaging — a name, a fixed set of deliverables and a fixed price for a common need turn a service into something that can be listed and bought ("the ninety-day reputation program"), which shortens the sales cycle and stabilizes delivery. Guarantees and terms — the risk-reversal of Chapter 4, and the conditions (duration, cancellation, payment) that make the offer safe to accept.

6.2.2 The offer ladder

Buyers at different stages of the journey (Chapter 2) can accept different levels of commitment. The offer ladder arranges your offers from the smallest to the largest, so that every stage has something the buyer can say yes to:

1. Free and useful is a guide, a checklist, a calculator or a webinar: the lead magnet exchanged for consent and a contact detail (→ see Book B9).
2. Low-commitment and diagnostic is an audit, an assessment, a first consultation or a sample: a small price or free, high value for the buyer, and qualifying for you.
3. The entry offer is a bounded first purchase with a fixed price and a clear result: the "start with reviews only" of Chapter 5.
4. The core offer is the main product or service, in its tiers.
5. Expansion is the add-ons, upgrades, retainers and renewals: the flywheel's force (Chapter 2).

A ladder with a missing rung loses the buyers who were standing on it. Most small businesses have rungs 4 and sometimes 1 and lack 2 and 3 — the steps that convert an interested stranger into a paying customer with limited risk.

6.2.3 Tiers and packaging: rules that make an offer easy to buy

Tiers. Offering the core service in two or three versions — often labeled good, better, best, or named for the customer situation they serve — does three things at once. It lets segments self-select by budget and need without a negotiation; it anchors the middle option, which most buyers choose when three are shown, so the price the business most wants to sell becomes the reasonable one; and it turns "no" into "not this tier", keeping the conversation open. Rules that make tiers work: three, rarely four (more produces paralysis); differences the buyer values, not arbitrary feature removal (scope, speed, level of support, guarantee — never the removal of something essential to the result); a clear recommendation ("most trades companies choose this one"); and prices that scale with value delivered, not with the effort of producing them. The top tier exists partly to make the middle one look moderate and partly to serve the customers who genuinely need it; it is rarely the volume seller.

Packaging. A package is a named, bounded set of deliverables at a fixed price for a common, recognizable need — "the ninety-day reputation program", "the year-end close package", "the first-hire HR kit". Packaging converts a service the buyer cannot picture into a product they can compare, and it stabilizes delivery because the team does the same thing repeatedly. Rules: the package solves one job the persona named (Chapter 4), not everything you could do; its scope is written, including what is excluded and what would be an add-on; its price is fixed or stated as a narrow range; it has a name the buyer can repeat; and it has a result the buyer can verify ("fifty reviews", "closed in five business days"). A business that packages its most common engagement usually finds that the package sells more easily than the bespoke version it replaced, at a better margin, because the buyer's anxiety (Chapter 4) is lower and the seller's cost to serve is known.

Guarantees and terms. The terms of an offer — duration, cancellation, payment schedule, what happens if the result is not reached — are part of the product as the buyer experiences it. A guarantee is the strongest risk-reversal available to a small business, precisely because larger competitors are slower to offer one. It must be specific (what is guaranteed, measured how, over what period), honored without argument (a disputed guarantee destroys more trust than no guarantee), and priced in (a small share of engagements will invoke it; the cost belongs in the margin). Payment terms are also a signal: a deposit signals seriousness on both sides; monthly billing lowers the entry barrier; a discount for annual payment trades margin for cash and commitment (→ see Book A7; Book G6, Invoicing, Payments and Collections).

CASE IN POINT A premium price with a discount-store website Fictional example; names and figures are invented. Halden & Roy, a two-partner architecture and interior design studio, positioned itself on craft and attention — fees 30% above the local average, justified by detailed design development and site supervision. Its website, built in 2018 by a friend, showed low-resolution photos, a countdown banner left over from a promotion, three typefaces and a "Get a FREE quote!!" button. Inquiries were plentiful and wrong: budget-conscious homeowners who balked at the fee, then vanished. The partners assumed a lead-quality problem and considered advertising. The mix audit (Section 6.7) showed a coherence problem: product and price said premium; physical evidence (the site) and promotion (the free-quote button) said discount. They rebuilt the site around six large project photographs, a "how we work" page with the fee range stated, and a single call to action — "book a 45-minute project conversation ($250, credited to your project)". Inquiries fell by half; the share that fit rose from one in five to two in three; the average project value rose 28%. No advertising was bought.

6.3 Price as a Marketing Signal

6.3.1 What a price says

Price is the only P that produces revenue; all the others produce costs. It is also the fastest signal a buyer reads: before they know anything about quality, they infer it from the price relative to the alternatives (Chapter 5). A price well below the alternatives says "cheaper, and probably less" — which is a position, if you can sustain it; a price well above says "better, and here is why" — which is a promise the rest of the mix must keep. A price that is hidden says "expensive, and they are embarrassed", and it sends the research-stage buyer (Chapter 2) to a competitor who states one. Book A7 (Pricing Strategy and Revenue Models) treats pricing methods in depth — cost-plus, value-based, competitive, subscription; this section treats price as communication.

6.3.2 Four pricing decisions marketing owns

Level relative to alternatives. Below, at or above — a positioning decision, not an accounting one. Most small businesses under-price relative to the value they deliver because they price from their costs and their fear, not from the buyer's alternatives and outcomes.

Structure. Hourly, per project, per month, per unit, per result, tiered — the structure changes how the buyer perceives risk and value. A fixed project price transfers risk to you and is easier to buy; an hourly rate transfers it to the buyer and invites scrutiny of every hour; a monthly retainer smooths your revenue and the buyer's budget; tiers let segments self-select and make the middle tier look reasonable.

Presentation. A price shown as a range on the website, anchored against the cost of the problem or of the alternative, with what is included spelled out, converts better than the same price discovered at the end of a sales process. Presentation also includes the order in which options appear and the unit (per month, per employee, per year) — each frames the number.

Promotions. Discounts, introductory prices, bundles, seasonal offers — useful to move a buyer at a specific moment, dangerous as a habit: a business that always discounts has taught its customers to wait, and has repositioned itself as the discount option without deciding to.

6.3.3 Pricing structures compared

The structure of a price shapes how the buyer perceives risk, how the seller's revenue behaves and what each party is motivated to do. Table 6.1 compares the structures a small business chooses among; the choice is a marketing decision because it changes the offer's meaning, and an economic one because it changes cash flow and margin (→ see Book A7 for the methods behind the numbers).

Table 6.1 — Pricing structures: what each signals, when to use it, what to watch

Most small businesses end up with a combination: a fixed-price entry package (rung 3 of the ladder), a retainer or subscription for the ongoing relationship (rung 4), and hourly or per-unit pricing only for work that genuinely cannot be scoped. The combination is itself a message — "we know what this costs, we will tell you before we start, and we stay with you afterwards" — and it is the one buyers in most services markets describe as fair when interviewed (Chapter 4).

DO THE MATH Anchoring the price against the cost of the problem Illustrative figures. A firm proposes a $24,000 CAD operations project to a 40-employee distributor. Presented alone, the number is compared with "doing nothing" (which appears free) and with a cheaper generalist ($14,000). Presented with the arithmetic, it is compared with the problem: the distributor's late deliveries cost about 3 lost orders a month at $2,800 of margin each, plus roughly 12 hours a week of the owner's firefighting — say $8,400 of margin and 50 hours a month. If the project halves the late deliveries, it recovers ≈ $4,200 of margin a month, or $50,400 a year, plus 300 hours of the owner's time. The project pays for itself in about six months on margin alone; the generalist, with no method and no guarantee, is not cheaper if it recovers nothing. The proposal page states this arithmetic before the price. The number did not change; its meaning did.

QUÉBEC & CANADA NOTE Price presentation and the law (as of September 2026) Prices are representations. Federally, the Competition Act, R.S.C. 1985, c. C-34 treats a false or misleading representation as reviewable (s. 74.01), and, since the 2022 amendments, expressly treats drip pricing — advertising a price that cannot be attained because of mandatory fixed charges added later — as a false or misleading representation, unless the added charge is imposed by law (such as sales tax). In Québec, the Consumer Protection Act, CQLR c P-40.1 governs how prices are displayed to consumers, including the requirement that the advertised price be the total the consumer must pay, apart from taxes; the Office de la protection du consommateur (OPC) enforces it. Regular-price claims ("was $200, now $120") require that the regular price be genuine. Business-to-business sales have fewer consumer-protection rules but the same Competition Act. → Chapter 10; → see Book H4; → see Book A7 for pricing methods. Practical rule: the price the buyer sees first is the price they pay, apart from taxes, and every "was" price is real and documented.

6.4 Place: Channels and Access

6.4.1 Place is access

For a physical product, place is distribution: where the product is stocked and sold, how it moves, who handles it. For a service, place is access: where and when the buyer can find, buy and receive it — your premises, their premises, online, by phone, through a partner, in a marketplace. Decisions about place include direct versus indirect routes (sell yourself or through distributors, resellers, marketplaces, referral partners — Chapter 2, Collection F), coverage (how many outlets or partners, in which territories — Chapter 3), availability (hours, response times, lead times, delivery radius), and the buying experience at the point of purchase — how easy it is to order, pay, schedule.

6.4.2 Place carries the position

Place is a position signal as strong as price. A craft product sold through a discount marketplace is repositioned by its neighbors; a premium service that can only be bought after four emails and a callback is repositioned by its friction; a local business that ships nationally has changed its market and its competitors. Choosing place therefore follows positioning and the ICP: where does the best-fit buyer expect to find this kind of offer, and what does that place say about it? Chapter 8 places the marketing channels — how the buyer learns about you — alongside the distribution channels — how the buyer obtains what you sell; the two are often the same pipe.

6.4.3 Availability, lead times and the buying experience

Three aspects of place are decided by small businesses without noticing that they are decisions. Availability — the hours the phone is answered, the days a technician can come, the time zone a service is delivered in — either matches the persona's life or fights it: a homeowner who can only call at 6 p.m. and a business that answers until 5 p.m. never meet. Lead time — how long between the yes and the delivery — is a competitive variable in most local markets, where the provider who can start next week wins against the one who can start in six, and a place where honesty pays: a stated, kept lead time beats an optimistic one that slips. The buying experience — how many steps, forms, calls and days separate "I want this" from "it is done" — is the friction of Chapter 2 made concrete, and it is where the offer ladder's lower rungs live or die: a diagnostic that takes three emails to schedule is not low-commitment. Each of the three can be measured (share of calls answered live; median lead time; steps to purchase), each can be stated on the website as a promise, and each, when it matches the position, becomes a differentiator no competitor can copy quickly (Chapter 5). Book J2 designs the processes behind them; here they are named as place decisions so that the mix audit (Section 6.7) asks about them.

6.5 Promotion: The Communication Mix

6.5.1 The elements

Promotion is the oldest word for what most people now call marketing communication — the set of means by which the business informs, persuades and reminds its segments. Kotler and Keller divide it into advertising, sales promotion, events and experiences, public relations and publicity, direct and database marketing, online and social-media marketing, mobile marketing, and personal selling (Kotler & Keller, 2016). This book treats the strategic choice among them in Chapter 8 (channels), the message in Chapter 5 (positioning and messaging), the rules in Chapter 10, and each element in its own book in Collections C and D. Here, three points connect promotion to the rest of the mix.

Promotion cannot repair the other Ps. Communication makes a good offer known; it cannot make a wrong offer right. A business whose leads do not convert usually has a product, price or process problem that promotion is being asked to shout over. Diagnose the mix (Section 6.7) before buying reach.

Promotion must tell the truth about the other Ps. What the message promises — speed, simplicity, expertise — the product, the process and the people must deliver, or the flywheel's friction (Chapter 2) destroys what the promotion built. The messaging hierarchy of Chapter 5 is written from what the business can prove, precisely so that promotion and delivery agree.

Promotion is regulated. Claims, testimonials, contests, comparative advertising, consent for messages, language — each element has rules (Chapter 10, Books H3 and H4). The mix decision "what will we say" carries the obligation "what can we prove".

6.5.2 Sales promotion in a small business

Sales promotions — discounts, bundles, limited offers, loyalty rewards, referral incentives, samples — are the element of promotion most abused by small businesses, because they produce a visible short-term response. Used deliberately, a promotion moves a buyer at a specific moment (the end of a season, a launch, a reactivation), rewards a behavior you want more of (a referral, a review, a renewal) or lets a buyer try before committing (the diagnostic rung of the ladder). Used habitually, promotions reposition the business as the discounter and train customers to wait. The test: does the promotion serve the position and the ladder, or substitute for them?

6.6 People, Process and Physical Evidence in Services

6.6.1 People

In a service, the people are the product. The buyer cannot inspect the work in advance; they judge the person who answers the phone, the consultant who runs the diagnostic, the technician at the door. Marketing decisions about people include who is visible (the owner's face and voice, in a small business, are the brand — Chapter 7), how they are presented (names, credentials, photographs, a way of speaking that matches the position), how they behave at the moments of truth (Chapter 2 — the first call, the first visit, the complaint) and how they are trained to represent the position (a premium position delivered with a discount manner is a contradiction). In an established business, the people decision includes which staff meet which segment: the persona who expects the owner will not accept a junior on the first visit without being told why.

6.6.2 Process

Process is the sequence of steps through which the service is delivered — from first contact to completion and follow-up — and how it feels to the buyer at each step: how long, how many forms, how much uncertainty, how much of the buyer's effort. A visible, named process is a proof of competence (Chapter 5) and a reducer of anxiety (Chapter 4): "the ninety-day plan" with its weeks and deliverables sells better than "we will take care of it". Process is also where most friction (Chapter 2) hides — the quote that takes four days, the onboarding that requires six forms — and therefore where delivery becomes a marketing problem. Book J2 (Processes, SOPs and the Definition of Done) builds processes; here the point is that the buyer-facing process is part of the offer and should be designed and communicated as such.

6.6.3 Physical evidence

Physical evidence is everything tangible the buyer perceives as a signal of the intangible service: the premises, the vehicle, the uniform, the website, the proposal document, the invoice, the packaging, the report, the email signature, the tone of the automated messages. Each signals a quality level; the buyer averages them, weighting the weakest. Halden & Roy's website (Section 6.2.2) is physical evidence contradicting the position. The audit of physical evidence is the cheapest improvement in most small businesses: a consistent template for proposals and reports, a clean invoice, a website whose photographs match the price, an email signature with a real name and a phone number.

6.6.4 Measuring the service Ps

The three service Ps feel qualitative and are not. Each can carry two or three numbers, counted weekly or monthly like any operating figure (Chapter 11), so that a drift is seen before it becomes a complaint or a lost renewal. People: the share of calls answered live within the standard; the share of first visits made by the person the buyer was told to expect; the number of customer-facing staff who can state the position and the promise in one sentence when asked (a five-minute test at the monthly meeting). Process: the median time from inquiry to quote, from yes to start, from start to completion; the number of steps and forms the buyer must complete before the first deliverable; the share of engagements where the buyer had to ask "what happens next?" (a question the process should have answered before it was asked). Physical evidence: the age of the oldest customer-facing template still in circulation; the share of automated messages reviewed in the last year; the score of the quarterly evidence audit (Section 6.7) — yes, partly, no — per item. None of these requires a system beyond the contact base and a spreadsheet; all of them turn the service Ps from opinions into levers. The mix audit asks the questions; these numbers answer them between audits.

WHO DOES THIS Owning the mix Solo: the owner decides every P and delivers most of them; the risk is the blind spot — the owner stops seeing the outdated website or the confusing quote. Remedy: the audit grid once a quarter, and one outsider (a customer, a peer) walked through the offer twice a year. Small team: the owner decides product, price and place; whoever owns marketing owns promotion and physical evidence; whoever delivers owns process and people — and the three meet monthly to check coherence, because contradictions live at the boundaries between them. Established: product management or the owner owns product and price; sales and operations own place, people and process; marketing owns promotion and physical evidence and runs the coherence audit as the guardian of the position. In every case, price and product remain the owner's decisions, because they commit the economics of the business.

6.7 Coherence: When One P Contradicts Another

6.7.1 The audit grid

The mix audit reads each P against the position and against the other Ps. It takes an hour with the grid of Table 6.2, filled honestly — ideally with a customer's or a peer's eyes.

Table 6.2 — The seven-Ps audit grid

The grid produces two outputs: the list of contradictions (a P that says something different from the position or from another P) and the two fixes to make first — chosen by the size of the contradiction and the cost of the fix, which are usually inversely related: the largest contradictions are often in physical evidence and process, and the cheapest to correct.

METHOD The one-hour mix audit When to use it: quarterly; after any change of position, offer or price; when leads convert poorly. Participants: owner and, if possible, one customer or peer as the outsider. Deliverable: Table 6.2 filled, two fixes with owner and date. 0–10 min — Restate the position (Chapter 5) in one sentence at the top of the grid; everything is read against it. 10–40 min — Fill the facts column for each P as the buyer experiences it: visit your own website on a phone, read your last three proposals, call your own number, walk through the order process, look at the invoice. Write what you see, not what you intend. 40–50 min — Score coherence for each P (yes / partly / no) and name the contradictions between Ps. 50–60 min — Choose two fixes: the largest contradiction you can correct within a month, and the cheapest one you can correct this week. Owner and date for each; into the task list (Chapter 11). Pitfalls: filling the grid from intentions rather than observations; letting the owner score their own physical evidence without an outsider; fixing promotion first because it is the most visible.

WATCH OUT Promotion as a cover for an incoherent mix When leads do not convert, the reflex is to change the message or buy more reach. If the price is hidden, the process is heavy, the website contradicts the position or the person on the first call does not sound like the promise, more promotion produces more disappointed prospects — at a cost. The order of repair is the reverse of visibility: physical evidence and process first (cheap, fast, large effect), price presentation second, offer ladder third, promotion last. A business that fixes promotion first spends the most to change the least.

6.7.2 Reading contradictions

Typical contradictions and what they signal to the buyer:

• A premium price with bargain evidence (Section 6.2.2) makes the buyer think: "they charge a lot and cannot afford a decent website — or they are hiding something."
• A "simple" offer with a complex process makes the buyer think: "if buying is this hard, what will delivery be like?"
• Expert positioning with junior people makes the buyer think: "the expert sold it; someone else will do it."
• Local positioning in a national place makes the buyer think: "who are they really for?"
• A guarantee in the promotion and fine print in the terms makes the buyer think: "the guarantee is not real."
• Consent-based positioning with an aggressive promotion frequency makes the buyer think: "they say permission and behave like spam."

Each contradiction is a decision the business made without deciding. The audit makes the decision explicit and lets the owner choose — often by changing the position to match what the business actually does, which is a legitimate outcome, rather than pretending to a position the mix cannot keep.

Costly Mistakes

1. Optimizing one P in isolation. Cutting the price without touching the evidence, redesigning the website without stating the price, adding a guarantee without fixing the process. Consequence: the buyer reads the weakest element; the improvement is discounted or contradicted.
2. A ladder with missing rungs. Offering only the core purchase, with nothing to say yes to at the research or evaluation stage. Consequence: interested strangers with no safe first step leave; the funnel loses them between interest and decision.
3. Hiding the price. Making the buyer ask, in the hope that value can be explained first. Consequence: research-stage buyers assume "expensive" and go to the competitor who states a range; the meetings that do happen start with a price shock.
4. Habitual discounting. Promotions every month. Consequence: customers wait for the next one; the business is repositioned as the discount option without having decided to be; margin erodes permanently.
5. Ignoring physical evidence. Letting the website, proposals, invoices and vehicles age while positioning as premium or expert. Consequence: the cheapest contradiction to fix goes unfixed for years and undermines every other investment.
6. Prices that drip. Advertising a price and adding mandatory fees at checkout or in the proposal. Consequence: a false or misleading representation under the Competition Act and, in Québec, a Consumer Protection Act issue — plus the trust lost at the exact moment of purchase.

Key Takeaways

• The mix is the execution of the position: product, price, place, promotion — and, for services, people, process and physical evidence — each must confirm the place you chose in the buyer's mind.
• The mix is a coherence test, not seven optimizations: the buyer reads all Ps together and weights the weakest.
• Offers are bounded, tiered and packaged, and arranged on a ladder — free, diagnostic, entry, core, expansion — so that every stage of the journey has a safe yes.
• Price is a signal read before anything else: level relative to alternatives, structure, presentation and promotions are marketing decisions; anchor the price against the cost of the problem, state it early, and never let it drip.
• Place is access and carries the position: where the buyer finds and obtains your offer says what it is.
• Promotion cannot repair the other Ps and must tell the truth about them; sales promotions serve the position and the ladder or they substitute for them.
• People, process and physical evidence are where services are judged: the owner's face, a visible named process, and tangible signals audited for their weakest link.
• Audit quarterly with the grid, from observation not intention, with an outsider; fix physical evidence and process first, promotion last.

Your Action Plan

The action plan below presents the six steps that turn this chapter into work, in the order in which each step consumes the output of the one before. For each step, the table gives the deliverable that must exist at the end, the timing on a small business's calendar, and the verification criterion by which anyone — a co-founder, a mentor, the owner's own future self — can tell that the step was done. The criterion column is the one to read first: a step whose criterion you cannot picture is a step you do not yet understand.

In La Machine

The mix is largely decided outside any software; La Machine holds the offer, price and promotion elements that the platform sells and communicates. The fields and the absences described below were read in the product's code on September 19, 2026, in a copy whose most recent change is dated September 7, 2026.

• Packages (business/packages) hold your offers as bounded, priced items, each carrying a name, a type — meetings or minutes — a quantity or a mark for unlimited, a price, a cost and a currency, so that proposals, the shop and the sequences describe the same offer with the same price. The page has no field for a tier and none for terms, so each rung of the ladder is its own package, and the scope and the terms of a rung are written in the proposal and on the page that sells it.
• Promotions (business/promotions) hold time-bound promotions as a whole-number percentage off, from zero to one hundred, with a start date, an end date and a status a person sets by hand. There is no promotion code, no fixed amount, no usage limit, no order minimum, no free month and no credit, and the percentage is displayed to the client rather than applied at payment. Using this page rather than ad hoc discounts still keeps a record of how often you promote — the test of Section 6.5.2.
• Shop (business/shop, product editor at business/shop-product-edit) is the place for offers presented for purchase online — the entry and diagnostic rungs in particular — with the price displayed before checkout, as the law requires; the checkout itself runs in demonstration mode in the copy read, so a real payment is taken elsewhere until that changes.
• Subscriptions are handled with the shop and packages for recurring offers (→ see Book N8, Sales, Offers, Shop and Subscriptions; Book L5).
• Branding (business/branding) and templates (business/marketing/templates) are the physical evidence La Machine produces on your behalf — the look and tone of every automated message, proposal and page; an incoherent template contradicts the position in every email.

What La Machine does not do: it does not decide your price level or structure, design your process or judge your physical evidence; and it does not check price presentation against the Competition Act or the Consumer Protection Act — the "price seen first is the price paid" rule is yours to apply. Book N8 covers packages, promotions, shop and subscriptions page by page; Book A7 covers pricing methods; Book J2 covers process design.

Self-Check

1. Name the seven Ps and say which three were added for services, and why.
2. Why is the mix described as a coherence test rather than a set of optimizations?
3. What are the five rungs of the offer ladder, and which two are most often missing in small businesses?
4. Give two ways price acts as a signal, and state the rule this chapter proposes about the price the buyer sees first.
5. In what order should an incoherent mix be repaired, and why is promotion last?

Answers. (1) Product, price, place, promotion (McCarthy, 1960), plus people, process and physical evidence (Booms & Bitner, 1981) — because services are produced and consumed at once, delivered by people, and cannot be inspected before purchase, so the buyer judges the deliverers, the delivery and the tangible signals (Section 6.1.1). (2) Because the buyer reads all the Ps together and infers the position from them, weighting the weakest; one contradicting element makes the whole offer less credible (Sections 6.1.2 and 6.7). (3) Free and useful; low-commitment diagnostic; entry offer; core offer; expansion — rungs 2 and 3, the safe first purchases, are most often missing (Section 6.2.2). (4) Its level relative to alternatives signals quality; a hidden price signals "expensive and embarrassed" and sends research-stage buyers elsewhere; the rule: the price the buyer sees first is the price they pay, apart from taxes, and every "was" price is real and documented (Section 6.3). (5) Physical evidence and process first, price presentation second, offer ladder third, promotion last — because promotion over an incoherent mix produces more disappointed prospects at a cost, while evidence and process are cheap to fix with large effect (Section 6.7.1, Watch Out).

Chapter References

Competition Act, R.S.C. 1985, c. C-34, s. 74.01. Justice Laws Website. Retrieved September 7, 2026, from https://laws-lois.justice.gc.ca/eng/acts/C-34/

Consumer Protection Act, CQLR c P-40.1. LégisQuébec. Retrieved September 7, 2026, from https://www.legisquebec.gouv.qc.ca/en/document/cs/P-40.1

Booms, B. H., & Bitner, M. J. (1981). Marketing strategies and organization structures for service firms. In J. H. Donnelly & W. R. George (Eds.), Marketing of services (pp. 47–51). American Marketing Association.

Kotler, P., & Keller, K. L. (2016). Marketing management (15th ed.). Pearson.

Lauterborn, B. (1990, October 1). New marketing litany: Four Ps passé; C-words take over. Advertising Age, 61(41), 26.

McCarthy, E. J. (1960). Basic marketing: A managerial approach. Richard D. Irwin.

Chapter 7 — Brand Basics

Ask a customer why they chose you and they will rarely say "the logo". Ask them what they expected before the first call — reliability, expertise, a certain price, a way of being treated — and they are describing your brand, whether or not you ever designed one.

What You'll Be Able to Do

• Distinguish brand from logo and from visual identity, and explain what a brand actually is.
• State your brand promise and personality in terms the buyer would recognize.
• Explain why consistency is the cheapest brand strategy for a small business, and audit yours across eight touchpoints.
• Inventory your brand assets and know where each one lives.
• Use the owner's presence deliberately in a small business, without making the company dependent on it.
• Protect the name: know what a trademark is, how to search for conflicts, when to register, and what the language rules require in Québec.

Why It Matters

Brand is the expectation that precedes the sale and the memory that follows it. It is what lets a stranger trust you enough to call, what makes a referral stick, what allows a price above the alternatives, and what makes the tenth message recognizable rather than ignored. Small businesses tend to treat brand as decoration — a logo bought once — when it is in fact the accumulated result of every promise kept or broken. For the owner, brand is the asset that survives a change of channel, product or even personnel; for the team, brand rules are what make a hundred small decisions consistent without a meeting for each.

In This Chapter

• 7.1 Brand, Logo and Identity
• 7.2 Brand Promise and Personality
• 7.3 Consistency as the Cheapest Brand Strategy
• 7.4 Brand Assets and Where They Live
• 7.5 Brand in a Small Business: The Owner's Face
• 7.6 Protecting the Name

7.1 Brand, Logo and Identity

7.1.1 Three different things

The American Marketing Association defines a brand as "a name, term, design, symbol, or any other feature that identifies one seller's goods or service as distinct from those of other sellers" (American Marketing Association, n.d.). That is the legal and technical core — the identifier. But the identifier only matters because of what it calls up: Kevin Lane Keller's concept of customer-based brand equity locates the brand's value in the associations the identifier triggers in the customer's memory — awareness of the name, and the strength, favourability and uniqueness of what the customer associates with it (Keller, 1993). David Aaker described brand equity as a set of assets linked to the name and symbol — awareness, perceived quality, associations, loyalty — that add to or subtract from the value of the product or service (Aaker, 1991). In plain terms, then:

• The logo is a mark — a drawn identifier.
• The visual identity is the system of identifiers — name, logo, colors, typefaces, imagery, tone — that make the business recognizable (Wheeler, 2017); Book B3 builds it.
• The brand is what people think and feel when they encounter those identifiers: the expectation before purchase and the memory after it. You design the identity; the customer holds the brand.

DEFINITION Brand A brand is the set of expectations, associations and memories that a name, a symbol or any other identifier calls up in the mind of a customer or prospect, and that distinguishes one business's offer from the alternatives. It is built by every encounter — what was promised, what was delivered, how it felt — and it is held by the customer, not by the company. A business manages its brand by choosing a promise it can keep, expressing it consistently through its identity and its behavior, and protecting the identifiers legally. A brand exists whether or not it was designed; the only question is whether it is deliberate.

7.1.2 Why it matters more for small businesses than they think

Large companies build brands with advertising budgets; small businesses build them with behavior — and behavior is free. A small business's brand is built in the moments of truth of Chapter 2: the phone answered on the second ring by someone who knows your name, the quote that arrives when promised, the technician who cleans up, the invoice that matches the quote. Each moment either confirms the expectation the buyer arrived with or corrects it. This is why brand is a fundamentals subject and not a luxury: the small business that understands brand as kept promises has an advantage that no competitor can buy quickly.

Two mechanisms make brand valuable in measurable ways. Mental availability — the probability that your name comes to mind when the buyer has the need — is what turns awareness (Chapter 1) into consideration; Byron Sharp and his colleagues have shown across categories that brands grow largely by being thought of by more people in more buying situations, which depends on being recognized through consistent, distinctive cues (Sharp, 2010; Romaniuk & Sharp, 2016). Risk reduction — a known name lowers the perceived risk of purchase, which shortens the sales cycle and supports a price above the anonymous alternative (Chapter 4). Both mechanisms depend on the same thing: being recognized and remembered for something specific.

7.2 Brand Promise and Personality

7.2.1 The promise

The brand promise is the one thing the customer can count on every time — the expectation you commit to meeting. It is derived from the positioning (Chapter 5) but it is broader and more durable: the position says whom you are for and why you are better; the promise says what it will feel like to deal with you, every time, in every channel. "Dates you can plan around." "Nobody falls through." "We answer." A promise is useful only if it is specific (a competitor could not say the same thing with a straight face), kept (the process and the people of Chapter 6 deliver it), and expressed (the identity and the messaging say it without stating it).

7.2.2 The personality

The brand personality is the character the business would have if it were a person — the tone in which the promise is made: precise or warm, formal or direct, playful or sober, expert or approachable. Personality shows in word choice, in the images used, in how the phone is answered, in the humor or the absence of it. It matters because buyers choose businesses the way they choose people to deal with, and because a consistent personality makes the business recognizable across messages even before the logo is seen. A small business rarely needs more than three adjectives and a short list of what we sound like / what we never sound like.

7.2.3 From promise to standards: the behavior behind the brand

A promise that is not translated into standards — observable behaviors with a number attached — remains a slogan. The translation is the most valuable hour in brand work for a small business, because it is where the promise meets operations (Chapter 6's people and process) and where it becomes something the team can be trained on and measured against. The method is simple: for each moment of truth in the customer journey (Chapter 2), ask what would a business that keeps this promise do here, every time? and write the answer as a standard — specific, observable, with a threshold.

For a promise of reliability ("on time, or we call before"): the phone answered live within three rings during business hours, or a callback within thirty minutes; a confirmation the evening before every appointment; a fifteen-minute arrival window; a call the moment the window is at risk; the invoice matching the quote or an explanation before the invoice, never after. For a promise of expertise ("you will understand your options"): every proposal opening with the customer's situation in their words; at least two options presented, with the trade-offs; no jargon without a plain-language equivalent; a written summary within a day of every meeting. For a promise of care ("you are not a ticket number"): a named person on every message; the customer's history known before the call; a follow-up after delivery, by a person, within a week.

Standards do three things a promise cannot. They make the brand trainable: a new employee can be taught "three rings, thirty minutes, the evening before" in an afternoon; they cannot be taught "be reliable". They make the brand measurable: the share of calls answered within three rings and the share of appointments confirmed the evening before are counted weekly, like any other operating number (Chapter 11), and a falling number is a brand problem before it is a complaint. And they make the brand honest: a standard the business cannot meet — because of capacity, tools or skills — is discovered in the standards exercise, and the promise is adjusted before it is published, rather than broken in front of customers. A small business should hold five to ten standards, each with an owner and a weekly or monthly measure; more than that, and none is kept.

CASE IN POINT A brand promise discovered, not invented Fictional example; names and figures are invented. Brière Électrique, a residential and light-commercial electrical contractor with eleven employees, asked its last forty customers one question: "what would you tell a neighbour about us?" Thirty-one answers contained a version of the same sentence — "they show up when they say, and they call if they are late." Nobody mentioned price, and only two mentioned the quality of the work (which was assumed). The owner had been considering a new logo and a slogan about "quality craftsmanship". Instead, the company made punctuality its promise — "On time, or we call before" — and built its behavior around it: a confirmation text the evening before, a fifteen-minute arrival window, an automatic call if the window would be missed, and a standing rule that a missed window without a call earned the customer a $50 CAD credit. The logo was left alone. Two years later, Brière's reviews mentioned punctuality in 70% of cases, referral share rose from 35% to 55% of new jobs, and the company's price — 8% above the local average — stopped being questioned. The brand was found in the customers' words and kept by the operations, not designed.

7.3 Consistency as the Cheapest Brand Strategy

7.3.1 Why consistency beats creativity for small brands

A small business is seen rarely. A prospect encounters it a few times before deciding — a referral, a search result, a website, an email, a call — and each encounter is a chance to be recognized as the same business making the same promise, or to be forgotten. Recognition is cumulative only when the encounters match: the same name spelled the same way, the same mark, the same colors, the same tone, the same promise. A business that is recognizable after three encounters has, in effect, tripled its advertising budget without spending a dollar — and one that is not has paid for three separate introductions to the same person. Consistency is what turns those few encounters into one impression; creativity, in a small brand, usually fragments them. The same name spelled the same way, the same colors, the same voice, the same promise stated the same way, the same process described in the same words — this is what makes a small business look larger and more reliable than its size, at almost no cost.

KEY CONCEPT Consistency beats creativity for small brands A large brand can afford variety because it is seen thousands of times; a small brand is seen a handful of times and must make those encounters add up. Every deviation — a different logo version on the invoice, a different tone in the automated email, a different promise on the proposal — spends recognition instead of building it. The rule for a small business is therefore not "be creative" but "be the same, everywhere, for years": one name, one mark, one palette, one voice, one promise. Creativity goes into the work and into the content; the identity stays still so that the work can be recognized as yours.

7.3.2 The eight touchpoints

Figure 7.1 arranges the touchpoints where consistency is won or lost around the promise. Each spoke asks one question: does this touchpoint say the same thing as the promise?

Figure 7.1 — The brand consistency wheel: the brand promise at the centre, eight touchpoints around it — website and landing pages; email, SMS and automated messages; proposals, contracts and invoices; social posts and content; premises, vehicles, uniforms and packaging; people, phone and first visit; deliverables and reports; reviews, referrals and what others say. Original figure.

1. Website and landing pages are the first encounter for most prospects: the place where the promise is stated and the identity is set.
2. Email, SMS and automated messages are the most frequent touchpoint, and the most neglected: the confirmation, the reminder, the invoice notice, written once by a developer and never read again by the owner.
3. Proposals, contracts and invoices are the documents that carry money; a proposal in a different typeface from the website, or an invoice from a generic accounting template, tells the buyer that the brand stops where the money starts.
4. Social posts and content carry the voice and the imagery across platforms; the tone of a reply to a comment is brand.
5. Premises, vehicles, uniforms and packaging are the physical evidence of Chapter 6, seen by many who will never visit the website.
6. People, the phone and the first visit are how the promise is embodied at the moments of truth.
7. Deliverables and reports are what the customer keeps and shows to others; often they are the only thing a referred prospect sees.
8. Reviews, referrals and what others say are the brand as the market tells it; you do not write these, but you earn them and you answer them.

7.3.3 The consistency audit

Once a quarter, score each spoke against the promise: says the same thing · partly · contradicts. The weakest spoke is the one the buyer believes — a premium website with a discount invoice is a discount brand to the customer who pays. The audit takes thirty minutes with the eight items in hand (open the website on a phone, read the last automated email you sent, look at the last invoice, listen to your own voicemail greeting) and produces one fix per quarter; over two years the eight spokes agree. Chapter 6's mix audit and this one overlap by design: physical evidence is the brand made tangible.

7.3.4 Voice and tone in practice

Of the eight spokes, the one small businesses find hardest to keep consistent is voice — because it is produced daily, by several people, in a hurry, and because nobody wrote down what it is. Voice is the personality (Section 7.2.2) made audible: the vocabulary, the sentence length, the level of formality, the use of "you" and "we", the presence or absence of humor, how bad news is delivered, how a request is refused. It shows most in the places least designed — the automated confirmation, the reply to a complaint, the note on the invoice, the answer to a review.

A one-page voice guide fixes most of the inconsistency. It holds: the three personality adjectives with one sentence each on what they mean in writing ("direct: we say the price and the date in the first paragraph, not the third"); a short list of words we use and words we avoid (the customer's vocabulary from the interviews of Chapter 4 on one side; jargon, superlatives and hedging on the other); three sample messages rewritten in the voice — a confirmation, a follow-up, a refusal — so that the team imitates examples rather than interprets adjectives; and the rules for hard moments: how we apologize (specifically, without excuses, with what we will do), how we say no (with a reason and an alternative), how we answer a negative review (thank, acknowledge, take it offline, never argue). Voice also distinguishes tone — the same voice adjusts its tone to the situation: warmer in a welcome, plainer in an invoice, calmer in a complaint — without changing character, exactly as a person does.

The test of a voice guide is the same as the test of the brand guide: a freelancer or a new employee given only the page writes an email a customer would recognize as coming from you. La Machine's templates (see In La Machine) are where the voice is written once and reused; the guide is what the templates are written from.

7.4 Brand Assets and Where They Live

7.4.1 The inventory

A brand asset is any element that makes the business recognizable and that would have to be recreated if lost. Romaniuk and Sharp use the term distinctive brand assets for the non-name elements — colors, shapes, characters, sounds, taglines, typefaces — that trigger the brand without the name being present, and show that their value depends on being both famous (known by the target) and unique (not shared with competitors) (Romaniuk & Sharp, 2016). For a small business, the inventory is short. Table 7.1 is the template: one row per asset, with what it is, where the master file lives, who may change it and its legal status. The last two columns are the ones most small businesses cannot fill, and that is the point of the exercise.

Table 7.1 — Brand asset inventory (template)

7.4.2 The brand guide

The brand guide is the document that records the assets and the rules — typically four to eight pages for a small business: promise and personality, name and logo usage, colors, typefaces, imagery rules, voice with examples, templates and where to find them, and who to ask. Its purpose is not aesthetics but delegation: a freelancer, a new employee or an agency can produce something on-brand without a meeting. Book B3 (Branding and Visual Identity) builds the identity and the full guide; the design system of La Machine itself is an example of the form (a single source of truth for colors, type, components and rules).

7.4.3 Images, rights and the library

Imagery is the asset most often mishandled in two opposite ways: too generic — stock photographs of smiling strangers that could belong to any business and that buyers have learned to ignore — and unlicensed — images found by a search and used without a license, which is copyright infringement whatever the intention, and a frequent source of demand letters to small businesses. Three rules fix both. Prefer your own: photographs of your premises, your team, your work, your customers' sites (with permission) — taken by a professional once a year, or by a staff member with a modern phone and ten minutes of guidance on light and framing — carry more trust than any stock image and cost less over time. License what you buy, and record the license: the source, the license type, the permitted uses (web, print, social, advertising), the date, stored beside the file; some free licenses prohibit commercial use or require attribution, and "royalty-free" is not "free". Keep one library, with approved images sorted by use (people, premises, work, products, abstract), the license recorded per file, and a rule that nothing enters the library without its rights. The library is a brand asset in the inventory of Table 7.1 — and one that grows every year at almost no cost, once the habit exists.

WATCH OUT Assets you do not own Three ownership failures recur in small businesses. The logo whose copyright belongs to the designer — in Canada, copyright in a commissioned work generally remains with its author unless assigned in writing; get the assignment signed (→ see Book H5, Intellectual Property). The domain name registered by the web agency in its own name, which becomes a hostage when the relationship ends. The social account created by an employee with a personal email, lost when they leave. The rule: every brand asset is registered, licensed or assigned in the business's name, with the credentials held by the owner — checked once a year with the inventory of Table 7.1.

7.5 Brand in a Small Business: The Owner's Face

7.5.1 The owner is the brand — at first

In a business of one to twenty people, the owner is the brand for most customers: they bought the owner's competence, presence and promise. This is an asset — trust attaches to people faster than to companies, and an owner who is visible (in content, at events, on the first call) shortens the sales cycle. It is also a risk: a brand that lives entirely in one person cannot be delegated, scaled or sold (→ see Book M1, Preparing Your Business for Sale, where "owner dependence" is one of the first factors a buyer discounts).

7.5.2 Using the owner's presence deliberately

Three practices let a small business use the owner's face without becoming its prisoner. Make the owner the voice of the promise, not the only keeper of it: the owner states the promise publicly; the process and the team keep it, visibly. Introduce the team early: names, faces and roles on the website and in the first conversation, so that trust starts transferring from the person to the company from the first encounter. Build the company's assets alongside the owner's: the named process, the cases, the reviews of the company, the templates — so that when the owner steps back, the brand remains. The test is simple: would a customer who has never met the owner recognize the promise and receive it?

7.5.3 Founder-led content without founder dependence

The most effective use of the owner's presence in the 2020s has been founder-led content: the owner writing, speaking or appearing on video about the problems the business solves, in their own voice, with their own opinions — the specific, first-hand content that Chapter 1 identified as the kind that still earns attention when generic content no longer does. It works because trust attaches to a person who shows their thinking, and because an owner who has run the business for fifteen years knows things no agency can invent. It also raises the dependence problem in an acute form: the audience follows the person.

Three rules keep founder-led content an asset of the company. Publish under the company's name and on channels the company owns: the newsletter, the website, the company's accounts — with the owner as author, not as the only brand; the list and the archive then belong to the business. Make the owner's opinions the company's positions: the method, the standards, the "how we work" that the content explains are written into the process and the brand guide, so that they survive the author. Bring the team into the content early: a technician explaining a job, an office manager describing the follow-up system, a second consultant with a different specialty — so that the audience learns that the company, not one person, holds the expertise. A buyer of the business (Book M1) will ask whether the audience would stay if the owner left; the honest answer is "partly" — and the three rules are what make that answer good enough.

WHO DOES THIS Owning the brand Solo: the owner is the brand and its guardian; the minimum is a two-page guide (promise, personality, logo files, colors, typeface, voice, templates) so that the first freelancer produces on-brand work, and a quarterly consistency audit. Small team: the owner holds the promise and approves changes to name, logo and voice; whoever owns marketing maintains the guide and the templates and runs the audit; everyone who touches a customer knows the promise in one sentence. Established: a marketing manager owns the guide, the asset library and the audit; a designer (internal or external) maintains the identity under Book B3's rules; the owner or general manager approves any change to the promise or the identifiers, because they commit the company for years. In every case, the brand promise is a decision of the owner, because it is a commitment about behavior, not about design.

7.6 Protecting the Name

7.6.1 What a trademark is, and what it is not

A trademark is a sign — a word, a design, a combination, and in Canada also a sound, a color, a shape or other non-traditional signs — used to distinguish the goods or services of one person from those of others. Registering a business name with a provincial or federal registry gives the right to use that name as a business; it does not give exclusive rights against others using a similar name for similar goods or services. A registered trademark does: in Canada, registration under the Trademarks Act, R.S.C. 1985, c. T-13 gives the owner the exclusive right to use the mark across Canada in association with the registered goods and services, for a renewable term (ten years per term since the 2019 amendments to the Act), and the right to stop confusingly similar marks (Canadian Intellectual Property Office, n.d.). Unregistered marks enjoy some protection through use (the common-law action for "passing off" and, in Québec, civil remedies), but proving it is slower and costlier than showing a registration certificate.

7.6.2 Before you fall in love with a name

Three checks belong at the very beginning — before the logo, the domain and the vehicles are paid for. Search the Canadian Trademarks Database (free, on the Canadian Intellectual Property Office's site; Canadian Intellectual Property Office, 2026) for identical and similar marks in your goods and services; a strong conflict found later costs a rebrand. Search the business registries — the Registraire des entreprises du Québec and the federal Corporations Canada database — and the domain and social handles you would need. Judge distinctiveness: a descriptive name ("Montréal Plumbing Services") is hard to register and easy for competitors to approach; a distinctive one (a coined word, an unexpected combination) is registrable and defensible. The compromise most small businesses reach is a distinctive name plus a descriptive line ("Brière — electrical contractors"): the name is protected, the description does the explaining.

7.6.3 When to register

Registration costs money and time (a government fee per class of goods or services, plus professional fees if you use a trademark agent; the process typically takes well over a year in Canada as of 2026 — verify current fees and timelines with CIPO). It is worth doing when the name has become an asset: when customers ask for you by name, when you sell beyond your immediate region, when you plan to license, franchise or sell the business, or when a competitor's name starts to resemble yours. A trademark agent is worth consulting for the application; the search you can do yourself first (→ see Book H5, Intellectual Property, for the full treatment, including copyright in your content and designs).

QUÉBEC & CANADA NOTE Trademarks and the language of commerce (as of September 2026) Two regimes meet on a Québec storefront. Trademarks are federal: the Trademarks Act, R.S.C. 1985, c. T-13 and the Canadian Intellectual Property Office govern registration; a mark registered in Canada protects the owner in every province. Language is provincial: the Charter of the French language, CQLR c C-11 requires that public signs, posters and commercial advertising be in French, and it regulates how a trademark in another language may appear. Since the amendments brought by Bill 96 and their regulations, in force for these rules on June 1, 2025, a recognized trademark in a language other than French may appear on public signage and in advertising, but French must be markedly predominant in the overall visual field, and generic or descriptive terms included in the mark must appear in French; on products, generic or descriptive terms in a non-French trademark must also appear in French on the product or its packaging. The Office québécois de la langue française (OQLF) administers the Charter and publishes guidance and examples. → see Book H4 for the full rules; → see Book H5 for trademark strategy. Practical rule: design your signage and packaging with French markedly predominant from the start — retrofitting is far more expensive than designing for it.

7.6.4 Beyond the name: copyright in what you publish

The name is protected by trademark; everything else the business creates — website copy, guides, photographs, videos, designs, the brand guide itself — is protected by copyright, which in Canada arises automatically on creation under the Copyright Act, R.S.C. 1985, c. C-42 and belongs, in general, to the author or, for works created by employees in the course of employment, to the employer. Two practical consequences for a small business. First, work you commission from freelancers and agencies is not yours unless it is assigned to you in writing — the logo (Section 7.4.2), but also the website copy, the photographs, the video; the assignment is a sentence in the contract, and its absence is discovered at the worst moment. Second, your content is an asset others may copy — a competitor lifting your guide, a marketplace reusing your photographs; a copyright notice on the website and the guides (© year, name) costs nothing and states the obvious, and a dated record of publication supports any later claim. Registration of copyright is optional in Canada and useful mainly as evidence; a trademark agent or a lawyer can advise when the content is valuable enough to defend. → see Book H5 for the full treatment of intellectual property, including the terms to put in every creative contract.

WHAT CHANGED Non-French trademarks on signage and products (June 2025) Until 2025, a registered or recognized trademark in another language could appear on Québec signage with a "sufficient presence" of French. Since June 1, 2025, French must be markedly predominant, and descriptive or generic elements within non-French trademarks must be translated on signage and on products (Office québécois de la langue française, 2026). Businesses with existing signage were given transition periods that ended in 2025; new signage must comply from the start. Check the OQLF's current guidance before ordering signs or packaging.

Costly Mistakes

1. Treating brand as a logo. Buying a mark and a slogan while the phone goes unanswered and the invoices arrive late. Consequence: a brand of unkept promises with a nice logo; money spent on the part of brand that matters least.
2. Inconsistency by neglect. Three logo versions in circulation, an automated email written in 2019, an invoice from a generic template, a voicemail greeting recorded on a bad day. Consequence: the few encounters a prospect has with you do not add up; the business looks smaller and less reliable than it is.
3. Assets the business does not own. The logo's copyright with the designer, the domain in the agency's name, the social account on an employee's personal email. Consequence: a hostage situation at the worst moment — a dispute, a departure, a sale of the business.
4. A brand that is only the owner. Every customer bought the owner personally; nothing transfers to the company. Consequence: the business cannot delegate sales, cannot grow beyond the owner's hours, and is discounted heavily by any buyer.
5. A name chosen without a search. Vehicles, signs and a domain paid for before the trademark database and the registries were checked. Consequence: a demand letter and a forced rebrand — typically $10,000–$50,000 CAD for a small business, plus the recognition lost.
6. Signage designed without the language rules. Non-French marks displayed without markedly predominant French, or descriptive terms untranslated. Consequence: OQLF complaints, orders to modify, and signage paid for twice.

Key Takeaways

• The logo is a mark; the identity is the system of marks; the brand is what the customer thinks and feels when they see them. You design the identity; the customer holds the brand.
• A small business builds its brand by behavior, in the moments of truth — which is why brand is a fundamentals subject, not a luxury.
• The brand promise is the one thing the customer can count on every time; it is derived from the position, found in customers' own words, and kept by process and people. Personality is the tone in which it is made — three adjectives suffice.
• Consistency beats creativity for small brands: one name, one mark, one palette, one voice, one promise, everywhere, for years. Creativity goes into the work and the content.
• Eight touchpoints carry the brand; the weakest is the one the buyer believes. Audit quarterly, fix one spoke per quarter.
• Inventory the brand assets — name, logo, colors, typefaces, promise line, imagery, voice, templates, domains and handles — and make sure each is owned by the business, in writing, with the credentials in the owner's hands.
• The owner is the brand at first: use that presence deliberately while building the company's assets — process, cases, reviews, team — so the brand survives the owner stepping back.
• A registered business name is not a trademark. Search before you commit; register when the name becomes an asset; design Québec signage and packaging with French markedly predominant from the start.

Your Action Plan

The action plan below presents the seven steps that turn this chapter into work, in the order in which each step consumes the output of the one before. For each step, the table gives the deliverable that must exist at the end, the timing on a small business's calendar, and the verification criterion by which anyone — a co-founder, a mentor, the owner's own future self — can tell that the step was done. The criterion column is the one to read first: a step whose criterion you cannot picture is a step you do not yet understand.

In La Machine

Brand assets and rules live in your brand guide and asset library; La Machine applies them to what it produces.

• Branding (business/branding) holds the visual identity the platform uses on your behalf — logo, colors, typefaces, image charter — so that every page, email, proposal and deck La Machine produces carries the same identity. It is the single place to update when the identity changes (→ see Book B3; Book N6).
• Templates (business/marketing/templates) and CTAs (business/marketing/ctas) carry the voice: the automated messages of spoke 2 in Figure 7.1 are written once here, in the personality you chose, and reused everywhere.
• Gallery (business/marketing/gallery) and the social visual studio (business/social/visual-studio, business/social/montages) hold approved imagery and produce on-brand visuals for posts — spoke 4 — from the branding settings rather than from scratch.
• Books, presentations and videos (business/assets/*) can be white-labeled with a partner's logo and colors through the deck configuration; the content and the license line stay intact (→ see Book N10).
• Website (business/website, business/website/builder) applies the branding to the pages that make up spoke 1.

What La Machine does not do: it does not define your promise or personality, does not hold your copyright assignments, licenses or registry documents (keep those with the inventory of Table 7.1), and does not search or register trademarks — the Canadian Trademarks Database and a trademark agent do that. It applies the identity you give it, consistently, which is the part of brand a small business most often fails to do by hand.

Self-Check

1. What is the difference between a logo, a visual identity and a brand?
2. Why does consistency matter more than creativity for a small brand?
3. Name the eight touchpoints of the consistency wheel, and state which one the buyer believes.
4. Give three brand assets a small business often fails to own, and the rule that prevents it.
5. What does registering a business name give you, what does a registered trademark add, and what does the Charter of the French language require of a non-French trademark on Québec signage since June 2025?

Answers. (1) A logo is a drawn identifier; the visual identity is the system of identifiers (name, logo, colors, typefaces, imagery, tone); the brand is the set of expectations, associations and memories those identifiers call up in the customer's mind — designed by the company, held by the customer (Section 7.1.1). (2) Because a small brand is seen only a handful of times before a decision; consistency makes those encounters add up into one recognizable impression, while variety fragments them (Section 7.3.1). (3) Website and landing pages; email, SMS and automated messages; proposals, contracts and invoices; social posts and content; premises, vehicles, uniforms and packaging; people, phone and first visit; deliverables and reports; reviews, referrals and what others say — the buyer believes the weakest one (Section 7.3.2). (4) The logo's copyright (left with the designer), the domain name (registered by the agency), the social account (on an employee's personal email); the rule: every asset registered, licensed or assigned in the business's name, with credentials held by the owner, checked yearly (Section 7.4.2). (5) A registered business name gives the right to use it as a business, not exclusivity; a registered trademark gives exclusive rights across Canada in the registered goods and services for renewable ten-year terms and the right to stop confusingly similar marks; since June 1, 2025, a non-French recognized trademark may appear on Québec signage only with French markedly predominant and with generic or descriptive terms in the mark shown in French (Section 7.6).

Chapter References

Charter of the French language, CQLR c C-11. LégisQuébec. Retrieved September 7, 2026, from https://www.legisquebec.gouv.qc.ca/en/document/cs/C-11

Trademarks Act, R.S.C. 1985, c. T-13. Justice Laws Website. Retrieved September 7, 2026, from https://laws-lois.justice.gc.ca/eng/acts/T-13/

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

American Marketing Association. (n.d.). Definitions of marketing. Retrieved September 7, 2026, from https://www.ama.org/the-definition-of-marketing-what-is-marketing/

Canadian Intellectual Property Office. (n.d.). Trademarks. Government of Canada. Retrieved September 7, 2026, from https://ised-isde.canada.ca/site/canadian-intellectual-property-office/en/trademarks

Canadian Intellectual Property Office. (2026). Canadian Trademarks Database. Government of Canada. Retrieved September 7, 2026, from https://ised-isde.canada.ca/cipo/trademark-search/

Keller, K. L. (1993). Conceptualizing, measuring, and managing customer-based brand equity. Journal of Marketing, 57(1), 1–22.

Office québécois de la langue française. (2026). Affichage public et publicité commerciale. Gouvernement du Québec. Retrieved September 7, 2026, from https://www.oqlf.gouv.qc.ca/

Romaniuk, J., & Sharp, B. (2016). How brands grow: Part 2. Oxford University Press.

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

Wheeler, A. (2017). Designing brand identity: An essential guide for the whole branding team (5th ed.). Wiley.

Chapter 8 — Channels Overview

Every week someone tells you about a channel you are not using. This chapter is the framework for deciding, once, which three you will use well — and for saying no to the other twenty-seven with a clear conscience.

What You'll Be Able to Do

• Define a channel as a place where attention and access live, and classify any channel as owned, earned or paid — direct or through partners.
• Describe the main channels available to a small business in 2026, what each is for, what it costs and how long it takes.
• Score channels on reach to your segments, cost, fit with your offer, control and time to results.
• Choose a starting set of three channels, sequenced, with a budget of money or hours and a measure for each.
• Run a channel test with a hypothesis, a metric, a sample and a stop rule — and know when to stop, scale or switch.
• Explain why owned assets come first, and what "renting your audience" costs.

Why It Matters

Channels are where the marketing budget actually goes — in dollars and, for most small businesses, in the owner's hours. They are also where the noise is loudest: every platform, agency and tool sells its channel as the answer. Without a framework, the owner either freezes or follows the most recent voice, and the business ends up present everywhere and effective nowhere. With one, channel decisions follow the persona (where they look), the position (what to say), the ladder (what to offer) and the constraints (what you can sustain) — and each channel is a test with a number, not an act of faith. For the team, the framework ends the recurring debate; for the owner, it protects the two scarcest resources, time and attention.

In This Chapter

• 8.1 A Channel Is Where Attention and Access Live
• 8.2 Owned Channels
• 8.3 Earned Channels
• 8.4 Paid Channels
• 8.5 Direct and Indirect: Partners as Channels
• 8.6 Choosing Three Channels, Not Thirty
• 8.7 Sequencing and Testing Channels

8.1 A Channel Is Where Attention and Access Live

8.1.1 Definition

DEFINITION Channel A marketing channel is a place — physical or digital — where the attention of your segment can be reached and where access to your offer can be given: a search engine's results page, an inbox, a social platform, an association's event, a partner's customer list, a shelf, a website. A channel is defined by who is there (which personas, at which stage of the journey), what it costs to be present (money, hours, or the work of being worth mentioning), who controls it (you, a third party, a platform) and how fast it works. Distribution channels — how the buyer obtains the product — overlap with marketing channels and are treated in Chapter 6 (place) and Collection F; this chapter treats the channels through which buyers learn about and are moved toward the offer.

Two classifications from Chapter 2 organize the whole landscape. Owned, earned and paid describes control and cost structure (Corcoran, 2009; Chaffey & Ellis-Chadwick, 2022). Direct and indirect describes whether you reach the buyer yourself or through an intermediary — a distributor, a reseller, a referral partner, a marketplace, an affiliate. Figure 8.1 places the common channels on that map.

Figure 8.1 — The channel map: owned, earned and paid channels, with the overlaps where most effective programs live; earned and paid channels send people to owned assets; partners are an indirect route that cuts across all three. Original figure.

8.1.2 The landscape in 2026

Weinberg and Mares cataloged nineteen "traction channels" — from search advertising and content marketing to trade shows, speaking engagements and existing platforms — and observed that most companies find one or two that work far better than the rest, usually not the ones they expected (Weinberg & Mares, 2015). The list has shifted since — AI assistants now answer questions that once led to a search result, short video has become a default format on most platforms, and messaging applications carry business conversations — but the structure holds: many channels exist, a business can work only a few, and the right few depend on the persona and the offer. Table 8.1 lists the channels most small businesses consider: for each, its class, what it is best for, its cost structure and time to results stated together in one column, the degree of control it leaves you, and the book that covers it. The detailed methods live in Collections C (digital and organic), D (influence, events and advertising) and F (partners).

Table 8.1 — The channel landscape for a small business (as of September 2026)

WHAT CHANGED Where research happens (2024–2026) Two shifts alter the map without changing its logic. AI assistants increasingly answer the research-stage questions that once produced a click to a website, so being cited by an assistant — which rewards specific, structured, first-hand content — matters alongside ranking in a search engine (→ see Book C5). Short video has become a default format across social platforms and search, so a small business's content plan usually needs a video component even in B2B (→ see Book C6). Neither shift changes the rule of this chapter: follow the persona, own the destination, choose three.

8.2 Owned Channels

Owned channels are the ones you control and that compound: the website and its landing pages, the email and SMS list, the content library, the customer base, and the physical presence — premises, vehicles, packaging. They cost mostly fixed effort to build and almost nothing per use, they cannot be taken away by a platform's policy change, and every other channel's job is to bring people to them.

Three owned channels deserve first investment in almost every small business. The website is where the position is stated, the proof displayed, the price shown and the lead captured — the conversion point of the whole system (→ see Book C2). The email list, built with consent, is the only channel through which a business can reach its audience a hundred times a year at a marginal cost near zero; it is the backbone of nurture, retention and reactivation (→ see Book C1). The content library — the guides, calculators, cases and answers that speak to each stage of the journey — feeds search, AI answers, social, email and sales conversations at once; it is the raw material of every earned channel (→ see Book B5). The customer base itself is a channel: existing customers are the audience for expansion, referrals and reviews (Chapter 2).

WATCH OUT Renting all your audience A business whose entire reach is a social following, a marketplace ranking or a paid campaign has an audience it rents. Platforms change algorithms, policies and prices without notice; accounts get suspended; marketplaces promote their own products. Businesses have watched the majority of their reach vanish in a week when a platform changed its distribution rules. The remedy is not to leave the platforms but to convert rented attention into owned assets continuously: every post, ad and listing leads somewhere you own — a page, a guide, a consent-based subscription — so that when the rent goes up or the door closes, the relationships remain. Measure the conversion: what share of this month's rented reach became owned contacts?

8.3 Earned Channels

Earned channels are the ones others give you: a referral, a review, a search ranking, a mention by an assistant, a press article, a share, a recommendation in a community. They cannot be bought directly, they carry the trust of a third party, and they are the strongest converters in most small businesses — a referred prospect arrives with trust already transferred (Chapter 2). They are also the slowest and least controllable: a ranking takes months, a review depends on a customer's mood, a mention depends on an editor.

Earned channels respond to two things. The first is being worth talking about — the promise kept (Chapter 7), the specific content (Chapter 1), the visible result. The second is the system of asking: the review requested at the moment of satisfaction, the referral made easy with a reason and a script, the content structured so search engines and assistants can find and cite it, the association approached with a talk its members want. Small businesses under-earn not because they are not worth recommending but because nobody asks, at the right moment, in a way that makes it easy. Books D1 (referrals and word of mouth), B10 (reviews and retention), C3 (search), C5 (AI answer engines) and D3 (public relations) each build one of these systems.

8.4 Paid Channels

Paid channels rent attention: search advertising (reaching people who are searching now), social and display advertising (reaching a defined audience whether or not they are searching), sponsored content and influencers (borrowing a niche's trust), events and trade shows (buying a place in front of a gathered audience), direct mail and print (reaching personas offline), and paid placements of many kinds. They are fast — results in days — and they stop the moment payment stops. Their economics are transparent when measured and ruinous when not: an auction-priced click that lands on a page that does not convert is money spent twice (Chapter 2).

Paid channels have three legitimate jobs in a small business. Acceleration — reaching a segment that owned and earned channels cannot yet reach, while those are built. Testing — learning quickly whether a message or an offer resonates, with a defined budget and a stop rule. Filling a known gap — a seasonal trough, a launch, a new territory. In every case, the paid channel sends people to an owned asset that captures consent, so that rented reach becomes an owned relationship (Section 8.2). Book D7 (Paid Advertising Fundamentals) and its neighbors cover the platforms; Chapter 10 and Book H4 cover the rules on claims, disclosure and contests.

Paid channels also produce the most numbers, and the numbers most easily misread. Platforms report what they can measure — impressions, clicks, "conversions" defined by the platform — and those figures flatter the channel: a click is not a lead, a form view is not a form, and a conversion attributed by a platform to its own advertisement may have been a customer who would have arrived anyway through a referral. The discipline is to read paid channels with your own numbers: leads created in your contact base with the paid source, customers won from those leads, and cost per customer computed from the invoice, not from the platform's dashboard. Two further habits protect the budget. Separate demand capture from demand generation in the reporting — search advertising on your own name or on a category the buyer was already searching captures existing demand and will always look efficient; social advertising to people who were not looking generates demand and will always look expensive per click while feeding the top of the funnel — judging both by the same cost-per-click hides what each is for. And track the rented-to-owned conversion (Section 8.2): of the people the paid channel brought, how many became consented contacts you can reach again without paying? A paid channel that produces customers and a growing owned list is an investment; one that produces customers and nothing else is a cost you will pay every month for as long as you want the customers.

DO THE MATH Channel economics: cost per lead, cost per customer, payback Illustrative figures. A firm compares three channels over a quarter. Search advertising: $3,600 CAD spent, 1,200 clicks at $3.00, 48 leads (4% of clicks), 6 customers (12.5% of leads) → cost per lead $75, cost per customer $600. Referral program: 40 hours of the owner's and the team's time at a loaded cost of $60/hour = $2,400, plus $1,500 in thank-you credits = $3,900; 26 referred leads, 11 customers (42%) → cost per lead $150, cost per customer $355. Organic social: 60 hours of content and posting = $3,600; 14 leads, 2 customers → cost per lead $257, cost per customer $1,800. With an average first-year gross margin of $4,000 per customer, search pays back in about two months, referrals in about one, and organic social in more than five. The decision is not "social does not work" but "at this stage, referrals and search earn the next dollar and hour; social's role is to support them, or it is paused." Cost per customer by channel — which requires a source on every lead (Chapter 2) — is the number that ends the channel debate.

8.5 Direct and Indirect: Partners as Channels

A partner is anyone whose customers, audience or trust you can borrow: a distributor or reseller who sells your product, a professional who recommends you (the accountant, the lawyer, the architect, the supplier's representative), an affiliate paid per result, a marketplace that lists you, a complementary business that co-markets with you, an association that endorses you. Indirect channels multiply reach and transfer trust — the persona already buys from or believes the partner — at the price of margin (commissions, discounts, fees), of control over the customer relationship, and of the effort to recruit, train and keep the partner engaged.

Partners belong in the channel decision for three reasons. They are often the fastest route to the decider: the accountant who advises fifty owner-managed firms is a channel to fifty owners at once (Chapter 4, the VITO rule). They are a segment to be marketed to, with their own persona, triggers and objections — a partner recommends you when doing so makes them look good and costs them nothing. And they are a durable asset once built, closer to owned than to paid in their economics. Collection F builds partner programs of every kind; Book D1 covers referral partners specifically. For this chapter, the rule is to score partners with the same scorecard as any channel (Section 8.6) — most small businesses will find one partner type in their top three.

8.6 Choosing Three Channels, Not Thirty

8.6.1 Why three

A channel done well requires consistency (presence week after week, because attention is won by repetition), competence (each platform and format has rules that take months to learn) and measurement (a source on every lead, a weekly count). A small business can sustain that for two or three channels; at five, each gets a third of what it needs and none works; at ten, the owner is a full-time content producer for platforms that produce nothing. Weinberg and Mares' finding — that one or two channels usually dominate — is the empirical version of the same rule (Weinberg & Mares, 2015).

KEY CONCEPT Three channels done well beat ten done badly Every channel has a threshold of effort below which it produces nothing: the newsletter sent quarterly, the social account posted to monthly, the search advertising run for two weeks, the association joined but never attended. Below the threshold, the channel costs money and hours and returns noise — and it is then judged "not working". Above it, results compound. Choose the smallest set of channels you can keep above their thresholds — usually three: one owned (which you always have: the website and the list), one earned (usually referrals or search), one paid or partner (to accelerate). Add a fourth only when the first three are running at their thresholds without the owner's weekends.

8.6.2 The scorecard

The channel scorecard turns the choice into an explainable decision. Score each candidate from 1 to 5 on five criteria; weight reach and fit most heavily; and read the total alongside the cost per customer from any channel you have already measured. The five criteria are each a question. Reach to the persona, weighted three, asks whether the decider of your primary ideal customer profile is there, at a moment that matters, as recorded in the persona's sources panel. Cost, weighted two, asks what money and hours it takes to reach the channel's threshold of effort, relative to your budget. Fit with the offer, weighted three, asks whether the channel suits the offer's price, complexity and cycle. Control, weighted one, asks who sets the rules and the price — you or a platform. Time to results, weighted one, asks whether the first customer arrives in weeks or in months. Table 8.2 shows the scorecard filled for the fictional business-to-business provider of this chapter: each row is one of those five criteria with its weight, each of the five columns that follow is a candidate channel with its score, and the total near the bottom ranks them before the last row states the decision.

Table 8.2 — The channel scorecard (illustrative scoring for a B2B professional-services firm)

The scorecard is not arithmetic for its own sake; it forces the five questions to be asked of every channel, and it records the reasoning so that the next "you should be on…" conversation has an answer.

METHOD The channel scorecard (two hours) When to use it: when building the first channel set; at the quarterly reset; when someone proposes a new channel. Inputs: the persona's sources panel (Chapter 4), the offer ladder (Chapter 6), the funnel counts by source (Chapter 2), the hours and dollars available per month (Chapter 1). Deliverable: Table 8.2 filled, three channels named with role, budget and measure. Step 1 — List candidates (20 min). From Table 8.1, keep every channel the persona named as a source, plus the ones you already use. Step 2 — Score reach and fit first (40 min). For each candidate: is the decider there, at which stage, with which question (reach); does the channel suit the price and complexity of the offer (fit)? Use evidence — interview notes, lead sources — not impressions. Step 3 — Score cost, control, time (30 min). Estimate the hours or dollars to reach the threshold of effort; note who controls the rules; estimate time to the first customer. Step 4 — Read the totals with the economics (15 min). Where you have cost per customer by channel (Chapter 2), let it override the score. Step 5 — Choose three (15 min). One owned, one earned or partner, one paid or accelerating; each with a role (stage, persona, offer), a monthly budget of hours or dollars, and a measure (leads and customers by source). Everything else is "support" (kept at minimal effort) or "reserve" (to test at a named date). Pitfalls: scoring reach by the channel's total audience instead of the persona's presence; scoring from enthusiasm; choosing three channels that all need the owner's evenings.

CASE IN POINT A B2B firm's first three channels Fictional example; names and figures are invented. Trémont Conseil, four people, advises owner-managed distributors on inventory and margin. It was present on four social platforms, ran an occasional advertisement, attended two trade shows a year and had never sent a newsletter. Leads: about six a quarter, source unknown. The scorecard, built from twelve buyer interviews, produced a different set. Channel 1 — referral partners: the distributors' accountants and two software vendors, each approached with a one-page description of the ICP and a "how to recognize a client who needs us" checklist; a quarterly breakfast for partners. Channel 2 — the newsletter: a monthly email to a consented list of 340 owners, each issue one specific case with numbers. Channel 3 — search advertising: on the twelve trigger phrases the interviews surfaced ("distributor margin erosion", "slow-moving inventory"), landing on a diagnostic offer. Social was cut to one platform, posting the newsletter's case. Two quarters later: 19 leads a quarter, sourced; cost per customer $410 from referrals, $780 from search; the newsletter produced three reactivations of past clients. The trade shows were kept for one year as "reserve" and dropped when no lead could be traced to them.

8.6.4 Channel roles by stage of the journey

Three channels are enough only if, between them, they cover the stages of the journey where the persona needs to meet you (Chapter 2). A set of three that all work the same stage — three ways of being found by people already searching, say — leaves the trigger stage and the post-purchase stage empty: nobody is being made aware, and nobody is being asked for the review. The scorecard's reach criterion should therefore be read by stage, and the chosen set checked against Table 8.3.

Table 8.3 — What each channel class does at each stage of the journey

Read against the table, most small-business channel sets show the same gap: strong in research and evaluation (search, the website), weak at the trigger stage (nobody is naming the problem before the buyer searches) and weakest after purchase (nobody asks). The remedy is rarely a fourth channel; it is usually to extend an existing owned channel to the missing stage — the newsletter that also names triggers for past contacts, the sequence that also asks for the review — or to give the partner channel its post-purchase role. A three-channel set that covers all five stages, even thinly, outperforms a five-channel set concentrated on two.

8.7 Sequencing and Testing Channels

8.7.1 Sequence: owned, then earned, then paid

The order of investment follows the logic of Chapter 2. First, the owned destination: a page that states the position and captures the lead with consent, and a way to follow up (a sequence). Without it, every other channel leaks. Second, the earned engine: the referral system and the review routine, because they cost hours rather than dollars, convert best and start the flywheel; content and search follow, because they compound slowly and need the destination to exist. Third, paid acceleration: once the page converts and the follow-up runs, advertising or a partner program brings a defined segment to a system that can receive it — and the cost per customer is knowable from the first week.

A business that starts with paid channels — the common reflex — pays to learn what the owned and earned steps would have taught it, at auction prices.

8.7.2 Test protocol

Every new channel is a test until it has produced customers at a known cost. A test has four parts (Chapter 2, Section 2.6.3): a hypothesis ("the accountants who serve our ICP will refer at least one client each in six months if given a checklist and a breakfast"), a metric (referred leads and customers, by partner), a sample or duration large enough to mean something (six months for a referral program; six to eight weeks for search advertising; three months for a newsletter's reactivation effect), and a stop rule — the date or budget at which you decide to scale (cost per customer below target: increase budget or hours), fix (leads arrive but do not convert: the destination or the follow-up is the problem, not the channel), or stop (no leads at threshold effort: the persona is not there).

8.7.3 Budgeting hours and dollars across three channels

A channel plan without a budget of hours is a plan the owner will abandon in the third week. Most small-business channels are paid for in time before they are paid for in money, and time is the constraint that bites first (Chapter 1). The budgeting exercise therefore starts from the hours actually available per week for marketing — honestly, after client work and administration — and allocates them, with the dollars, across the three channels and the owned destination, leaving a reserve for the test.

Consider a firm with six hours a week of marketing time (four from the owner, two from an assistant) and $1,500 CAD a month of budget. A workable allocation: owned destination and list — one hour a week to keep the website's proof and prices current and to process the weekly count, plus the email tool's subscription (about $60 a month at typical small-business pricing as of 2026, verified on the vendor's page); earned channel (referral partners) — two hours a week: one partner conversation, one thank-you, the quarterly breakfast prepared over time, plus $200 a month in credits and breakfasts; owned channel (newsletter) — one and a half hours a week for the monthly issue, written from a case the firm already has; paid channel (search advertising) — half an hour a week to review the campaign, plus $900 a month in media and, for the first quarter, a freelancer's setup at a one-time cost taken from the reserve; reserve — one hour a week and the remaining $340 a month for the test of the quarter and for the unplanned. The allocation is written into the task table of Chapter 11 as recurring tasks with durations, and the weekly count reports, next to leads and customers by source, hours actually spent by channel — because the channel that quietly consumes eight hours for two leads is the one the scorecard must see.

Two rules make the budget hold. Automation buys hours before it buys results: the review request, the acknowledgment, the newsletter's scheduling and the social posting of the newsletter's case are automated first (Chapter 13; Book B8), which is what makes six hours enough. The reserve is spent, not hoarded: it exists to run the quarter's test with a stop rule; a reserve that is never spent is a channel that was never tested. When the hours are exhausted before the plan, the plan shrinks to what the hours allow — never the reverse — because a channel below its threshold of effort (Section 8.6.1) is the most expensive channel of all.

WHO DOES THIS Channel ownership in the three situations Solo: the owner chooses the three channels and runs them with automation doing the repetition (sequences, scheduled posts, review requests); the realistic budget is four to six hours a week plus a small paid test. Small team: the owner chooses; whoever owns marketing runs the owned and earned channels and briefs a freelancer or an agency for the paid one — keeping the accounts, the data and the source tracking in the company's name (Chapter 12). Established: the marketing manager owns the scorecard and the tests; specialists or agencies run channels against a cost-per-customer target; the owner reviews the channel mix quarterly. In every case, the owner decides which three — because the choice commits the company's hours and money — and no channel runs without a measure.

QUÉBEC & CANADA NOTE What each channel obliges you to do (as of September 2026) Each channel carries its own rules, detailed in Chapter 10 and Collection H. Email and SMS: consent, identification and unsubscribe under CASL (s. 6, s. 10, s. 11). Telephone: the CRTC's Unsolicited Telecommunications Rules, the National Do Not Call List and internal do-not-call lists for telemarketing (Canadian Radio-television and Telecommunications Commission, n.d.). Advertising in any channel: no false or misleading representations, tested performance claims, disclosed sponsored content and influencer relationships under the Competition Act (s. 74.01) and the Competition Bureau's guidance. Direct mail: not covered by CASL, but the use of personal information — including purchased lists — falls under P-39.1 and PIPEDA. Door-to-door and off-premises selling to consumers in Québec: the Consumer Protection Act, CQLR c P-40.1 regulates itinerant merchants, including permit requirements and cancellation rights. All public advertising and commercial publications in Québec: French, under the Charter of the French language. Score a channel's cost to include the compliance work it requires.

Costly Mistakes

1. Being everywhere. Five or ten channels, each below its threshold of effort. Consequence: hours and dollars spent on noise; every channel judged "not working"; the owner's evenings consumed by platforms that produce nothing.
2. Choosing channels by fashion or by vendor. Adopting the platform everyone talks about, or the one the agency sells, instead of the one the persona named. Consequence: presence where the decider is not; results that never arrive; a new fashion next year.
3. Paid before owned. Buying clicks that land on a page that does not convert, with no follow-up sequence. Consequence: the click is bought twice; the business concludes that advertising is expensive when the destination was the problem.
4. Renting the whole audience. Building reach only on platforms and marketplaces, without converting it into consented contacts. Consequence: a policy or algorithm change removes most of the reach in a week, with nothing left.
5. No measure per channel. Running channels without a source on every lead. Consequence: cost per customer by channel unknown; budget follows the loudest voice; the channel debate never ends.
6. Stopping too early, or never. Judging a referral program after six weeks, or running a paid campaign for a year without a stop rule. Consequence: the slow compounding channels are abandoned before their threshold; the fast leaking ones are funded forever.

Key Takeaways

• A channel is where attention and access live; classify each as owned, earned or paid — and direct or through partners — and remember that earned and paid channels exist to bring people to owned assets.
• Owned first: website, consented list, content library and customer base compound and cannot be taken away. Earned channels carry third-party trust and respond to being worth talking about plus a system of asking. Paid channels accelerate, test and fill gaps — and stop when payment stops.
• Partners are channels — often the fastest route to the decider — and a segment to be marketed to in their own right.
• Three channels done well beat ten done badly: every channel has a threshold of effort below which it returns nothing; choose the smallest set you can keep above threshold — one owned, one earned or partner, one paid.
• Score channels on reach to the persona, cost, fit, control and time, weighting reach and fit; let measured cost per customer by channel override the score.
• Sequence owned → earned → paid; treat every new channel as a test with a hypothesis, a metric, a duration and a stop rule; scale, fix or stop on the number.
• Never rent your whole audience: convert rented attention into owned relationships continuously, and measure the conversion.
• Each channel carries rules — CASL, telemarketing rules, the Competition Act, privacy law, consumer protection, French — which are part of the channel's cost.

Your Action Plan

The action plan below presents the seven steps that turn this chapter into work, in the order in which each step consumes the output of the one before. For each step, the table gives the deliverable that must exist at the end, the timing on a small business's calendar, and the verification criterion by which anyone — a co-founder, a mentor, the owner's own future self — can tell that the step was done. The criterion column is the one to read first: a step whose criterion you cannot picture is a step you do not yet understand.

In La Machine

La Machine runs the owned channels and records what the others send.

• Website and builder (business/website, business/website/builder, preview at business/website/preview) and landing pages (business/marketing/landing-pages) are the owned destination of Section 8.7.1: position stated, price shown, lead captured with consent and source through forms (business/marketing/forms) and lead magnets (business/marketing/lead-magnets).
• Email queue, campaigns, sequences and templates (business/marketing/email-queue, campaigns, sequences, templates) run the email channel — the newsletter, the nurture and the follow-up that make every other channel convert — against the consent register (business/marketing/consent) and the do-not-reach list.
• Social (business/social: accounts, calendar, composer, montages, outbox, posts, visual studio, settings) runs the organic social channel: scheduling, publishing and on-brand visuals from one place, so that the channel stays above its threshold of effort without the owner's evenings.
• Powerdialer (business/marketing/powerdialer) is a placeholder page as of September 15, 2026: the product places no call, so permitted calls are made outside it, with the telephone entries of the do-not-reach list checked first (→ see Book N5).
• Responses, the log and sales records (business/marketing/forms/responses, business/marketing/log) give the counts by source that the scorecard and the test protocol need, since funnels keep no counts; CRM intake (crm/intake) receives the leads that paid and partner channels send.
• Partner programs and referral tracking live in the partner functions (→ see Book N9, Partner Programs in La Machine); SEO and analytics in the website module (→ see Book N7).

What La Machine does not do: it does not buy advertising — paid search and social campaigns are configured in the platforms' own tools and tracked here by source; it does not attend your association's breakfast or make the referral ask for you, though it schedules the task and sends the thank-you; and it does not choose your three channels — the scorecard does, and the owner decides.

Self-Check

1. What four things define a channel, and what are the two classifications this chapter uses?
2. Why should a small business invest in owned channels before earned and paid, and what job do paid channels legitimately do?
3. What is the "threshold of effort", and how does it justify the rule of three?
4. Name the five criteria of the channel scorecard and the two that carry the most weight.
5. What are the four parts of a channel test, and what are the three possible decisions at the stop rule?

Answers. (1) Who is there (personas, stage), what it costs, who controls it, how fast it works; classified as owned / earned / paid and as direct / indirect through partners (Section 8.1.1). (2) Because owned assets compound and cannot be taken away, and every other channel exists to bring people to them — a paid click landing on a page that does not convert is bought twice; paid channels accelerate, test and fill known gaps, always landing on owned assets that capture consent (Sections 8.2, 8.4, 8.7.1). (3) The level of consistent effort below which a channel produces nothing — a quarterly newsletter, a two-week ad campaign; a small business can keep only two or three channels above their thresholds, so three done well beat ten done badly (Section 8.6.1). (4) Reach to the persona, cost, fit with the offer, control, time to results — reach and fit weigh most (Section 8.6.2). (5) Hypothesis, metric, sample or duration, stop rule; at the stop rule: scale, fix (the destination or follow-up, not the channel) or stop (Section 8.7.2).

Chapter References

An Act to promote the efficiency and adaptability of the Canadian economy by regulating certain activities that discourage reliance on electronic means of carrying out commercial activities, S.C. 2010, c. 23 (Canada's Anti-Spam Legislation), ss. 6, 10, 11. Justice Laws Website. Retrieved September 7, 2026, from https://laws-lois.justice.gc.ca/eng/acts/E-1.6/

Act respecting the protection of personal information in the private sector, CQLR c P-39.1. LégisQuébec. Retrieved September 7, 2026, from https://www.legisquebec.gouv.qc.ca/en/document/cs/P-39.1

Charter of the French language, CQLR c C-11. LégisQuébec. Retrieved September 7, 2026, from https://www.legisquebec.gouv.qc.ca/en/document/cs/C-11

Competition Act, R.S.C. 1985, c. C-34, s. 74.01. Justice Laws Website. Retrieved September 7, 2026, from https://laws-lois.justice.gc.ca/eng/acts/C-34/

Consumer Protection Act, CQLR c P-40.1. LégisQuébec. Retrieved September 7, 2026, from https://www.legisquebec.gouv.qc.ca/en/document/cs/P-40.1

Personal Information Protection and Electronic Documents Act, S.C. 2000, c. 5. Justice Laws Website. Retrieved September 7, 2026, from https://laws-lois.justice.gc.ca/eng/acts/P-8.6/

Canadian Radio-television and Telecommunications Commission. (n.d.). National Do Not Call List. Retrieved September 7, 2026, from https://lnnte-dncl.gc.ca/

Chaffey, D., & Ellis-Chadwick, F. (2022). Digital marketing (8th ed.). Pearson.

Corcoran, S. (2009, December 16). Defining earned, owned and paid media. Forrester Research.

Weinberg, G., & Mares, J. (2015). Traction: How any startup can achieve explosive customer growth. Portfolio/Penguin.

Chapter 9 — Marketing and Sales Alignment

A lead arrived on Friday at 4:50 p.m. Marketing counted it as a success. Sales saw it Tuesday morning, called once, left no message and marked it "no answer". Both departments were right about their own numbers — and the company lost a customer nobody will ever count.

What You'll Be Able to Do

• Draw the line between marketing and sales in your business, and say what each function owes the other.
• Define lead, marketing-qualified lead, sales-qualified lead and opportunity by observable events, so that everyone classifies a contact the same way.
• Design the qualification and hand-off, with the context that travels with each lead.
• Explain why speed to lead matters, quantify what slow response costs you, and set a response standard you can keep.
• Write a marketing-sales service-level agreement with shared metrics and a review rhythm.
• Install the feedback loops — win/loss reasons, objections, content gaps — that make each function better at the other's job.

Why It Matters

Marketing and sales are two halves of one process: the conversion of a stranger into a customer. When they are aligned, leads are defined the same way, handed over fast with context, followed up systematically, and the reasons for wins and losses flow back into the message. When they are not, marketing generates leads sales ignores, sales complains about lead quality, both report success, and the leak sits invisibly between them — usually the largest leak in the funnel (Chapter 2). In a small business the two functions are often one person, which does not remove the problem: it hides it inside one head, where the marketing hat produces a lead that the sales hat forgets. For the owner, alignment is the cheapest revenue increase available; for the team, it is the end of the oldest argument in business.

In This Chapter

• 9.1 Where Marketing Ends and Sales Begins
• 9.2 Lead Definitions: Lead, MQL, SQL, Opportunity
• 9.3 Qualification and Hand-off
• 9.4 Speed to Lead
• 9.5 Shared Metrics and the Service-Level Agreement
• 9.6 Feedback Loops: Win/Loss, Objections, Content Gaps
• 9.7 One Team in a Small Business

9.1 Where Marketing Ends and Sales Begins

9.1.1 Two functions, one process

Kotler, Rackham and Krishnaswamy described the relationship between marketing and sales in most companies as a spectrum from undefined (the two grow up independently and collide) through defined (rules to prevent disputes) and aligned (clear boundaries, joint planning, shared language) to integrated (shared systems, metrics and rewards) — and observed that the friction between them, when present, costs the company far more than either function's inefficiency alone (Kotler, Rackham, & Krishnaswamy, 2006). Frank Cespedes made the complementary point that sales is where strategy meets the customer, and that a strategy sales cannot execute — because the leads, the message or the offer do not match what happens in the field — is not a strategy (Cespedes, 2014).

For a small business the spectrum collapses to a practical question: at what moment does a contact stop being marketing's responsibility and become sales'? Marketing's job ends when a person has been identified, is known to fit the ideal customer profile (Chapter 3), has shown intent, and has been handed over with the context that lets a conversation start well. Sales' job begins with that conversation and ends with a signed customer — after which both functions own the customer's next experiences (onboarding, reviews, referrals, expansion — Chapter 2's flywheel).

9.1.2 What each function owes the other

Marketing owes sales: leads that fit the profile, delivered fast, with source, consent status and context; a message in the market that matches what sales will say in the room; content that answers the objections sales hears; and a nurture path for the leads sales cannot work now.

Sales owes marketing: prompt and systematic follow-up of every lead handed over; a recorded outcome for each — accepted, recycled, won, lost — with the reason; the objections and questions heard in the field; and the buyers' own words, which are marketing's raw material (Chapter 4).

KEY CONCEPT Definitions before dashboards Every alignment failure traces to a missing definition. If "lead" means "anyone who filled a form" to marketing and "someone ready to buy" to sales, marketing will report 60 leads and sales will report 6 worth calling, and both will be right. Before any dashboard, both functions agree — in writing — on what a lead, a marketing-qualified lead, a sales-qualified lead and an opportunity are, each defined by an event anyone can observe. The definitions are the treaty; the dashboard is only its accounting.

9.2 Lead Definitions: Lead, MQL, SQL, Opportunity

9.2.1 The four stages, defined by events

A stage is a state a person is in, defined by an observable event (Chapter 2). The four stages between "stranger" and "customer" that most businesses need are:

DEFINITION Lead, marketing-qualified lead (MQL), sales-qualified lead (SQL), opportunity A lead is a person or organization that has identified itself to the business and signalled a possible need — by submitting a form, requesting a quote, downloading a document, replying to a message, calling, or signing up at an event. A lead has a source and a consent status. A marketing-qualified lead (MQL) is a lead that fits the ideal customer profile (Chapter 3) and has shown intent beyond the first identification — visiting the pricing page, requesting a quote or a demonstration, downloading a second document, replying to a nurture message; the criteria are written and applied by marketing, ideally automatically. A sales-qualified lead (SQL) is an MQL that sales has contacted and accepted as worth pursuing, having confirmed a real need, the involvement of a decider (Chapter 4), and a plausible timing. An opportunity is an SQL in an active buying process: a meeting held, a proposal or quote sent, a decision date agreed. Beyond opportunity lies the customer — signed and paid. Each transition is counted; each stage has an owner and a service level (Figure 9.1).

The vocabulary varies (some businesses say "prospect" for SQL, "deal" for opportunity); the discipline does not: a person cannot be in a stage without the event that defines it having happened, and two people looking at the same record put it in the same stage.

Figure 9.1 draws that lifecycle as a single line with five stages and the event that opens each of them. It is read left to right — lead, marketing-qualified lead, sales-qualified lead, opportunity, customer — with the owner of each stage named beneath it and the event that must have happened written above the transition into it. The figure exists to settle one argument: a stage is entered by an event, not by an opinion.

Figure 9.1 — The lead lifecycle: lead → MQL → SQL → opportunity → customer, with the defining event, the owner of the next step and an illustrative service level for each; disqualified or stalled leads recycle to marketing nurture with the reason recorded; shared metrics are reviewed by both functions. Original figure.

9.2.2 Writing your MQL criteria

The MQL is the hinge of the system, because it is where marketing hands work to sales. Its criteria combine fit (does the contact match the ICP — sector, size, geography, role?) and intent (has the contact done something that people who buy tend to do?). Fit criteria come straight from the ICP document (Chapter 3); intent criteria come from your own data: look at the last twenty customers and note which actions they took before buying — the pricing page, a quote request, a second download, a reply. Those actions are your intent signals. A simple points system — fit criteria worth more than intent signals, a threshold that triggers the hand-off — is enough for most small businesses; Book B8 (Marketing Automation and Funnels) builds the scoring in detail. The test of good MQL criteria is the acceptance rate: if sales accepts fewer than about two thirds of MQLs, the criteria are too loose; if it accepts nearly all and complains of too few, they are too tight.

9.2.3 Recycling, not discarding

A lead that does not become an SQL is not waste; it is a person who fits, or may fit later, and who has given consent to be contacted. "Not now", "no budget this year", "no answer after six attempts", "not the decider" are recycle reasons, each with a nurture path: a sequence for the "not now" that checks back at the date they named, a monthly newsletter for the rest, a re-qualification when the intent signals return. The recycle arrow in Figure 9.1 is where most small businesses leak leads they paid for; closing it is usually worth more than any new channel (Chapter 1, Do the Math).

9.2.4 Lead scoring in practice

The MQL criteria of Section 9.2.2 can be applied by a person reading each record, or by a score the system computes from the record's fields and behavior. Scoring is worth setting up when leads exceed what one person can read daily — roughly thirty a month — and it is worth keeping simple: a handful of fit points, a handful of intent points, a threshold, and a quarterly check of whether the threshold predicts acceptance. Table 9.1 is a starting model for a B2B service; the values are illustrative and should be replaced by your own, derived from the behavior of your last twenty customers before they bought.

Table 9.1 — A simple lead-scoring model (illustrative)

Three cautions. Fit outweighs intent: a lead with every intent signal but the wrong sector is a curious visitor, not a customer, and a model that lets intent alone cross the threshold floods sales with unfitting leads. The threshold is calibrated against acceptance: if sales accepts fewer than two thirds of leads at 60, raise the threshold or the fit weights; if it accepts nearly all and asks for more, lower it. Scores decay: an intent signal from eleven months ago is not intent; most systems reduce scores over time or reset them after a defined period, so that the MQL of today reflects the behavior of the last quarter. Book B8 (Marketing Automation and Funnels) builds and tunes the model; the point here is that the score is the MQL definition made automatic, and it is only as good as the definition it encodes.

9.3 Qualification and Hand-off

9.3.1 Qualification questions

Qualification is the conversation — or the form — that confirms fit and establishes need, decider and timing. The classic frameworks ask about budget, authority, need and timing; more recent ones emphasize the buyer's problem, its cost, the decision process and the consequence of not acting. For a small business, four questions asked early and recorded are enough: What is happening that made you reach out now? (the trigger, Chapter 4). Who else is involved in this decision? (the decision unit). What have you tried or considered? (the alternatives, Chapter 5). When do you need this solved, and what happens if it is not? (timing and cost of inaction). The answers qualify the lead and feed the persona's evidence file (Chapter 4) — qualification is research.

9.3.2 The hand-off package

A hand-off without context forces the salesperson to start from zero and the buyer to repeat themselves — friction on both sides (Chapter 2). The hand-off package travels with the lead, automatically where possible: source (channel, campaign, page); consent status; fit (ICP match, segment); intent signals (what they did, when); the persona they likely match; what they asked or downloaded; the qualification answers if a form or a first call collected them; and the promise made (a callback within the hour, a specific document). The package is what the CRM record should show on one screen the moment the salesperson opens it.

9.3.3 Who takes it, and how fast

The hand-off names an owner — a person, not a team — and a service level: the time within which the first contact attempt is made, the number of attempts over how many days before the lead is recycled, and the channels used (call, then email, then text, where consent allows). Round-robin assignment, territory rules or product rules decide the owner; what matters is that every lead has one name on it within minutes of arriving, and that the system shows which leads have passed their service level without a contact attempt. Section 9.4 explains why the first hour matters so much.

CASE IN POINT The lead nobody called back Fictional example; names and figures are invented. Solstice HVAC, a heating and air-conditioning contractor with twenty-two employees, spent about $4,000 CAD a month on search advertising and received around 90 web inquiries a month. Its salespeople worked from an inbox shared with dispatch. When the owner finally measured (Chapter 2), she found that the median time to first contact was 26 hours, that 31% of inquiries received no contact attempt at all, and that no inquiry received more than two attempts. The inquiries that were contacted within an hour converted to appointments at 38%; those contacted after a day, at 9%. Solstice changed nothing about its advertising. It routed inquiries to a named salesperson by rotation, set a 15-minute callback standard during business hours and an automatic text plus next-morning call outside them, and required six attempts over ten days before an inquiry could be marked "no answer". The following quarter, the same 90 inquiries produced 34 appointments instead of 17, and 21 installations instead of 11 — at an average of $9,800 CAD, about $98,000 of additional quarterly revenue from a routing rule and a standard.

9.4 Speed to Lead

9.4.1 The evidence

Speed to lead — the time between a lead's arrival and the first contact attempt — is one of the best-documented predictors of conversion. A study published in Harvard Business Review audited the response of more than two thousand US companies to web-generated leads and found that firms which attempted to contact a lead within an hour of receiving the inquiry were nearly seven times as likely to qualify the lead (have a meaningful conversation with a decision maker) as those that tried even an hour later — and more than sixty times as likely as companies that waited twenty-four hours or longer; yet the average first response took more than forty hours, and nearly a quarter of companies took more than twenty-four hours or never responded (Oldroyd, McElheran, & Elkington, 2011). The mechanism is not mysterious: the buyer is at their desk, the problem is on their mind, and — as Chapter 2 showed — they have probably contacted two or three other providers. The first competent response often wins by default.

9.4.2 A standard you can keep

The right standard depends on your market and your capacity, but the shape is the same everywhere: minutes, not hours, during business hours; an automatic acknowledgment plus a first-thing-next-morning call outside them; and a defined sequence of attempts across channels before a lead is recycled. For most small businesses: first attempt within 15 to 60 minutes during hours; an immediate automatic reply that confirms receipt, sets the expectation ("a person will call you within the hour") and offers a booking link; six attempts over ten business days mixing call, email and — with consent — text; and a recorded outcome for each attempt. The standard is written into the service-level agreement (Section 9.5) and enforced by the system, which shows every lead past its deadline.

9.4.3 The acknowledgment and the sequence, word by word

The automatic acknowledgment is the most-read message a business sends, and the least designed. It has four jobs, in order: confirm that the request arrived and repeat what the person asked for, in their words if a form captured them ("You asked about scheduling for a 30-person shop"); set the expectation precisely ("Marie will call you before 11:30 this morning" during hours; "we will call you tomorrow between 8:00 and 9:00" outside them) — a specific promise, kept, is the first proof of the brand (Chapter 7); offer control — a booking link to choose a time, and a direct number and name in case they prefer to call; and give one useful thing while they wait — the price range page, the case that matches their situation, the two questions the call will cover, so that the conversation starts ahead. The acknowledgment identifies the sender and carries an unsubscribe mechanism like any commercial electronic message (Chapter 10), even though the person asked for it.

The attempt sequence over ten business days alternates channels and adds value at each step rather than repeating "just following up". A typical shape for a lead who inquired: day 1 — call within the hour; if no answer, a voicemail with the name, the reason and a time you will try again, followed within five minutes by a short email or text (with consent) that says the same and offers the booking link; day 2 — a second call at a different time of day; day 3 — an email that adds something: the case that matches their trigger, or the answer to the question most people in their situation ask; day 5 — a third call, and a text if consented; day 8 — an email that names the alternatives honestly ("if you have solved this another way, tell us and we will stop; if the timing is wrong, tell us when to check back"); day 10 — a final call and a closing email that leaves the door open and states that you will not keep calling — the lead is then recycled to the nurture path with the reason "no answer after six attempts", not deleted. Every attempt is logged with its outcome, so that the acceptance and speed metrics (Section 9.5.2) are read from data rather than remembered. A sequence of this shape is not harassment; it is the persistence a person who inquired expects and, in most markets, the difference between the third of leads that are reached and the third that are lost.

DO THE MATH Speed-to-lead economics Illustrative figures, consistent with the direction of the research cited. A firm receives 60 inquiries a month. Contacted within an hour, inquiries convert to qualified conversations at 35%; contacted after a day, at 8%. Today, 20% are contacted within the hour and 80% after a day: 60 × (0.20 × 0.35 + 0.80 × 0.08) = 4.2 + 3.8 = 8 qualified conversations. With a routing rule and a standard that brings 80% under the hour: 60 × (0.80 × 0.35 + 0.20 × 0.08) = 16.8 + 1.0 = 17.8 qualified conversations — more than double, from the same inquiries and the same advertising. At a 30% close rate and $6,000 CAD of first-year value, that is roughly $17,600 of additional monthly revenue, or about $211,000 a year. The cost: a routing rule, an automatic acknowledgment, and a named owner per lead — none of which requires a new channel.

9.5 Shared Metrics and the Service-Level Agreement

9.5.1 The agreement

A service-level agreement (SLA) between marketing and sales is a one-page treaty stating what each delivers to the other, in numbers, and how both will know. It is written together, signed by both (or by the owner, in a business where both hats sit on one head), and reviewed on a rhythm. Table 9.2 is a template.

Table 9.2 — The marketing-sales service-level agreement (template with illustrative values)

9.5.2 The shared metrics

Six numbers, read together every two weeks, make the agreement live. Lead-to-MQL rate (are we attracting the right people?). MQL acceptance rate (do the MQL criteria produce leads sales agrees are worth working? — target roughly two thirds or more). Speed to first contact (median and share within standard). SQL-to-opportunity rate (do accepted leads reach a real buying process?). Win rate (opportunity to customer). Cost per customer by source (which channels produce customers, not clicks — Chapter 8). A seventh, over time: cycle length from lead to customer, by source. Each number has a shared owner and a target; when one falls, the review asks which function's commitment slipped — without blame, because the agreement made the commitments explicit.

THE RULE Every lead has one owner and a deadline, and every outcome is recorded A rule of practice, non-negotiable in an aligned system: no lead sits without a named owner and a service-level deadline, and no lead leaves the pipeline without a recorded outcome and reason. Unowned leads expire in shared inboxes; unrecorded outcomes make win rates, lost reasons and channel economics unknowable. The system — not the manager — enforces the rule by assigning the owner on arrival, showing every lead past its deadline, and refusing to close a record without a reason. In La Machine, intake assignment and CRM tasks do this (see In La Machine).

QUÉBEC & CANADA NOTE The lead's consent and the follow-up sequence (as of September 2026) A web inquiry is personal information and, usually, the beginning of a lawful basis for follow-up. Under CASL, an inquiry or application made to the business creates implied consent for commercial electronic messages for six months after the inquiry (CASL s. 10(10)(e)); a purchase creates it for two years (s. 10(10)(a)). Follow-up emails and texts to the lead must still identify the sender and offer an unsubscribe mechanism (s. 6), and any unsubscribe must be honored within ten business days (s. 11). Telephone follow-up of a person who contacted you is not telemarketing in the usual sense, but calls to prospects who did not inquire are subject to the CRTC's Unsolicited Telecommunications Rules and the National Do Not Call List. Under P-39.1 (Law 25) and PIPEDA, the inquiry's personal information may be used for the purpose it was collected — responding and following up — and the person must be informed of that purpose. In plain terms: the six-attempt sequence over ten days is lawful for a person who inquired, provided each message identifies you and can be stopped; after six months without express consent, electronic follow-up stops. → Chapter 10; → see Book H3.

9.6 Feedback Loops: Win/Loss, Objections, Content Gaps

9.6.1 Win/loss reasons

Every closed opportunity — won or lost — is a data point about the market, and it is worthless if the reason is "price" or "went with competitor" typed in a hurry. A useful win/loss record has four fields: the outcome, the primary reason in the buyer's words (from a two-question call to the buyer, not the salesperson's guess), the alternative chosen (including "did nothing"), and what would have changed the outcome. Ten of these a quarter, read together, tell marketing which pillar is not landing (Chapter 5), which objection lacks proof (Chapter 4), and which segment converts below expectations (Chapter 3). Calling lost buyers is uncomfortable and consistently the most instructive research a small business can do.

9.6.2 Objections and the content gap

The objections sales hears three times are marketing's next assignment: a page, a case, a comparison, a guarantee, a price range — placed where the persona meets the objection in the journey (Chapter 2). The SLA (Table 9.2) commits marketing to respond within a month. The reverse loop matters too: marketing tells sales what the market is asking (the search queries, the form questions, the content downloaded most), so the first conversation starts from what the buyer already read.

9.6.3 Buyer quotes

The most valuable output of sales conversations is the buyer's own language — the trigger described in their words, the fear named, the outcome wanted. A shared document or a CRM field where salespeople paste one quote per conversation becomes, within a quarter, the best source of headlines, email subject lines and case-study openings the company will ever have (Chapter 4, Chapter 5). The fortnightly review reads the new quotes aloud.

9.6.4 The fortnightly review: a thirty-minute agenda

The review is where the SLA lives or becomes theatre (Section 9.7.2). It is short, because it is frequent, and it follows the same agenda every time so that preparation is a matter of pulling six numbers and three lists. Minutes 0–10 — the six metrics, read from the funnel and the log, each against its target and its previous value: lead-to-MQL rate, MQL acceptance rate, speed to first contact (median and share within standard), SQL-to-opportunity rate, win rate, cost per customer by source. A number that moved by more than its usual variation gets one sentence of explanation from its owner — not a debate. Minutes 10–15 — leads past their deadline: the list from the system, each with its owner, and the reason; the point is not blame but the pattern (Friday afternoons? one source? one person overloaded?). Minutes 15–22 — the recycle and loss reasons of the fortnight: what the buyers said, grouped; any objection heard three times becomes a marketing task with an owner and a date, on the spot. Minutes 22–27 — the buyer quotes: read aloud, two or three, chosen by the salespeople; marketing notes the ones that could become headlines or subject lines. Minutes 27–30 — decisions: one change to a definition, a threshold, a sequence or a piece of content, with an owner and a date — or an explicit "no change, and here is why". The decisions of the last three reviews are listed at the top of the next agenda, with their status; a review that produced no decisions three times running is the signal of Section 9.7.2.

In a solo business the review is a fifteen-minute Friday ritual with the same agenda, written down; the act of writing "no change" is what keeps the marketing hat honest about what the sales hat learned. In an established business the review is attended by both managers and, once a month, by the owner, who reads the six numbers and the decisions and asks one question: which of these decisions changed a number?

WHO DOES THIS Alignment in the three situations Solo: the owner wears both hats, so alignment is a matter of process: the automatic acknowledgment, the routing to a task with a deadline, the six-attempt sequence, and a Friday habit of recording outcomes and one quote per conversation; the risk is the marketing hat forgetting what the sales hat learned, which the recorded quotes prevent. Small team: one person markets, one or two sell; the SLA is written on one page, signed by the owner, and reviewed in thirty minutes every two weeks; the CRM enforces owner, deadline and outcome. Established: a marketing manager and a sales manager co-own the SLA, the shared metrics and the fortnightly review; win/loss calls are made by someone outside the deal (marketing, or an external interviewer); the owner or general manager sees the six numbers monthly and arbitrates definition changes. In every case, the owner signs the definitions and the service levels, because they commit both functions to each other.

9.7 One Team in a Small Business

9.7.1 The advantage of being small

Large companies spend years aligning marketing and sales because the functions sit in different buildings with different incentives. A small business has none of that distance: the person who writes the email can hear the salesperson's calls, the salesperson can edit the landing page, and the owner sees both funnels on one screen. The advantage is real, and it is squandered when the small business copies the large company's structure — separate tools, separate reports, separate meetings — instead of building one process with two hats.

9.7.2 One process, two hats

The practical design for a small business: one contact base (Chapter 13) in which marketing and sales work the same records; one set of stage definitions (Section 9.2) written in it; one weekly count of the funnel from visitor to customer, by source (Chapter 2); one fortnightly review where the six shared metrics, the win/loss reasons, the objections and the quotes are read; and one owner per lead at every moment. Whether the hats sit on one head or three, the process is identical — which is also what makes it possible to hire the first salesperson or the first marketer without inventing a new system (Chapter 12).

WATCH OUT Alignment theatre Two failure modes look like alignment and are not. The joint meeting without definitions: marketing and sales meet weekly, discuss "lead quality" and leave with opinions, because nobody agreed what a lead is. The dashboard without an owner: the funnel is displayed on a screen, the numbers fall, and nobody's commitment was attached to any of them. Alignment is not a meeting or a dashboard; it is written definitions, named owners, deadlines the system enforces, recorded outcomes, and a review that changes something. If the last three reviews changed nothing — no criterion, no content, no sequence — the process is theatre.

Costly Mistakes

1. No shared definition of a lead. Marketing counts forms, sales counts buyers, both report success. Consequence: the largest leak in the funnel — between hand-off and first contact — is invisible to both, and the argument about "lead quality" never ends.
2. Leads in a shared inbox. No owner, no deadline. Consequence: median response times of a day or more, a third of inquiries never contacted, and paid leads converting at a fraction of their potential — exactly the pattern the research describes.
3. One attempt, then "no answer". Consequence: leads that were merely busy are discarded; the cost per customer by channel doubles for reasons that have nothing to do with the channel.
4. Discarding instead of recycling. "Not now" and "no budget this year" leads deleted or forgotten. Consequence: the cheapest future customers — people who already fit and already consented — are paid for a second time when they re-enter through an advertisement.
5. Win/loss reasons typed by the salesperson. "Price" and "competitor" in every record. Consequence: marketing repairs the wrong pillar; the real objection — usually risk, proof or trust — stays unanswered.
6. Alignment theatre. Meetings and dashboards without definitions, owners, deadlines or recorded outcomes. Consequence: the appearance of a process and the results of none.

Key Takeaways

• Marketing and sales are two halves of one process — the conversion of a stranger into a customer — and the leak between them is usually the largest in the funnel.
• Definitions come before dashboards: lead, MQL, SQL and opportunity are each defined by an observable event, agreed in writing and applied to every record.
• The MQL is the hinge: fit from the ICP plus intent signals from your own customers' behavior; tune the criteria to an acceptance rate of about two thirds or more.
• Recycle, do not discard: every non-converting lead has a reason and a nurture path; closing the recycle loop is worth more than most new channels.
• The hand-off travels with context — source, consent, fit, intent, persona, questions, promise — to one named owner with a deadline.
• Speed to lead decides outcomes: contact within the hour converts several times better than contact the next day; set a standard in minutes, an automatic acknowledgment, and six attempts over ten days before recycling — lawful for a person who inquired, under CASL's six-month implied consent, provided each message identifies you and can be stopped.
• The SLA makes commitments explicit — volume and fit, hand-off, speed, persistence, outcome data, feedback, review — and six shared metrics read every two weeks make it live.
• Feedback loops turn selling into research: win/loss reasons in the buyer's words, objections heard three times, and one buyer quote per conversation.
• A small business builds one process with two hats, not two departments — and that is its advantage.

Your Action Plan

The action plan below presents the six steps that turn this chapter into work, in the order in which each step consumes the output of the one before. For each step, the table gives the deliverable that must exist at the end, the timing on a small business's calendar, and the verification criterion by which anyone — a co-founder, a mentor, the owner's own future self — can tell that the step was done. The criterion column is the one to read first: a step whose criterion you cannot picture is a step you do not yet understand.

In La Machine

The lifecycle of Figure 9.1 runs across the marketing module and the CRM.

• CRM intake (crm/intake) is the client's own profile form, in three steps — identity, consent and subscription — and not a router that receives leads from forms, landing pages, magnets and imports; the hand-off therefore assigns its owner on the record itself. CRM tasks (crm/tasks) redirects to Productivity → Tasks (productivity/tasks), whose activities carry the follow-up sequence, each with a scheduled date and an assignee typed as free text, so that every lead has one name and one date; the list is not filtered by person and has no count of late tasks, so leads past their service level are found by reading the scheduled dates. This is where the rule of Section 9.5.2 is kept, by a person who reads the list, since nothing on the page raises an alert.
• Contacts and companies (business/marketing/contacts, business/marketing/companies) hold the hand-off package — source, consent status, segment and fit tags, intent signals from forms and responses, persona tag, notes — on the record the salesperson opens; client profiles and details (crm/client-profile, crm/client-details) continue the record after the sale.
• Forms, landing pages and lead magnets (business/marketing/forms/responses, business/marketing/landing-pages/responses, business/marketing/lead-magnets/responses) record what each lead asked for — the intent signals of the MQL criteria — with the source; sequences (business/marketing/sequences) send the automatic acknowledgment within minutes and run the nurture paths for each recycle reason, against the consent register (business/marketing/consent) and the do-not-reach list.
• Tags, segments, responses and sales records give the counts and transitions — lead to MQL, MQL to SQL, SQL to opportunity, win rate — that the fortnightly review reads, and the log (business/marketing/log) shows what left; funnels keep no stage counts, and powerdialer (business/marketing/powerdialer) is a placeholder, so the calling part of the attempt sequence is made outside the product (→ see Book N5).
• Plan tasks and Runs, in the Control center section (/taches, /executions), can carry the fortnightly review and the "objection heard three times → content task" rule as recurring tasks with verification criteria; no page creates a plan task, so they are loaded through the programming interface, a script or the assistant, each recurring one with a calendar expression, because the product reads the frequency label only as French words (→ see Book N2).

What La Machine does not do: it does not write your stage definitions or MQL criteria — you write them and express them as tags and segments, since the module has no scoring object (→ see Book B8, Book N5); it does not make the calls or hold the conversations; and it does not interview lost buyers — the win/loss call is a human task it schedules and whose result it records. Book N4 (CRM, Contacts and Prospect Data) covers intake, tasks and profiles page by page.

Self-Check

1. At what moment does a contact stop being marketing's responsibility and become sales', and what must travel with it?
2. Define MQL and SQL, and say what acceptance rate suggests the MQL criteria are well tuned.
3. What did the Harvard Business Review study of web-lead response find about contacting a lead within an hour, and what does it imply for a small business's standard?
4. Name the six shared metrics of the SLA.
5. Why are win/loss reasons typed by the salesperson unreliable, and what should replace them?

Answers. (1) When the person has been identified, fits the ICP, has shown intent and is handed over with context — source, consent status, fit, intent signals, persona, what they asked, qualification answers and the promise made — to one named owner with a deadline (Sections 9.1.1 and 9.3.2). (2) An MQL fits the ICP and has shown intent beyond identification (pricing page, quote request, second download, reply); an SQL is an MQL that sales has contacted and accepted after confirming need, a decider and timing; an acceptance rate of about two thirds or more suggests the criteria are well tuned — much lower means too loose, near 100% with too few means too tight (Section 9.2). (3) Firms that attempted contact within an hour were nearly seven times as likely to qualify the lead as those that waited even an hour longer, and more than sixty times as likely as those that waited a day; the standard should be minutes during business hours, an automatic acknowledgment plus a next-morning call outside them, and a defined sequence of attempts (Oldroyd et al., 2011; Section 9.4). (4) Lead-to-MQL rate, MQL acceptance rate, speed to first contact, SQL-to-opportunity rate, win rate, cost per customer by source — plus cycle length over time (Section 9.5.2). (5) Because they record the salesperson's guess — usually "price" or "competitor" — rather than the buyer's reason; a two-question call to the buyer recording the primary reason in their words, the alternative chosen and what would have changed the outcome (Section 9.6.1).

Chapter References

An Act to promote the efficiency and adaptability of the Canadian economy by regulating certain activities that discourage reliance on electronic means of carrying out commercial activities, S.C. 2010, c. 23 (Canada's Anti-Spam Legislation), ss. 6, 10, 11. Justice Laws Website. Retrieved September 7, 2026, from https://laws-lois.justice.gc.ca/eng/acts/E-1.6/

Act respecting the protection of personal information in the private sector, CQLR c P-39.1. LégisQuébec. Retrieved September 7, 2026, from https://www.legisquebec.gouv.qc.ca/en/document/cs/P-39.1

Personal Information Protection and Electronic Documents Act, S.C. 2000, c. 5. Justice Laws Website. Retrieved September 7, 2026, from https://laws-lois.justice.gc.ca/eng/acts/P-8.6/

Canadian Radio-television and Telecommunications Commission. (n.d.). National Do Not Call List. Retrieved September 7, 2026, from https://lnnte-dncl.gc.ca/

Cespedes, F. V. (2014). Aligning strategy and sales: The choices, systems, and behaviors that drive effective selling. Harvard Business Review Press.

Kotler, P., Rackham, N., & Krishnaswamy, S. (2006). Ending the war between sales and marketing. Harvard Business Review, 84(7/8), 68–78.

Oldroyd, J. B., McElheran, K., & Elkington, D. (2011). The short life of online sales leads. Harvard Business Review, 89(3), 28.

Chapter 10 — Marketing Ethics, Consent and Compliance

The newsletter went to 4,200 addresses collected over nine years. Three people complained to the regulator. The investigation did not ask whether the newsletter was good; it asked, for each address, when and how did this person agree to receive it? — and the company could not answer.

What You'll Be Able to Do

• Explain permission marketing as a strategy, not a legal constraint — and why it is the cheaper one.
• Classify every contact by consent type — express, implied, none — with its date, source and expiry, and know what each type allows.
• State the core rules of anti-spam and telemarketing law in Canada, and what they require of every message and call.
• Meet the privacy obligations that apply to a small business's marketing data: purposes, information, consent, a person in charge, a published policy, tracking technologies, retention.
• Make claims, use testimonials and disclose relationships lawfully — with a file behind every claim.
• Build the outbound gate: exclusion lists and the blocking filter that runs before every send.
• Document compliance so that the question "when and how did this person agree?" always has an answer.

Why It Matters

Ethics and compliance are treated here as design requirements of the marketing system, not as an appendix — because in a small business they are cheap to build in and ruinous to retrofit. A contact base with consent recorded is an asset that can be used for years; one without it is a liability that cannot be used at all. Claims with files behind them survive a complaint; claims without them become penalties and lost trust. Beyond the law, permission is also the better strategy: messages people asked for are read, and a business known for respecting its audience earns the referrals that interruption never will. For the owner, this chapter is protection; for the team, it is the set of rules the system enforces so that nobody has to remember them under pressure.

In This Chapter

• 10.1 Permission Marketing as Strategy
• 10.2 Consent: Express, Implied, Unbundled
• 10.3 Anti-Spam and Telemarketing Rules
• 10.4 Privacy and Tracking
• 10.5 Truthful Claims, Testimonials and Disclosures
• 10.6 Exclusion Lists and the Blocking Filter
• 10.7 Documenting Compliance

10.1 Permission Marketing as Strategy

10.1.1 The strategic argument

Seth Godin's distinction between interruption marketing — messages forced on people who did not ask for them — and permission marketing — anticipated, personal, relevant messages sent to people who agreed to receive them — was made in 1999, before most of today's rules existed (Godin, 1999). The argument was economic, not legal: attention is scarce (Chapter 1), interruption gets more expensive as it gets more common, and a relationship built on permission compounds — each message the person wanted makes the next one more welcome. Twenty-seven years later, the law in Canada has caught up with the economics: the rules of this chapter mostly require what permission marketing recommended anyway.

Four principles summarize both the strategy and the law. Consent — you contact people who agreed, in a way they understood, for the purposes they were told. Identification — every message says who sent it and how to reach you. Honesty — what you claim is true, provable and not misleading; what you promise, you deliver. Respect for the person's choices — stopping when asked, promptly and without argument; using information only for what it was collected for; letting people see and correct what you hold.

10.1.2 The four legal pillars, and where they are treated

Chapter 1 named the four pillars of Canadian marketing law: anti-spam (CASL), privacy (P-39.1/Law 25 in Québec; PIPEDA federally), truthful marketing (Competition Act; Consumer Protection Act in Québec) and language (Charter of the French language). This chapter gives each an overview sufficient to design a compliant marketing system, with the exact obligations that matter most to a small business. The full treatment belongs to Collection H: Book H2 (Privacy and Personal Information), Book H3 (Anti-Spam and Telemarketing Compliance) and Book H4 (Consumer Protection, Advertising Law, Competition and Language Requirements). Nothing here is legal advice; the statutes were consulted in their consolidated versions on September 7, 2026, and they change.

KEY CONCEPT Compliance is a filter before every send, not a policy on a shelf A compliance policy that lives in a binder protects nobody, because the violation happens at the moment a message leaves — in a hurry, by an employee, from a list nobody checked. Compliance that works is built into the system that sends and calls: a consent field the sender cannot bypass, an exclusion list checked automatically, an unsubscribe link inserted by the template, identification pulled from settings, a log of what went to whom under which consent. Figure 10.1 (Section 10.6) draws that gate. Once it exists, compliance costs almost nothing per message; without it, every send is a gamble.

10.2 Consent: Express, Implied, Unbundled

10.2.1 Two laws, one contact record

Consent is regulated twice over for a marketing contact in Québec: CASL decides whether you may send commercial electronic messages to the address, and privacy law (P-39.1, or PIPEDA elsewhere in Canada) decides whether you may collect, use and disclose the personal information at all — the name, the address, the behavior recorded. The two regimes overlap but differ in detail, and a contact record needs to satisfy both. The practical unit is the consent record: for each contact, the type of consent, its date, its source (the form, the purchase, the inquiry, the conversation), the wording they saw, and — for implied consent — its expiry.

DEFINITION Express, implied and unbundled consent Express consent is an affirmative act by the person — ticking an unticked box, typing an address into a subscription form, saying yes when asked — after being told, clearly, what they are agreeing to and who is asking. Implied consent is consent the law infers from a relationship or a situation — an existing business relationship, a published address — for a limited time and purpose, without the person having said yes. Unbundled consent means that agreement to marketing is requested separately from anything else: not buried in terms of sale, not a condition of a download, not one box covering three purposes. Under CASL, express consent does not expire until withdrawn; implied consent expires with the relationship that created it. Under P-39.1, consent must be clear, free, informed, given for specific purposes and requested for each purpose in clear and simple language — and, in writing, presented separately from any other information (P-39.1 s. 14).

10.2.2 What each type of consent allows

Table 10.1 sets out the types of consent that Canada's anti-spam legislation recognizes, one per row: express consent; the implied consent that arises from an existing business relationship, from an existing non-business relationship, from a conspicuously published address and from a disclosed address; and the messages that are exempt. For each type, the columns give how it arises, how long it lasts, what it allows you to send, and the record you must keep to prove it. The two facts to retain are that implied consent expires and that the burden of proof is on the sender; the paragraphs after the table draw the consequences for the system.

Table 10.1 — Consent types under CASL and what they allow (as of September 2026)

Two features of the table drive system design. First, implied consent expires — which means the contact base must hold dates and compute expiries, and the sending function must respect them. Second, the burden of proof is on the sender: under CASL s. 13, a person who alleges they have consent has the onus of proving it. A consent you cannot document is, for enforcement purposes, a consent you do not have.

THE RULE What CASL requires of every commercial electronic message Under An Act to promote the efficiency and adaptability of the Canadian economy by regulating certain activities that discourage reliance on electronic means of carrying out commercial activities, S.C. 2010, c. 23 (CASL), a commercial electronic message — an email, a text message, a message through a social platform that encourages participation in a commercial activity — may be sent only if (a) the recipient has consented, expressly or impliedly, and (b) the message identifies the sender and anyone on whose behalf it is sent, gives contact information valid for at least 60 days, and contains an unsubscribe mechanism (CASL s. 6(1)–(3)). A request for express consent must state its purpose and identify the requester clearly and simply (s. 10(1)). An unsubscribe request must be given effect without delay, and no later than 10 business days (s. 11). The maximum administrative monetary penalty is $1,000,000 per violation for an individual and $10,000,000 for any other person (s. 20(4)); the Canadian Radio-television and Telecommunications Commission (CRTC) enforces. In plain terms: consent you can prove, a message that says who you are and how to stop, and a stop that happens fast. → see Book H3.

10.2.3 Consent under Law 25

Privacy law adds its own layer, applicable to the personal information itself. Under P-39.1, a business must inform the person at the time of collection of the purposes, the means of collection, their rights of access and rectification and their right to withdraw consent, and — where applicable — the third parties to whom the information may be communicated and the possibility of communication outside Québec (s. 8). Consent must be clear, free and informed, given for specific purposes, requested for each purpose in clear and simple language, and a written request for consent must be presented separately from any other information (s. 14). Personal information may not be used within the enterprise for purposes other than those for which it was collected without the person's consent, apart from listed exceptions such as purposes consistent with the original ones or necessary to deliver a requested service (s. 12). Federally, the Office of the Privacy Commissioner's guidelines on obtaining meaningful consent set out comparable expectations for PIPEDA: emphasize key elements, allow control over the level of detail, provide clear yes-or-no options, be innovative and creative in presenting information, consider the consumer's perspective, make consent a dynamic and ongoing process (Office of the Privacy Commissioner of Canada, 2018).

THE RULE Law 25: unbundled, purpose-specific consent and information at collection Under the Act respecting the protection of personal information in the private sector, CQLR c P-39.1 (as reformed by Law 25): at collection, tell the person the purposes, the means, their rights of access, rectification and withdrawal, and any communication to third parties or outside Québec (s. 8). Ask for consent per purpose, in clear and simple language, and — in writing — separately from everything else; consent is valid only for the time necessary to achieve its purposes, and consent not given in accordance with the Act is without effect (s. 14). Use the information only for the purposes collected unless consent or an exception applies (s. 12). The Commission d'accès à l'information (CAI) may impose administrative monetary penalties of up to $50,000 for a natural person and, otherwise, $10,000,000 or 2% of worldwide turnover for the preceding fiscal year, whichever is greater (s. 90.12); penal fines reach $25,000,000 or 4% of worldwide turnover (s. 91), doubled for a subsequent offense (s. 92.1). In plain terms: one box per purpose, in plain words, on its own — and the information used only for what the person was told. → see Book H2.

10.2.4 The consent audit

METHOD The consent audit (half a day for a few thousand contacts) When to use it: before any campaign to an existing list; at the quarterly review; before importing any list. Deliverable: every contact classified by consent type with date, source and expiry; a count by class; the excluded set. Step 1 — Inventory the sources (30 min). List every way an address entered your base: website forms (which wording? which box?), purchases, quotes, event sign-ups, business cards, imports, purchased lists, the owner's address book. Step 2 — Classify each source (1 hour). For each source, decide which consent type it produces under Table 10.1 — or none. A purchase produces implied consent for two years; a subscription form with a clear purpose and an unticked box produces express consent; a scraped or purchased list produces none; a business card produces disclosed-address implied consent for role-relevant messages only. Step 3 — Date and compute (1 hour). For every contact, record the consent type and the date of the event that created it; compute the expiry for implied consent (two years or six months). Where the date is unknown, treat the consent as none. Step 4 — Exclude and separate (30 min). Move contacts with no provable consent, expired implied consent, unsubscribes and bounces to the excluded set — kept, not deleted, so that they are never re-imported and so that a lawful re-permission attempt can be considered (Book H3 explains the narrow ways to seek express consent from such contacts, which do not include sending them a commercial email). Step 5 — Fix the sources (1 hour). Rewrite every form so that it produces express consent that satisfies both laws: purpose stated, requester identified, one unticked box per purpose, separate from the terms, the record stored with date, wording and source. Pitfalls: assuming that "they are customers" means consent (it does — for two years, for messages about your products); treating a downloaded guide as consent for a newsletter (only if the box said so); deleting excluded contacts (you lose the suppression list).

DO THE MATH What a clean list is worth, and what a dirty one costs Illustrative figures. A firm has 4,200 addresses. The audit finds 1,900 with express consent, 1,100 with valid implied consent (purchases and inquiries within the periods), and 1,200 with no provable consent or expired periods. Sending to the 3,000 lawful contacts: with a 42% open rate and a 4% click rate typical of a wanted newsletter, 1,260 opens and 120 clicks per issue. Sending to all 4,200 would add, at best, a few dozen clicks from people who did not agree — and expose the firm to complaints from 1,200 people, each a potential violation with a maximum penalty of $10,000,000 for a business (CASL s. 20(4)), plus the far more common consequence: spam complaints that push the sending domain onto blocklists, so that the 3,000 lawful contacts stop receiving the newsletter too. The 1,200 excluded contacts are not lost value; they are 1,200 people to invite back through lawful channels — a re-permission postcard, a phone call to the business contacts, a relevant offer on the website — over the next year. The clean list of 3,000 that arrives in the inbox is worth more than the dirty list of 4,200 that arrives in the spam folder.

10.3 Anti-Spam and Telemarketing Rules

10.3.1 CASL in practice

CASL applies to commercial electronic messages sent to an electronic address — email, SMS, instant messages, messages through social platforms — whenever the message encourages participation in a commercial activity, whether or not the sender expects profit. Section 10.2 covered consent and the three content requirements (identification, contact information, unsubscribe). Four practical points complete the picture for a small business. Transactional and service messages — a receipt, a shipping notice, a warranty notice, a response to a request for a quote — are exempt from the consent requirement or excluded from the definition, but a promotion slipped into a receipt makes the whole message commercial. Every message identifies the sender: the legal name, a mailing address and at least one of a phone number, an email address or a web address; La Machine's marketing settings hold these for every template. The unsubscribe must be easy: a link or reply that works without a login, honored within ten business days at the latest (s. 11), and honored for all commercial messages unless the person chose otherwise. Referral messages — asking a customer to forward, or sending to a referred person — have their own narrow exemption for a single message that names the referrer; beyond that, consent is needed (Book H3).

Enforcement is real and continuing: the CRTC publishes its enforcement actions, which have included administrative monetary penalties against small and mid-size businesses as well as large ones, and undertakings that require compliance programs (Canadian Radio-television and Telecommunications Commission, n.d.). The private right of action that CASL once provided for, at sections 47 to 51, was repealed before it ever came into force; the Act as consolidated to September 30, 2026 shows those sections as repealed.

10.3.2 Telemarketing: the CRTC's rules and the National Do Not Call List

Telephone marketing to consumers is governed by the CRTC's Unsolicited Telecommunications Rules, made under the Telecommunications Act, S.C. 1993, c. 38. Their core obligations for a business that makes marketing calls: register with the National Do Not Call List (DNCL) operator and subscribe to the list for the area codes called, scrub calling lists against the National DNCL regularly (numbers may not be called for the period specified in the rules after the list is downloaded), maintain an internal do-not-call list honored for the period the rules require, identify the caller and the business at the start of the call and provide contact information on request, respect the calling hours the rules set, and follow the specific rules for automatic dialing-announcing devices. An existing business relationship — a purchase within the preceding eighteen months, a contract in force or recently expired, an inquiry within the preceding six months — exempts a call from the National DNCL requirement (not from the internal list or the other rules), and calls to businesses are treated differently from calls to consumers (Canadian Radio-television and Telecommunications Commission, n.d.). The periods and calling hours are the kind of detail that changes: verify the current rules on the National DNCL site before designing a calling program (→ see Book H3; → see Book E2 for outbound sales practice).

QUÉBEC & CANADA NOTE Text messages, calls and the itinerant merchant (as of September 2026) Three channels raise questions small businesses often get wrong. SMS: a text message is a commercial electronic message under CASL — consent, identification and an unsubscribe (reply STOP) are required, and the CRTC's rules on automatic dialing-announcing devices may also apply to automated texts and calls. Calls to existing customers: the eighteen-month and six-month existing-business-relationship windows of the telemarketing rules are not the same as CASL's two-year and six-month windows for electronic messages — record both. Door-to-door and off-premises consumer sales in Québec: the Consumer Protection Act, CQLR c P-40.1 treats a merchant who solicits or concludes contracts away from their establishment as an itinerant merchant, with permit requirements, mandatory contract content and a cancellation right for the consumer; the Office de la protection du consommateur (OPC) administers it. → see Book H3, Book H4.

10.4 Privacy and Tracking

10.4.1 What Law 25 asks of a small business's marketing

Law 25 applies to every enterprise that collects personal information in Québec, regardless of size, and its obligations are proportionate but not optional. For marketing data, seven matter most:

1. A person in charge of the protection of personal information must be designated. By default it is the person exercising the highest authority — the owner — who may delegate in writing; the title and contact information of the person in charge must be published on the website (P-39.1 s. 3.1).
2. Governance policies and practices must exist, proportionate to the enterprise, covering the keeping and destruction of information, the roles of personnel across the information's life cycle and a complaint process; they are approved by the person in charge and published in simple and clear language on the website (s. 3.2).
3. Information must be given at collection: purposes, means, rights, withdrawal, third parties, communication outside Québec (s. 8), and, when technology that can identify, locate or profile a person is used — analytics that build a profile, tracking pixels, some cookies —, information about that technology and about the means to activate those functions (s. 8.1), which implies that such functions are not active by default.
4. Consent must be obtained as described in Section 10.2.3 (s. 14), and use must be limited to the purposes for which the information was collected (s. 12).
5. Communication outside Québec is regulated: before communicating personal information outside the province (which includes storing it with a provider whose servers are elsewhere), a privacy impact assessment is required and the information must receive adequate protection (s. 17).
6. Confidentiality incidents call for a register of incidents, and for notification of the CAI and of the persons concerned when an incident presents a risk of serious injury (ss. 3.5–3.8).
7. Retention and destruction are bounded: information is kept only as long as necessary for its purposes and then destroyed or anonymized (s. 23).

Federally, PIPEDA sets comparable principles — accountability, identifying purposes, consent, limiting collection, use and retention, accuracy, safeguards, openness, individual access and challenging compliance — for businesses in provinces without substantially similar legislation and for interprovincial and international flows (Office of the Privacy Commissioner of Canada, 2018).

10.4.2 Tracking: pixels, analytics and cookies

Marketing tracking — analytics that follow a visitor across pages, advertising pixels that build audiences for retargeting, cookies that recognize a returning visitor — collects personal information the moment it can identify, locate or profile a person. Under P-39.1 s. 8.1, the visitor must be informed of the technology and of the means to activate the identifying or profiling functions; the CAI's guidance treats this as requiring that non-essential functions be off by default and turned on by the visitor, with a clear explanation. For a small business, the practical design is: essential cookies only until the visitor chooses; a banner that explains, in plain words, what each category does and lets the visitor turn categories on; a published policy that names the technologies and the third parties (analytics and advertising platforms) and states the retention; and an internal record of the choice. Analytics can be configured to minimize identification; advertising audiences built from site visitors are profiling and require the opt-in (→ see Book B7 for analytics; Book H2 for the full regime).

WATCH OUT The purchased list and the scraped address Buying a list of contacts, scraping addresses from websites or exporting a professional network's connections into your email tool are the three most common ways small businesses acquire unlawful contacts. A purchased list rarely carries consent you can prove — the seller's assurance is not your record; scraped addresses at best qualify for the narrow "conspicuous publication" implied consent, only for messages relevant to the person's role and only if no statement refuses unsolicited messages; exported connections are personal information collected for another purpose. Under CASL s. 13 you must prove consent; under P-39.1 s. 12 you may use information only for the purposes it was collected for. The safer path is to build the list yourself: a lead magnet worth the address (Book B9), forms that produce express consent, and business contacts approached in ways the rules allow — by mail, by phone within the telemarketing rules, or by a single relevant message where an implied-consent basis genuinely exists, recorded.

10.5 Truthful Claims, Testimonials and Disclosures

10.5.1 The Competition Act's civil regime

The Competition Act, R.S.C. 1985, c. C-34 treats a range of marketing conduct as reviewable under its civil provisions — no intent to deceive is needed for a civil finding. The conduct most relevant to a small business (s. 74.01): making a representation to the public that is false or misleading in a material respect (s. 74.01(1)(a)); making a performance, efficacy or length-of-life claim not based on an adequate and proper test, with the burden of proof on the person making the claim (s. 74.01(1)(b)); making a warranty or guarantee claim that is misleading or that there is no reasonable prospect of honoring (s. 74.01(1)(c)); ordinary selling price claims — "was $200" — where the product was not sold at that price in substantial volume or offered at it in good faith for a substantial time (s. 74.01(2)–(3)); and, since the 2022 amendments, drip pricing — advertising a price that cannot be attained because of mandatory fixed charges added later, unless imposed by law (s. 74.01(1.1)). The general impression conveyed, as well as the literal meaning, is considered (s. 74.03(5)). Remedies range from orders to cease and to publish corrective notices to administrative monetary penalties up to the greater of $10,000,000 ($15,000,000 for subsequent orders), three times the benefit derived, or 3% of annual worldwide gross revenues for a corporation (s. 74.1). The Competition Bureau enforces and publishes guidance.

THE RULE Competition Act: no misleading claims, tested performance claims, disclosed relationships Under the Competition Act, R.S.C. 1985, c. C-34: do not make a representation to the public that is false or misleading in a material respect, judged by its general impression as well as its literal meaning (s. 74.01(1)(a), s. 74.03(5)); base every performance, efficacy or durability claim on an adequate and proper test conducted before the claim is made, and be ready to prove it (s. 74.01(1)(b)); make "regular price" claims only when the regular price is genuine (s. 74.01(2)–(3)); never let the advertised price drip (s. 74.01(1.1)); and disclose material connections — a paid influencer, an incentivized review, an affiliate commission — so that the audience is not misled about the independence of the endorsement (Competition Bureau guidance; Ad Standards' Canadian Code of Advertising Standards). Penalties for corporations reach the greater of $10,000,000, three times the benefit or 3% of worldwide gross revenues (s. 74.1). In plain terms: for every claim, a file; for every endorsement, the truth about who paid. → see Book H4.

10.5.2 Testimonials, reviews and influencers

A testimonial is a representation and follows the same rules: genuine, current, reflecting the honest opinion and the typical experience of a real customer; any performance figure in it tested like any other claim. Reviews must not be fabricated, purchased and presented as spontaneous, or filtered to hide typical negative experiences; the Competition Bureau has taken enforcement action against fake and misleading reviews and publishes guidance on online reviews and on influencer marketing, which expects material connections — payment, free products, affiliate links, family or employment ties — to be disclosed clearly and close to the endorsement (Competition Bureau, n.d.). Ad Standards' Canadian Code of Advertising Standards and its influencer disclosure guidelines set the industry's self-regulatory expectations (Ad Standards, n.d.). Quoting a customer by name is also a use of their personal information: obtain and keep their written agreement (P-39.1, PIPEDA).

10.5.3 Contests, consumer advertising and language

Contests and promotions with a chance element are regulated federally — the Competition Act requires adequate and fair disclosure of the number and value of prizes, the regional allocation and the odds (s. 74.06), and the Criminal Code restricts lotteries, which is why contests include a skill-testing question and a no-purchase entry route. In Québec, the former regime of registration, duties and reporting to the Régie des alcools, des courses et des jeux for publicity contests was abolished by legislation in 2023; the federal rules and the Consumer Protection Act continue to apply — verify the current requirements before launching a contest (→ see Book H4). Consumer advertising in Québec follows the Consumer Protection Act, CQLR c P-40.1 — including its rules on price display, on advertising directed at children under thirteen (prohibited, with narrow exceptions), and on prohibited business practices — administered by the OPC. Language: commercial publications, websites, signage and advertising in Québec must comply with the Charter of the French language, CQLR c C-11 (Chapter 7, Section 7.6.3, for trademarks on signage; → see Book H4).

CASE IN POINT The claim that could not be proven Fictional example; names and figures are invented. Vireo Nutrition, a nine-person maker of a plant-based protein blend, advertised "clinically proven to increase energy" on its packaging and site. The phrase came from a supplier's marketing sheet; the "clinical proof" was a small, unpublished study on a different product. A competitor complained to the Competition Bureau; the Bureau's inquiry asked for the adequate and proper test behind the claim (s. 74.01(1)(b)). Vireo had none. It resolved the matter with a consent agreement: withdrawing the claim, publishing a corrective notice, paying an administrative penalty it could ill afford, and reprinting $38,000 CAD of packaging. The lesson its founder drew was procedural: a claims register — every marketing claim, its wording, the evidence file, the date and the approver — reviewed before any claim is published. Two years later, Vireo's marketing rested on verifiable statements ("22 g of protein per serving", "made in Québec", "rated 4.7 by 640 customers as of March 2026") and on named customers' stories with signed authorizations. Its conversion rate did not fall; its exposure did.

10.6 Exclusion Lists and the Blocking Filter

10.6.1 The gate

Everything in this chapter converges on one mechanism: the outbound gate — the set of checks that every message and every call passes before it leaves, automatically, so that compliance does not depend on anyone's memory (Figure 10.1).

Figure 10.1 — The outbound gate: a contact and a message pass five gates — excluded? consent valid and unexpired? deliverable? relevant? compliant? — and either send (logged) or block (reason recorded). Original figure; periods are those of CASL as of September 2026.

1. Excluded? The contact is on the do-not-reach list, is a partner's account, is in dispute, is outside the territory, or is on a do-not-call list (Chapter 3, Section 3.5.2). If yes, block.
2. Consent valid? The record shows express consent, or implied consent whose period has not expired — two years from a purchase, six months from an inquiry. If no, block.
3. Deliverable? Not unsubscribed, not withdrawn, no hard bounce, a valid address or number. If no, block — and never re-import.
4. Relevant? The message matches the segment and persona (Chapters 3 and 4), the frequency cap has not been exceeded, and the message suits the contact's stage. If no, hold.
5. Compliant? Identification and unsubscribe present, claims in the message backed by a file, French where required, disclosures included. If no, hold.

Gates 1 to 3 are enforced by the sending system from data on the record; gates 4 and 5 are designed into the segment definitions, the templates and the review of new content. Every block records its reason; the reasons feed the consent audit (Section 10.2.4) and the nurture path (Chapter 9): a contact blocked for expired implied consent is a candidate for lawful re-permission, not a lost cause.

10.6.2 Exclusion lists are permanent and shared

The exclusion set — unsubscribes, withdrawals, complaints, bounces, do-not-reach — is kept, not deleted, so that a future import or a new tool cannot resurrect a contact who said stop. It is shared across channels: a person who unsubscribed from email did not thereby consent to text messages; a person who asked never to be called did not ask to be emailed instead, but the spirit of the request, and often the law, is that the business leaves them alone. And it is applied before purchase of any list, tool or agency service: the first question to an external provider is how your exclusion list is applied to their sends.

10.7 Documenting Compliance

10.7.1 The compliance file

The investigation in this chapter's opening asked one question the company could not answer. The compliance file exists so that it always can. For a small business it holds nine things, most of them produced automatically by a well-configured system:

1. The consent records hold, per contact, the type, the date, the source, the wording seen, the response and the expiry (Section 10.2).
2. The form and script wording is kept in every version of every consent request, dated, so that the wording a given contact saw can be shown.
3. The published privacy policy and governance practices are filed with the name and contact of the person in charge and the version history (P-39.1 ss. 3.1–3.2).
4. The unsubscribe and withdrawal log records each request, when it was received and when it was honored.
5. The exclusion lists and the scrubbing log are kept, including the National DNCL downloads and scrubs for any calling program.
6. The claims register lists each marketing claim with its evidence file, its date and its approver (Section 10.5).
7. Authorizations are the signed agreements for testimonials, names, photographs and cases.
8. The complaint process and its records show how a person complains, who handles it and what happened (s. 3.2).
9. The incident register and the retention schedule say what is kept how long, and what was destroyed when (ss. 3.5, 23).

An annual compliance review — half a day — reads the file against the current versions of the statutes and regulators' guidance, because the rules change: the periods, the penalties, the guidance on cookies and on influencers have all changed in the last five years. Chapter 11 puts the review in the task table; the book's bibliography records the facts and their dates in its volatile-facts register.

WHO DOES THIS Compliance in the three situations Solo: the owner is the person in charge of personal information by default (P-39.1 s. 3.1) and publishes their own title and contact; compliance is a matter of choosing tools that enforce the gate and spending one afternoon on the consent audit and the policy; a template privacy policy adapted with care is acceptable to start, reviewed by a professional when the business grows. Small team: the owner remains the person in charge or delegates in writing to whoever owns marketing; the team knows the four principles in one sentence each; the annual review is a half-day with the marketing owner and, ideally, a lawyer or privacy professional for an hour. Established: a named privacy officer (often the operations or finance lead), a documented program, an annual review with counsel, training for everyone who touches customer data, and a claims register approved before any campaign. In every case, the person in charge is named and published, and the gate is in the system, not in a memo.

Costly Mistakes

1. "They are our customers, so we can email them forever." Implied consent from a purchase lasts two years under CASL; from an inquiry, six months. Consequence: a list that turns unlawful silently, and a newsletter sent to thousands of expired consents — each a potential violation and a spam complaint.
2. Consent bundled into the terms. One ticked box covering the purchase, the terms and "marketing communications". Consequence: consent without effect under P-39.1 s. 14, unprovable under CASL s. 13, and a list built on sand.
3. Buying or scraping a list. Consequence: contacts with no provable consent, used for a purpose they were not collected for; complaints; a blocklisted domain that stops the lawful newsletter too.
4. Claims from the supplier's brochure. "Clinically proven", "#1", "fastest" — published without the adequate and proper test. Consequence: a Competition Bureau inquiry the business cannot answer, corrective notices, penalties, reprinted packaging.
5. Tracking on by default. Analytics and advertising pixels that profile visitors from the first page, with a banner that only says "we use cookies". Consequence: collection without the information and activation P-39.1 s. 8.1 requires, on every visit, from every visitor.
6. Compliance in a binder. A policy nobody reads, a suppression list in a spreadsheet, unsubscribes handled by hand when someone remembers. Consequence: the violation happens at the moment of sending, by the person in a hurry — exactly where a gate would have stopped it.

Key Takeaways

• Permission is the strategy, not only the law: consent, identification, honesty and respect for the person's choices are what make marketing work in a world of scarce attention — and what the four legal pillars require.
• Consent is recorded per contact: type, date, source, wording, expiry. Express consent lasts until withdrawn; implied consent from a purchase lasts two years, from an inquiry six months (CASL s. 10(10)); the burden of proof is on the sender (s. 13).
• Every commercial electronic message identifies the sender and offers an unsubscribe honored within ten business days (CASL ss. 6, 11); penalties reach $10,000,000 per violation for a business (s. 20(4)).
• Law 25 asks for unbundled, purpose-specific consent, information at collection, a published person in charge and governance policy, off-by-default profiling technologies, limited use, care with data leaving Québec, incident handling and retention limits (P-39.1 ss. 3.1, 3.2, 8, 8.1, 12, 14, 17, 23); penalties reach $10,000,000 or 2% of turnover administratively, $25,000,000 or 4% penally.
• Telemarketing follows the CRTC's rules: register and scrub against the National DNCL, keep an internal list, identify yourself, respect calling hours; the existing-relationship windows differ from CASL's.
• Claims need files: no misleading representations, tested performance claims, genuine regular prices, no drip pricing, disclosed material connections (Competition Act ss. 74.01, 74.1); testimonials genuine and typical, with signed authorizations.
• The outbound gate makes compliance automatic: excluded? consent valid? deliverable? relevant? compliant? — send and log, or block and record why. Exclusion lists are permanent and shared across channels and providers.
• Document it: consent records, wording versions, policy and person in charge, unsubscribe log, scrubbing log, claims register, authorizations, complaint records, incident register, retention schedule — reviewed annually against the current rules.

Your Action Plan

The action plan below presents the eight steps that turn this chapter into work, in the order in which each step consumes the output of the one before. For each step, the table gives the deliverable that must exist at the end, the timing on a small business's calendar, and the verification criterion by which anyone — a co-founder, a mentor, the owner's own future self — can tell that the step was done. The criterion column is the one to read first: a step whose criterion you cannot picture is a step you do not yet understand.

In La Machine

La Machine enforces the parts of this chapter that can be enforced by a system, and records the rest.

• Consent register (business/marketing/consent, detail at business/marketing/consent/:id) holds, for each contact, the consent type, date, source and wording — the record CASL s. 13 asks you to be able to produce — and gives an implied consent an expiry of 24 months from its documentation, while the law counts two years from a purchase and six months from an inquiry, so the expiry is typed from the event; as read on September 15, 2026, an expiry passing does not rewrite the opt-in flag until the contact is written or the application restarts. A lead magnet's consent box writes to it at capture; a form's checkbox records an answer (→ see Book N5).
• Do-not-reach (business/marketing/do-not-reach) is the permanent, shared exclusion list of Section 10.6.2 — unsubscribes, withdrawals, complaints, bounces, partner accounts, disputes — where each entry blocks the channel it names, checked before every message the product sends; the sender identification of each line of business is checked on every message, and the marketing settings page (business/marketing/settings) is a placeholder.
• Sequences (business/marketing/sequences) are the one path that sends: each message passes the gate, and excluded contacts and incomplete identification are blocked with the reason recorded; launching a campaign (campaigns) sends nothing, and the email queue page (email-queue) is a manual fallback; the email transport carries the identification and an instruction to reply to unsubscribe, a public unsubscribe writes the exclusion list, and a reply mailbox is read by a person within the legal limit.
• The marketing log (business/marketing/log) is the record of what left — who received what, and when — that the compliance file needs, and refusals are counted from extracts; powerdialer (business/marketing/powerdialer) is a placeholder that places no call.
• Forms and responses (business/marketing/forms, business/marketing/forms/responses) carry the requests, but only a lead magnet's unticked consent box documents a consent, its label kept as the proof sentence; contacts import (business/marketing/contacts/import) asks for no consent status, so an imported list has its consents documented on the Consent page before it can become a mailing list (→ see Book N4, Book N5).
• Administration and compliance functions — roles, access, backups, retention — are covered in Book N12; the website module (business/website) is where the privacy policy, the person in charge and the tracking banner are published.

What La Machine does not do: it does not write your privacy policy, name your person in charge, conduct the privacy impact assessment for data leaving Québec, or scrub against the National DNCL for you — the registration and subscription are yours (the do-not-reach list holds your internal list and the numbers you import from your scrubs); it does not judge whether a claim has an adequate and proper test behind it; and it is not a substitute for a lawyer's or privacy professional's review of your program. It gives you the gate, the records and the log; the decisions and the files are yours. → see Book N5, Book N12, Books H2, H3, H4.

Self-Check

1. Under CASL, how long does implied consent last after a purchase and after an inquiry, and who bears the burden of proving consent?
2. What three things must every commercial electronic message contain, and how quickly must an unsubscribe be honored?
3. What does "unbundled" consent mean under P-39.1 s. 14, and what must a business tell a person at the time of collection under s. 8?
4. A supplier's brochure says the product is "clinically proven". What does the Competition Act require before you repeat that claim, and who bears the burden of proof?
5. Name the five gates of the outbound gate and say which are enforced by the system and which are designed into the content.

Answers. (1) Two years after a purchase or lease (or a contract in force or expired within that period); six months after an inquiry or application (CASL s. 10(10)); the person who alleges consent has the onus of proving it (s. 13) (Section 10.2.2). (2) Identification of the sender (and of anyone on whose behalf it is sent), contact information valid for at least 60 days, and an unsubscribe mechanism (s. 6(2)–(3)); the unsubscribe must be given effect without delay and no later than ten business days (s. 11) (Section 10.2.2, The Rule). (3) Consent requested for each specific purpose, in clear and simple language, and — in writing — presented separately from any other information, so that agreeing to marketing is never buried in the terms of sale or made a condition of something else; at collection the person must be told the purposes, the means of collection, their rights of access and rectification, their right to withdraw consent, and any third parties or communication outside Québec (ss. 8, 14) (Section 10.2.3). (4) An adequate and proper test, conducted before the claim is made, that you hold and can produce — the supplier's assurance is not your test — and the burden of proving it is on the person making the representation (s. 74.01(1)(b)) (Section 10.5.1). (5) Excluded? · consent valid and unexpired? · deliverable? · relevant? · compliant? — gates 1 to 3 are enforced by the sending system from data on the record; gates 4 and 5 are designed into the segments, templates and content review (Section 10.6.1).

Chapter References

An Act to promote the efficiency and adaptability of the Canadian economy by regulating certain activities that discourage reliance on electronic means of carrying out commercial activities, S.C. 2010, c. 23 (Canada's Anti-Spam Legislation), ss. 6, 10, 11, 13, 20. Justice Laws Website. Retrieved September 7, 2026, from https://laws-lois.justice.gc.ca/eng/acts/E-1.6/

Act respecting the protection of personal information in the private sector, CQLR c P-39.1, ss. 3.1, 3.2, 3.5, 8, 8.1, 12, 14, 17, 23, 90.12, 91, 92.1. LégisQuébec. Retrieved September 7, 2026, from https://www.legisquebec.gouv.qc.ca/en/document/cs/P-39.1

Charter of the French language, CQLR c C-11. LégisQuébec. Retrieved September 7, 2026, from https://www.legisquebec.gouv.qc.ca/en/document/cs/C-11

Competition Act, R.S.C. 1985, c. C-34, ss. 74.01, 74.03, 74.06, 74.1. Justice Laws Website. Retrieved September 7, 2026, from https://laws-lois.justice.gc.ca/eng/acts/C-34/

Consumer Protection Act, CQLR c P-40.1. LégisQuébec. Retrieved September 7, 2026, from https://www.legisquebec.gouv.qc.ca/en/document/cs/P-40.1

Personal Information Protection and Electronic Documents Act, S.C. 2000, c. 5. Justice Laws Website. Retrieved September 7, 2026, from https://laws-lois.justice.gc.ca/eng/acts/P-8.6/

Telecommunications Act, S.C. 1993, c. 38, s. 41. Justice Laws Website. Retrieved September 7, 2026, from https://laws-lois.justice.gc.ca/eng/acts/T-3.4/

Ad Standards. (n.d.). The Canadian Code of Advertising Standards. Retrieved September 7, 2026, from https://adstandards.ca/code/

Canadian Radio-television and Telecommunications Commission. (n.d.). Canada's Anti-Spam Legislation (CASL): Enforcement actions. Government of Canada. Retrieved September 7, 2026, from https://crtc.gc.ca/eng/internet/anti.htm

Canadian Radio-television and Telecommunications Commission. (n.d.). National Do Not Call List: Rules for telemarketers. Retrieved September 7, 2026, from https://lnnte-dncl.gc.ca/

Commission d'accès à l'information du Québec. (n.d.). Entreprises : protection des renseignements personnels. Gouvernement du Québec. Retrieved September 7, 2026, from https://www.cai.gouv.qc.ca/

Competition Bureau. (n.d.). Deceptive marketing practices. Government of Canada. Retrieved September 7, 2026, from https://competition-bureau.canada.ca/

Godin, S. (1999). Permission marketing: Turning strangers into friends, and friends into customers. Simon & Schuster.

Office of the Privacy Commissioner of Canada. (2018, May). Guidelines for obtaining meaningful consent. Government of Canada. Retrieved September 7, 2026, from https://www.priv.gc.ca/en/privacy-topics/collecting-personal-information/consent/gl_omc_201805/

Chapter 11 — Typical Tasks, Cadence and Workflows

Everything in the first ten chapters is a decision. This chapter is the list of things somebody has to do, at what rhythm, with what result — so that the marketing system runs whether or not the owner had a good week.

What You'll Be Able to Do

• Describe the marketing operating rhythm — daily, weekly, monthly, quarterly, annual — and say what belongs at each cadence.
• Read and write a task in the La Machine format: task, objective, trigger, frequency, automation level, input, output, verification criterion, role, tools.
• Distinguish one-off setup tasks, recurring tasks and workflows, and sequence the setup over the first ninety days.
• Size the recurring workload in hours per week, and know which tasks to automate first.
• Write a verification criterion for any marketing task — observable, countable, time-bound.
• Build the annual marketing calendar: seasons, triggers, campaigns, content, reviews and compliance dates on one page.

Why It Matters

Marketing fails in small businesses less from bad strategy than from irregularity: the newsletter that skips three months, the review request nobody sends, the lead that waits until Tuesday, the consent audit that never happens. A task table with a cadence and a verification criterion for each task is what turns intention into rhythm — and what makes delegation and automation possible, because a task that is written can be handed to a person or a machine. For the owner, this chapter converts the whole book into a schedule that fits the hours actually available. For the team, it is the definition of done for every marketing task, which ends the question "did we do marketing this week?" with a number instead of a feeling.

In This Chapter

• 11.1 The Marketing Operating Rhythm
• 11.2 The Task Table in the La Machine Format
• 11.3 One-Off Tasks: The Setup
• 11.4 Recurring Tasks
• 11.5 Workflows Worth Automating First
• 11.6 Verification Criteria for Marketing Tasks
• 11.7 The Marketing Calendar

11.1 The Marketing Operating Rhythm

11.1.1 Five cadences

A marketing system runs at five cadences at once, like the rings of Figure 11.1. Each cadence has a purpose, a duration and a small set of tasks; the discipline is to keep each task at its cadence — neither skipping the weekly count because the month was busy, nor rewriting the positioning every week because a competitor posted something.

Figure 11.1 — The marketing operating rhythm: five concentric cadences — daily, weekly, monthly, quarterly, annual — with the tasks that belong to each and the time each typically takes in a solo or small-team business. Original figure.

Daily (15 minutes, mostly automatic). Leads are acknowledged and routed the moment they arrive (Chapter 9); replies in the inbox and on social channels are answered; new leads are checked for source and consent. In a well-configured system, the human part is the replies; the rest happens without anyone.

Weekly (two to four hours). The count — stages, transitions and sources for the week (Chapter 2); the list of leads past their service level (Chapter 9); one piece of content produced or scheduled (Book B5); the newsletter or the sequences checked; review requests sent for the week's completed work (Chapter 5, Chapter 7); one partner touch (Chapter 8); hours spent by channel logged (Chapter 8). The weekly slot booked in Chapter 1 is this ring.

Monthly (half a day). The review — what moved, what we tried, what we learned, what we stop and start (Chapter 2); the newsletter issue; consent expiries processed and re-permission attempts scheduled (Chapter 10); win/loss reasons read (Chapter 9); the proof file updated with the month's numbers and cases (Chapter 5); budget against actual, in hours and dollars.

Quarterly (one day). The reset — ICP fit rate, win rate and margin by segment (Chapter 3); the channel scorecard re-read with cost per customer by source (Chapter 8); the mix audit and the brand consistency audit (Chapters 6 and 7); the consent audit (Chapter 10); the persona evidence file reviewed for patterns (Chapter 4); the journey map revisited (Chapter 2); the next quarter's calendar (Section 11.7).

Annual (two days). Strategy and budget (Book B4); the positioning review (Chapter 5); the brand guide and the asset inventory (Chapter 7); the compliance review against the current statutes and regulators' guidance (Chapter 10); the tool stack and its costs (Chapter 13); the marketing calendar for the coming year.

11.1.2 Why cadence beats intensity

A system that runs at a steady cadence outperforms one that runs in bursts for three reasons. Compounding: earned channels — search, reviews, referrals, the list — grow with consistent presence and stall without it (Chapter 8). Capacity: leads generated at a steady rate can be served; leads generated in a burst arrive when the team is busiest. Learning: a weekly count over twelve weeks produces a trend; a count every three months produces an anecdote. Cadence is also what makes the system survive the owner's absence — a vacation, a busy season, an illness — because the tasks are scheduled, delegated or automated rather than remembered.

KEY CONCEPT A task without a verification criterion is not a task "Do the newsletter" is an intention. "Send the September issue to the consented list by the 15th, with one case and one call to action, reaching at least 95% of the list with no excluded contact delivered" is a task: it says what is done, by when, and how anyone can tell it was done correctly. The verification criterion — the definition of done — is what makes a task delegable to a colleague, assignable to an automation and countable in the weekly review. Every row of Table 11.1 has one; Section 11.6 shows how to write them. A task list without criteria is a to-do list; a task list with them is an operating system.

11.2 The Task Table in the La Machine Format

11.2.1 The ten columns

La Machine describes every task — in marketing as in any other function — with ten fields, so that a task can be created once, scheduled, assigned to a person or a workflow, executed and verified. The fields are: Task (a verb and an object); Objective (the output of the system it serves — awareness, demand, trust, retention, learning; or a compliance obligation); Trigger (the event or the date that starts it); Frequency (one-off, daily, weekly, monthly, quarterly, annual, or event-driven); Automation level (Manual — a person does it; Semi-auto — the system prepares or prompts, a person decides or finishes; Auto — the system does it and a person reviews the log); Input (what the task needs to start); Output (what it produces); Verification criterion (how anyone can tell it was done correctly — observable, countable, time-bound); Role (who owns it — owner, marketer, seller, assistant, or "system" for automated tasks); and Tools (the pages or tools used). The same format serves the books of Collection N. In the product itself, a plan task answers the same questions under its own names — the role is its owner, chosen from a list fixed in the code — keeps input and output in its objective and its procedure, and adds whether a person must approve each run; and since no page creates a plan task, each guide's tasks are loaded into the platform through its programming interface, a script or the assistant (→ see Book N2, Piloting Tasks, Executions and Projects, Chapter 2).

11.2.2 The table

Table 11.1 lists the tasks of the marketing system built in this book — the one-off setup tasks that create the system, the recurring tasks that run it, and the workflows that automate it. Durations are typical for a solo or small-team business; established businesses spend more time on the same tasks, not on more of them. Every task refers to the chapter that explains it. For legibility on a printed page, the ten fields are laid out in six columns: the objective follows the task name; trigger, frequency and automation level share a column; input and output share a column; role and tools share a column.

Table 11.1 — The marketing task table (format La Machine)

The table holds 46 tasks: 15 one-off setup tasks, 25 recurring tasks (3 daily, 8 weekly or fortnightly, 5 monthly, 5 quarterly, 4 annual) and 6 workflows, of which every workflow and 5 recurring tasks run at the Auto level and 9 more at Semi-auto. It is the table to load — adapted to your offer, your roles and your hours — into the task system of La Machine (see In La Machine).

11.3 One-Off Tasks: The Setup

11.3.1 Sequence, not simultaneity

The fifteen setup tasks are the ninety-day plan of Chapter 14 in operational form, and their order matters because each consumes the output of the one before: the charter frames the count; the count and the customer ranking feed the ICP; the ICP selects the interviewees for the persona; the persona and the ICP fill the positioning canvas; the positioning shapes the offer ladder and the brand promise; the persona's sources drive the channel scorecard; the funnel stages become the lead definitions; and compliance — consent audit, policy, exclusions, claims register — runs alongside from the first week, because nothing should be sent before it is done. A business that runs the tasks in parallel to "save time" produces a persona without an ICP, a website without a position and a campaign without consent — and redoes them.

11.3.2 A realistic timeline

For a solo owner with six hours a week plus an assistant's two: weeks 1–3 — S1, S2, S3, S4, S12 (charter, count, ICP, exclusions, consent audit); weeks 4–7 — S5, S13, S14 (interviews and persona; policy and tracking; claims register); weeks 8–9 — S6, S7 (positioning; offer ladder); weeks 10–11 — S8, S9, S10 (mix audit; brand promise, standards and voice; channels); weeks 12–13 — S11, S15 (lead definitions, SLA and intake; the calendar). Thirteen weeks — a quarter — and the recurring ring starts turning fully in week 3, with the count, and grows a task at a time. An established business with a marketing manager compresses the same sequence into six to eight weeks; it does not skip steps.

CASE IN POINT Thirteen weeks at a two-partner firm Fictional example; names and figures are invented. Lacoste & Nguyen, a two-partner bookkeeping and advisory firm with four staff, decided in January to run marketing as a system. Its constraints: six hours a week between the partners, two from the office coordinator, a budget of $1,200 CAD a month. Weeks 1–3: charter (target: 8 qualified inquiries a month), funnel and count installed, ICP written from 71 clients ranked by margin (owner-managed trades and construction firms with 5 to 40 employees; exclusions: personal tax, start-ups under a year), consent audit (1,340 contacts: 610 express, 380 implied and valid, 350 none — moved to excluded). Weeks 4–7: nine interviews, a persona ("Dispatch-desk Danielle", the owner's spouse who runs the books), privacy policy and person in charge published, claims register opened (two claims removed for lack of evidence). Weeks 8–9: positioning as "the bookkeeping firm that gives trades owners a monthly cash-flow picture in one page", offer ladder with a $290 "one-page cash-flow diagnostic" as entry rung. Weeks 10–13: mix audit (the website said "full-service accounting"; rewritten), brand promise ("your numbers, by the 10th, explained in twenty minutes"), channels (accountant-adjacent partners — an insurance broker and a supplier's rep —, the monthly one-page newsletter, search advertising on "bookkeeping for contractors"), lead definitions and SLA, calendar. By the end of the second quarter the recurring ring ran on five hours a week; inquiries averaged eleven a month, of which seven fit; three came from the diagnostic offer, four from partners.

11.4 Recurring Tasks

11.4.1 The weekly time budget

The recurring tasks of Table 11.1 fit a small business only if their hours are counted honestly. Table 11.2 sizes them for three situations; the numbers are typical, not prescriptive, and the point is the shape: the daily ring is nearly free once automated, the weekly ring is where the hours go, and the monthly and quarterly rings are half-days that are scheduled or do not happen.

Table 11.2 — Typical hours per week for the recurring ring

DO THE MATH What automation buys back Illustrative figures. Before automation, a solo owner's recurring marketing takes about 11 hours a week: 2 hours acknowledging and routing leads by hand and chasing follow-ups, 1.5 hours sending review requests and thank-yous, 1 hour checking who may still be emailed, 1.5 hours assembling and sending the newsletter, 1 hour posting it to social channels, and 4 hours on the count, content, partners and reviews. Automating W1 to W6 removes or halves the first five items: leads acknowledged and routed in seconds, review requests sent on completion, consent expiries processed at month end, the newsletter drafted from a template and posted automatically. The recurring load falls to about 6 hours a week — 5 hours saved, 260 hours a year. At a loaded cost of $75 CAD an hour for the owner's time, that is $19,500 a year, against tool costs that rarely exceed $1,500 a year at small-business pricing (as of 2026, verify on the vendors' pages). More important than the money: the five hours saved are the hours that make the weekly count, the content and the partner touch actually happen — the tasks that produce customers.

11.4.2 Daily, weekly, monthly in practice

Daily work is a fifteen-minute habit at a fixed time — replies, a glance at the task list for leads past deadline — and it is the ring most owners skip when busy, which is why the acknowledgment and routing must be automatic: the buyer's first hour cannot depend on the owner's day. Weekly work lives in the slot booked in Chapter 1 and follows the same order every week — count first (fifteen minutes), then deadlines and recycles, then the content piece, then review requests and the partner touch, then the queue check — so that the count is never displaced by the content. Monthly work is a half-day, scheduled in the calendar for the year, whose centrepiece is the review (Chapter 2) and whose deliverables are the newsletter issue and the updated proof file. The quarterly and annual rings are days blocked in advance; a quarterly reset that is "postponed to next month" has, in practice, been canceled.

WHO DOES THIS Roles across the rings Solo: the owner holds every role; the system holds the Auto tasks; an assistant or a freelancer can take R3, R6, R12 and R15 as they are written, because each has a verification criterion. Small team: the marketer owns the weekly and monthly rings and the compliance tasks; the seller owns R2, R5 and R10; the owner owns the quarterly and annual rings and attends the fortnightly review. Established: the marketing manager owns the table, assigns tasks to specialists and agencies with the same criteria, and reports the rings' health monthly; the owner reads the quarterly reset. In every situation, the verification criterion is what is delegated, not the task name — and the Auto tasks are reviewed by a person weekly, because an automation that fails silently fails completely.

11.5 Workflows Worth Automating First

11.5.1 The six

Not everything should be automated; six workflows should be, first, because they are frequent, rule-based, time-critical or legally required — and because a human doing them by hand does them late.

1. Lead acknowledgment and routing (W1). The speed-to-lead standard of Chapter 9 cannot be met by a person checking an inbox; the workflow sends the acknowledgment within minutes, assigns the owner by rule and creates the task with its deadline.
2. Review request after completion (W2). The moment of satisfaction passes quickly (Chapter 5); a request triggered by the completion event, with a reminder a week later, produces reviews at several times the rate of requests remembered.
3. Unsubscribe and exclusion enforcement (W3). This is a legal obligation with a ten-business-day maximum under CASL s. 11 (Chapter 10), and one that must apply across every channel and tool; a person doing it by hand will one day miss one.
4. Consent expiry (W4). Consent expires two years after a purchase and six months after an inquiry (CASL s. 10(10)): these are dates a system computes and a person forgets; the workflow makes the contact unsendable and opens a lawful re-permission task.
5. "Not now" nurture (W5). This is the most valuable recycle reason (Chapter 9): the workflow sends the check-back on the date the buyer named, and creates a task the moment they reply.
6. Newsletter production and distribution (W6). The workflow pre-fills the template with the month's case and call to action, sends it on approval and posts it to the social channel automatically — so that the newsletter's threshold of effort (Chapter 8) is met on a bad month as well as a good one.

11.5.2 What not to automate yet

Three families of tasks stay manual in a small business, at least at first: the conversations — sales calls, interviews, partner touches, replies to complaints — because their value is the human attention; the judgments — the ICP, the positioning, the channel decisions, the review of blocked messages — because they commit the company; and the creation of the first version of anything — the persona, the case, the hierarchy — because automation reproduces, it does not discover. Automation then takes the repetition of what humans decided: the persona's sequence, the case's distribution, the hierarchy's templates. Book J7 (Business Process Automation) and Book N3 (Workflows, Automation and Human Approval) give the method for choosing and building workflows, including the approval steps that keep a human in the loop where the risk requires it.

WATCH OUT Automation that fails silently An automated workflow that stops — a broken integration, an expired credential, a changed field — produces no error a customer will report; it produces silence: leads unacknowledged, reviews unrequested, expiries unprocessed. Every Auto task in Table 11.1 therefore has a weekly human check of its log (R9), and the two most critical — acknowledgment and exclusion enforcement — get a test entry every week: a test lead that must receive the acknowledgment and appear as a task, a test unsubscribe that must block. The check takes ten minutes; the silent failure it catches would have cost the month.

11.6 Verification Criteria for Marketing Tasks

11.6.1 Three properties

A verification criterion answers the question how would anyone — including a person who did not do the task — know it was done correctly? It has three properties. Observable: it refers to something that can be seen or read — a sent message in the log, a filled field, a published page — not to a state of mind ("the team is aligned") or an intention ("we focused on quality"). Countable: it has a number or a yes/no — "sent to 100% of the consented list", "five strangers repeat the position", "no lead past deadline at day's end". Time-bound: it says by when — "within 5 minutes", "by the 15th", "before the quarter ends". A criterion with the three properties can be checked in a minute, delegated without a meeting and automated where the task allows.

11.6.2 Bad, better, best

Table 11.3 shows the same tasks with criteria that fail and criteria that work.

Table 11.3 — Rewriting verification criteria

11.6.3 Criteria as the currency of delegation

The criterion is what a task is once it leaves the owner's head. A colleague given "do the newsletter" will produce their idea of a newsletter; given the working criterion, they will produce yours. An agency given "run our advertising" will optimize what the platform shows; given "leads created in our contact base with the paid source, at a cost per customer below $600 CAD, reported weekly", it will optimize what you need (Chapter 12). And an automation can only be built from a criterion — the workflow is the criterion made executable. This is why La Machine's task format carries the criterion as a required field, and why the books of this library close every chapter with an action plan whose steps have one.

11.7 The Marketing Calendar

11.7.1 What the calendar holds

The marketing calendar is one page per year — twelve columns, or a wheel — on which five layers are placed: seasons and triggers (when your buyers' triggers cluster: year-end for accountants, spring for landscapers, September for anything involving school or hiring); campaigns (the two to four concentrated efforts of the year — a launch, a seasonal offer, an event — each with its channel, offer and dates); content (the theme of each month's newsletter and content pieces, drawn from the persona's questions and the trigger calendar); reviews and resets (the monthly half-days, the quarterly days, the annual two days, blocked in advance); and compliance dates (the annual compliance review, the consent audits, the DNCL subscription renewal for businesses that call, trademark or domain renewals, and the dates when a rule you rely on changes). The calendar is the visible form of the cadence: a month with nothing on it is a month in which the system stopped.

METHOD Building the annual marketing calendar (two hours) When to use it: at the annual planning; when a new channel or offer is added. Inputs: the persona's triggers (Chapter 4), the channel plan (Chapter 8), the offer ladder (Chapter 6), the hours budget (Section 11.4), last year's counts by month (Chapter 2). Deliverable: one page, twelve months. Step 1 — Triggers and seasons (20 min). From the persona cards and last year's leads by month, mark when your buyers' triggers cluster and when your capacity is lowest. The calendar must generate leads before the trigger season and hold back when capacity is full. Step 2 — Fixed dates first (15 min). Block the monthly reviews, the quarterly resets, the annual review, the compliance dates and any renewal. These are immovable; everything else fits around them. Step 3 — Campaigns (30 min). Choose two to four concentrated efforts, each with an offer from the ladder, a channel from the plan, a start and end date and a target expressed as a transition rate or a count. Place them ahead of the trigger seasons. Step 4 — Content themes (30 min). Give each month a theme from the persona's questions by stage; the newsletter's case and the content piece of each week follow the theme. A month's theme is one sentence a buyer would recognize. Step 5 — Hours check (15 min). Add up the hours the calendar implies per week, month by month; where a month exceeds the budget, remove or move something. Never keep a month that exceeds the hours — it is the month the system will skip. Step 6 — Load and share (10 min). Enter the recurring tasks and the campaign tasks into the task system with their criteria; share the one-page calendar with everyone who touches a customer. Pitfalls: a calendar of campaigns without the recurring ring; campaigns placed during the trigger season instead of before it; a calendar nobody can see.

11.7.2 Reading the calendar during the year

The calendar is a plan, and the monthly review (Chapter 2) is where it meets reality: a campaign that underperforms is stopped or fixed at the review, not at year-end; a month whose hours were exceeded is examined for the task that consumed them; a trigger season that arrived early moves the next campaign. Two rules keep the calendar honest. Change it in writing — a moved campaign is re-dated on the page with a note, so that the annual review can see what was planned and what happened. Protect the rings — campaigns are added around the recurring tasks, never in place of them; a business that suspends the newsletter to run a campaign has traded a compounding asset for a burst.

QUÉBEC & CANADA NOTE Compliance dates that belong on the calendar (as of September 2026) Several obligations of Chapter 10 recur and are easily forgotten without a calendar entry. The annual compliance review against the current versions of CASL, P-39.1, the Competition Act, the Consumer Protection Act and the Charter of the French language — the statutes and regulators' guidance change, and the review is where the change is caught. The quarterly consent audit, with the monthly processing of implied-consent expiries (two years after a purchase, six months after an inquiry — CASL s. 10(10)). The National Do Not Call List subscription, which a business making marketing calls to consumers must keep current for the area codes it calls, with regular scrubs of its lists under the CRTC's Unsolicited Telecommunications Rules. Trademark renewal every ten years under the Trademarks Act and domain renewals annually, in the business's name (Chapter 7). The review of the published privacy policy and of the person in charge (P-39.1 ss. 3.1–3.2) whenever a tool, a third party or a purpose changes — and at least annually. Put each on the calendar with an owner; the cost of the entry is a minute, the cost of the omission is Chapter 10.

Costly Mistakes

1. Tasks without criteria. "Do the newsletter", "follow up leads", "post on social". Consequence: everyone's idea of done; nothing delegable; nothing automatable; the weekly question "did we do marketing?" answered by a feeling.
2. Setup tasks run in parallel. The website rebuilt before the positioning, the persona written before the ICP, the campaign sent before the consent audit. Consequence: each artefact redone once the one it depended on exists — at twice the cost.
3. The weekly count displaced by the content. The slot spent on a post; the count "next week". Consequence: no trend, no leaks located, decisions by anecdote — the learning loop never closes.
4. Bursts instead of cadence. Marketing when sales dip, silence when busy. Consequence: leads arriving when the team cannot serve them, earned channels stalling in the silence, no compounding.
5. Automation that fails silently. A stopped workflow nobody checks. Consequence: a month of unacknowledged leads or unprocessed expiries discovered when a customer, or a regulator, points it out.
6. The quarterly reset "postponed". Consequence: an ICP two years old, a channel plan never re-scored, a mix audit never run — the system running on decisions taken for a business that no longer exists.

Key Takeaways

• Marketing runs at five cadences — daily, weekly, monthly, quarterly, annual — and each task belongs to one; cadence beats intensity because earned channels compound, capacity can absorb a steady flow, and weekly counts produce trends.
• A task without a verification criterion is not a task. The criterion — observable, countable, time-bound — is what makes a task delegable, automatable and countable.
• The La Machine task format has ten fields: task, objective, trigger, frequency, automation level, input, output, verification criterion, role, tools; Table 11.1 holds forty-six marketing tasks in it — fifteen setup tasks, twenty-five recurring tasks, six workflows.
• Setup runs in sequence over about thirteen weeks, each task consuming the output of the one before; compliance runs alongside from week one.
• The recurring ring fits about six hours a week in a solo business, eleven in a small team, twenty-five in an established one — if the six workflows are automated first: acknowledgment and routing, review requests, exclusion enforcement, consent expiry, "not now" nurture, newsletter production.
• Do not automate conversations, judgments or first versions; automate the repetition of what humans decided — and check every automation's log weekly, with a test entry for the critical ones.
• The calendar holds five layers — triggers and seasons, campaigns, content themes, reviews and resets, compliance dates — built in two hours, checked against the hours budget, changed in writing, and never allowed to displace the recurring ring.

Your Action Plan

The action plan below presents the seven steps that turn this chapter into work, in the order in which each step consumes the output of the one before. For each step, the table gives the deliverable that must exist at the end, the timing on a small business's calendar, and the verification criterion by which anyone — a co-founder, a mentor, the owner's own future self — can tell that the step was done. The criterion column is the one to read first: a step whose criterion you cannot picture is a step you do not yet understand.

In La Machine

This chapter is the one that loads most directly into the platform, because the task format is La Machine's own.

• Plan tasks (Control center → Plan tasks, /taches) hold every row of Table 11.1 once it is loaded — no page creates a plan task, so the table goes in through the programming interface, a script or the assistant — and the verification criterion is required at creation, so that a task cannot be created without its definition of done, although the Mark as done button does not check it; Runs (/executions) record each run of a task with its state, its times and the result reported, which is how the weekly review sees, for example, that the count was done on Monday and the newsletter sent on the 15th (→ see Book N2, Piloting Tasks, Executions and Projects).
• Workflows (→ see Book N3, Workflows, Automation and Human Approval) implement W1 to W6 as automations with the approval steps you choose: the acknowledgment and routing on lead creation, the review request on completion, the exclusion enforcement on unsubscribe or bounce, the consent expiry by date, the "not now" check-back, the newsletter draft on the 10th. The messages they send appear in the marketing log (business/marketing/log), which records what left and is the record the weekly check (R9) reads.
• Sequences, campaigns and templates (business/marketing/sequences, campaigns, templates) are the message side of the workflows; the consent register and do-not-reach (business/marketing/consent, do-not-reach) are the compliance side, applied automatically to every send.
• Productivity tasks (productivity/tasks, where crm/tasks redirects) carry the lead-related tasks (R1, R2, R5), each attached to a contact or a company with a scheduled date and an assignee typed as free text, and without recurrence or criterion, while crm/intake is the client profile form rather than a lead queue; responses, tags and segments supply the weekly count (R4), since funnels (business/marketing/funnels) keep no stage counts.
• The Dashboard (Control center → Dashboard, /tableau-de-bord) shows the progress of the plan's tasks by wave, owner, status and chain, across functions — the marketing ring alongside the others — and across every business of the deployment; it computes the runs of the last fourteen days without displaying them, so whether the cadence held this week is read on Runs (/executions).

What La Machine does not do: it does not decide your cadences, write your criteria or choose which tasks to automate — Table 11.1 is a starting point you adapt, and the criteria are yours; it does not replace the weekly human check of the automation logs, which is a task in the table precisely because automation fails silently; and it does not hold the annual calendar as a document — the calendar is a page you build, from which the tasks are loaded. Book N2 shows tasks and executions page by page; Book N3 shows how to build the six workflows.

Self-Check

1. Name the five cadences of the marketing operating rhythm and one task that belongs to each.
2. What are the ten fields of a task in the La Machine format, and which one makes a task delegable and automatable?
3. Why must the fifteen setup tasks run in sequence rather than in parallel? Give one example of a dependency.
4. Which six workflows should a small business automate first, and which three families of tasks should stay manual at first?
5. Rewrite "leads followed up" as a working verification criterion, and say which three properties it now has.

Answers. (1) Daily — acknowledge and route leads (automatic); weekly — the count; monthly — the review and the newsletter; quarterly — the reset (ICP numbers, channel scorecard, audits); annual — strategy, budget, compliance review (Section 11.1.1). (2) Task, objective, trigger, frequency, automation level, input, output, verification criterion, role, tools; the verification criterion (Sections 11.2.1 and 11.6.3). (3) Because each consumes the output of the one before — for example, the persona's interviewees are selected from the ICP, so a persona written before the ICP describes the wrong people; the positioning canvas is filled from the persona and the ICP; the campaign requires the consent audit (Section 11.3.1). (4) Lead acknowledgment and routing; review request after completion; unsubscribe and exclusion enforcement; consent expiry; "not now" nurture; newsletter production and distribution. Conversations, judgments and first versions stay manual (Section 11.5). (5) "No lead past its service level at 17:00; every lead has six logged attempts over ten business days before recycle; every recycle has a reason" — observable (the log), countable (zero past deadline; six attempts), time-bound (17:00; ten business days) (Section 11.6).

Chapter References

An Act to promote the efficiency and adaptability of the Canadian economy by regulating certain activities that discourage reliance on electronic means of carrying out commercial activities, S.C. 2010, c. 23 (Canada's Anti-Spam Legislation), ss. 10, 11. Justice Laws Website. Retrieved September 7, 2026, from https://laws-lois.justice.gc.ca/eng/acts/E-1.6/

Act respecting the protection of personal information in the private sector, CQLR c P-39.1, ss. 3.1, 3.2. LégisQuébec. Retrieved September 7, 2026, from https://www.legisquebec.gouv.qc.ca/en/document/cs/P-39.1

Trademarks Act, R.S.C. 1985, c. T-13. Justice Laws Website. Retrieved September 7, 2026, from https://laws-lois.justice.gc.ca/eng/acts/T-13/

Canadian Radio-television and Telecommunications Commission. (n.d.). National Do Not Call List: Rules for telemarketers. Retrieved September 7, 2026, from https://lnnte-dncl.gc.ca/

Chapter 12 — Expertise, Roles and When to Get Help

Nobody has all the skills modern marketing requires — not the owner, not the agency, not the marketing manager. The question is never "how do I learn everything?" but "which skills must stay in the house, which can be bought, and which can be handed to a machine?"

What You'll Be Able to Do

• Name the seven skill families of modern marketing and rate your business honestly on each.
• Decide, as a solo owner, what you do yourself, what you automate and what you buy — and in what order.
• Recognize the signals that it is time for a first marketing hire, and write the role from the task table rather than from a template.
• Structure the marketing function in an established business: manager, specialists, suppliers, and the metrics that hold them together.
• Compare freelancers, agencies and fractional executives on deliverables, fees, control and fit — and choose by situation.
• Brief and manage external help so that you keep your accounts, your data and your learning.
• Say what La Machine and AI take over, and what remains a human's job.

Why It Matters

The marketing budget of a small business is mostly people — the owner's hours, an employee's salary, a supplier's fees — and the most expensive mistakes are staffing mistakes: the agency that produces a brochure and owns the advertising account; the hire made for a channel instead of a system; the owner who does everything and therefore does marketing last. Chapter 11 said what must be done and how often; this chapter says who should do it, in each of the three situations, and how to buy help without losing control. For the owner, it is the difference between a marketing function that grows with the business and one that has to be rebuilt every time a person leaves. For the team, it is clarity about roles — and about what the machine does so that they do not have to.

In This Chapter

• 12.1 The Skills of Modern Marketing
• 12.2 Solo: What You Do Yourself and What You Automate
• 12.3 Small Team: The First Marketing Hire
• 12.4 Established: Marketing Manager, Specialists, Agencies
• 12.5 Freelancers, Agencies and Fractional Executives
• 12.6 Briefing and Managing External Help
• 12.7 What La Machine and AI Take Over

12.1 The Skills of Modern Marketing

12.1.1 Seven families

The work described in Chapters 1 to 11 draws on seven families of skill. No person masters all seven; a business needs access to each, at the level its situation requires.

1. Strategy is defining the market, the ICP, the personas, the position, the offer ladder and the channel plan (Chapters 3 to 8), and deciding what not to do: the skill of choosing.
2. Content and writing is turning the hierarchy into pages, emails, cases, guides, scripts and posts in the brand's voice (Chapters 5 and 7; Books B5, B6): the skill of saying one thing clearly to one reader.
3. Data and measurement is defining stages, counting transitions, recording sources, reading the funnel, running tests and computing the cost per customer (Chapters 2 and 9; Book B7): the skill of knowing.
4. Channels is the operating knowledge of each channel — search, email, social, advertising platforms, events, partners (Chapter 8; Collections C and D): the skill of being present where the persona is, competently.
5. Automation and tools is configuring the contact base, the sequences, the workflows and the integrations, and keeping one source of truth (Chapters 11 and 13; Books B8, J7): the skill of making the system run without hands.
6. Design and production is the visual identity, the layouts, the images, the video and the physical evidence of the brand (Chapters 6 and 7; Book B3): the skill of making the promise look like itself.
7. Compliance and ethics is consent, privacy, claims, language, the outbound gate and the compliance file (Chapter 10; Collection H): the skill of never having to explain a violation.

DEFINITION Marketing skill families and proficiency levels The skill families are the seven groups of competence a marketing system requires — strategy, content, data, channels, automation, design, compliance. A business rates its access to each on three proficiency levels: aware (knows what the skill is for and can brief someone who has it), capable (can execute the recurring tasks of Chapter 11 in that family to their verification criteria) and expert (can design the system, teach it and handle the exceptions). A small business needs to be aware in all seven, capable in strategy, data and compliance — the families that commit the company — and may buy capability in content, channels, design and automation until its volume justifies hiring it. Expertise is rented until it is used weekly.

12.1.2 Which skills must stay in the house

Three families cannot be outsourced without outsourcing the business. Strategy — whom to serve and what to promise — is the owner's decision, as every chapter of this book has insisted; a supplier can facilitate the exercise and challenge the answers, never make them. Customer knowledge — the interviews, the quotes, the win/loss reasons — must be heard by the people who decide, or the persona becomes a supplier's document about strangers. Compliance ownership — the person in charge of personal information, the consent records, the claims register — is a legal responsibility of the enterprise (P-39.1 s. 3.1) that a supplier can support but not carry. Everything else — writing, design, channel operation, automation configuration — can be bought, provided it is briefed from the in-house strategy and executed on assets the business owns (Section 12.6).

KEY CONCEPT Own the strategy and the customer knowledge; standardize execution; buy the rest The division of labor that works for a small business has three parts. The owner owns the choices — ICP, position, promise, channels, budget — and the direct knowledge of customers that informs them. Execution is standardized into the tasks of Chapter 11, each with a verification criterion, so that it can be handed to a colleague, a supplier or a workflow without a meeting. The rest is bought — by the hour, by the project, by the month — from people who do it weekly for many clients and are therefore better and faster than an owner doing it monthly for one. A business that reverses the division — buys the strategy, keeps the execution — pays a supplier to decide who it is and spends its own scarce hours posting.

Table 12.1 applies that division of labor to the seven skill families of Section 12.1.1. Each row is a family — strategy, content and writing, data and measurement, channels, automation and tools, design and production, compliance and ethics — and each column is one of the three situations of this book: the solo owner with automation, the small team with a part-time marketer and a seller, and the established business with a marketing manager, specialists and suppliers. Each cell says who holds the skill in that situation and what they do with it; read across a row to see how a family moves from the owner's hands to a team's, and down a column to see what one situation asks of its people.

Table 12.1 — Skill families and who holds them in the three situations

12.2 Solo: What You Do Yourself and What You Automate

12.2.1 The owner's four hours

The solo owner has, realistically, four to six hours a week for marketing (Chapter 11). The allocation that works: one hour of strategy and knowledge — reading the count, a partner conversation, a customer interview every month; two hours of content in your own voice — the case, the newsletter, the answer to the question three prospects asked; one hour of review — the queue, the blocked messages, the automation logs, the leads past deadline; and the remainder for the quarter's test. Everything repetitive is automated (the six workflows of Chapter 11); everything requiring a skill you lack and use monthly is bought by the hour (a designer for the brand guide, a writer to polish, an integrator to configure the first workflows).

12.2.2 What not to do yourself

Three things a solo owner should not do personally, even when able: design the identity (a designer produces in two days what an owner produces badly in two weeks; the owner decides, the designer executes); configure paid advertising from scratch (the platforms' auctions punish beginners; a freelancer's setup with a cost-per-customer target and the account in your name pays for itself in the first month or is stopped); and write the privacy policy alone for a business that collects sensitive data (a template and an hour with a professional beat a week of the owner's research). The test for buying help: would an expert doing this weekly do it in a fraction of my time, and does it commit the company's choices? If yes to the first and no to the second, buy it.

DO THE MATH The owner's hour versus the freelancer's hour Illustrative figures. An owner bills or produces about $150 CAD an hour of value in the business. Configuring a first automation — acknowledgment, routing, review requests — takes the owner about twelve hours of learning and doing; a freelancer who configures La Machine workflows weekly does it in three hours at $90 an hour. Owner's cost: 12 × $150 = $1,800 of displaced value; freelancer's cost: $270, plus one hour of the owner's briefing ($150). The freelancer is six times cheaper — and the owner spends the eleven saved hours on the interviews and the case that only the owner can do. The same arithmetic applies to the brand guide (designer, two days), the paid-search setup (specialist, four hours) and the privacy policy (professional, one hour of review). The owner's hour is the scarcest resource in the business; spending it on tasks an expert does faster is the most common hidden cost in small-business marketing.

12.3 Small Team: The First Marketing Hire

12.3.1 The signals

A first marketing hire is justified when the recurring ring exceeds what the owner and automation can carry and when the system exists to hand over. Four signals, together: the recurring tasks of Chapter 11 consume more than ten hours a week and are being skipped; leads are arriving faster than they are worked (leads past deadline in every weekly review); the owner is the bottleneck for content (cases and newsletters wait weeks for the owner's time); and the ICP, positioning, hierarchy and task table exist in writing — because a hire without them has nothing to execute and will invent a strategy. A hire made before the system exists produces a busy person and no results; a hire made after it produces results in the first quarter.

12.3.2 The profile

The first hire in most small businesses is a marketing generalist — a coordinator or manager who can run the recurring ring end to end: the count, the content from the hierarchy, the newsletter, the review requests, the partner touches, the queue checks, the compliance tasks, and the briefing of freelancers for design, video and paid channels. The profile is capable (Section 12.1.1) in content, data and automation, aware in the rest, and comfortable with tools. It is not a specialist in one channel — a business with one channel specialist and no generalist has a channel and no system — and it is not a strategist: the owner remains the strategist, and a good generalist prepares the data and the options for the owner's decisions.

Write the job from Table 11.1, not from a template: the recurring tasks the hire will own, with their verification criteria, are the job description; the hours they represent (Table 11.2) are the workload; the setup tasks still pending are the first ninety days. A candidate who reads the task table and asks good questions about the criteria is the right candidate.

12.3.3 Compensation and status

Salary ranges vary by region, experience and sector, and they change; consult the Government of Canada's Job Bank wage data for the relevant occupational group — professional occupations in advertising, marketing and public relations — and its regional breakdowns before setting a range (Government of Canada, 2026). As an illustration only, as of 2026 a part-time or junior marketing coordinator in Québec typically costs a small business in the order of $45,000 to $60,000 CAD a year full-time equivalent before payroll costs, and an experienced generalist manager $65,000 to $95,000 — figures to verify against current data. Add the employer's payroll contributions and benefits (commonly 12% to 20% in Québec, depending on the plan) to compare with a supplier's fee (Section 12.5). Many small businesses start with a part-time hire — three days a week — or with a fractional arrangement (Section 12.5) and convert to full-time when the ring and the volume justify it.

QUÉBEC & CANADA NOTE Employee or contractor, and the language of work (as of September 2026) Two rules matter when the first marketing role is created. Status. Whether the person is an employee or a self-employed contractor is not a matter of what the contract says but of the facts — control over the work, ownership of tools, chance of profit and risk of loss, integration into the business — assessed by the Canada Revenue Agency and Revenu Québec; misclassifying an employee as a contractor exposes the business to unpaid source deductions, contributions and penalties (Canada Revenue Agency, n.d.). A marketing coordinator who works your hours, with your tools, only for you, is almost certainly an employee, with the rights of the Act respecting labour standards, CQLR c N-1.1 (→ see Book H7, Employment Law Essentials (Québec and Canada)). Language. In Québec, the Charter of the French language, CQLR c C-11 gives workers the right to carry on their activities in French and regulates job postings and the requirement of another language; a posting that requires English must state the reasons and the employer must have taken reasonable means to avoid the requirement. → see Book I3 (Recruiting, Hiring and Onboarding) for the process; Book H7 for the rules.

CASE IN POINT The first marketing hire that worked, and the one that did not Fictional example; names and figures are invented. Two firms of similar size hired their first marketer in the same year. Ferland Mécanique (industrial maintenance, 26 employees) hired a "social media specialist" because the owner believed the company should be on video. Eight months later: 140 posts, 900 followers, four inquiries traceable to social, and a specialist who had never seen an ICP because none existed; the role was ended amicably at a cost of about $52,000 CAD including recruitment and severance. Rousseau Sécurité (alarm and access systems, 31 employees) spent the previous quarter on the setup tasks of Chapter 11 — ICP, persona, positioning, channel plan, task table — then hired a part-time marketing coordinator (three days a week) with a job description drawn from the twenty-five recurring tasks and their criteria. In the first quarter the coordinator ran the count, the newsletter, the review requests and the partner touches, briefed a freelancer for search advertising with a cost-per-customer target of $700, and prepared the monthly review. Inquiries rose from nine to twenty-two a month, of which fifteen fit the ICP; the owner's marketing hours fell from ten to three a week. The difference was not the person; it was whether there was a system to hand over.

12.4 Established: Marketing Manager, Specialists, Agencies

12.4.1 The structure

Above roughly fifty employees, or a marketing budget that supports two or more people, the function acquires a structure: a marketing manager who owns the system — the hierarchy, the task table, the channel plan, the metrics, the compliance program — and reports the six numbers monthly; specialists for the channels that carry the most volume (content, paid, automation), in-house when the work is weekly and continuous, through an agency or freelancers when it is periodic or requires expensive expertise; a designer in-house or on retainer; and a marketing operations capability — the person who keeps the stack, the integrations and the data clean (Chapter 13). Sales and marketing co-own the SLA (Chapter 9); the owner or general manager reads the six numbers and arbitrates definitions and budget.

12.4.2 What changes, and what does not

Three things change with scale. Metrics replace observation: the owner no longer sees every lead; the dashboard and the fortnightly review are how the system is known. Governance replaces memory: the brand guide, the hierarchy, the task criteria and the compliance file are the institutional knowledge, and they are maintained as documents with owners and versions. Suppliers multiply: the established business may have an agency for paid channels, a video producer, a PR consultant, a web developer — each briefed from the same hierarchy and measured against the same cost-per-customer logic. What does not change: the owner still decides whom to serve and what to promise; the customer knowledge still has to reach the people who decide; and the marketing manager's first job is to keep the system's cadence, not to launch campaigns.

WHO DOES THIS The marketing manager's job in one paragraph The marketing manager of an established small business owns four things: the system (hierarchy, task table, channel plan, calendar — current and followed), the numbers (the count, the six SLA metrics, cost per customer by source — reported monthly, read fortnightly with sales), the suppliers (briefed from the hierarchy, measured on your numbers, holding no account or data you do not own) and the compliance program (with the person in charge). The manager proposes strategy changes with evidence and executes the owner's decisions; they do not make the choice of whom to serve. The best predictor of a good manager is not campaign creativity but whether the recurring ring runs every week without the owner asking.

12.5 Freelancers, Agencies and Fractional Executives

12.5.1 Three kinds of external help

External help comes in three forms, each with a fit. A freelancer is one person selling one or two skills — writing, design, paid-search setup, automation configuration, video — by the hour or the project; fast to engage, inexpensive per hour, dependent on your brief and your management. An agency is a team selling a bundle — strategy, creative, channels, reporting — usually on a monthly retainer with a minimum term; it brings breadth and process, costs more, and tends to work on its own accounts and templates unless the contract says otherwise. A fractional executive — a fractional chief marketing officer or marketing director — is a senior person selling one or two days a week of judgment: strategy facilitation, the hierarchy, the channel plan, the hiring and management of freelancers and agencies; the right form when the business needs senior thinking it cannot afford full-time and has execution capacity to direct. Table 12.2 compares the three forms on their typical deliverables, their fees (illustrative Canadian ranges, dated), the situation each fits, and the control and risk questions to settle before signing.

Table 12.2 — External help compared (fee ranges are illustrative, Canadian market, as of 2026 — verify before budgeting)

12.5.2 Choosing by situation

Solo: freelancers, by the hour, for design, automation setup and paid-channel setup; a writer to polish if the owner's hours are the constraint; no agency — the owner cannot yet brief or manage one, and the retainer exceeds the useful spend. Small team: a generalist hire plus freelancers for specialty and volume; a specialized agency for one paid channel above threshold, measured on cost per customer; a fractional marketing director for two quarters to build the system if the owner cannot. Established: a marketing manager who manages agencies and freelancers; a full-service or specialized agency where production capacity is needed; a fractional CMO to challenge strategy annually. In every case the test is Section 12.1.2: the strategy, the customer knowledge and the compliance ownership stay inside.

WATCH OUT Agencies that own your accounts The most expensive supplier failure is discovered at the end of the relationship: the advertising accounts, the analytics property, the domain, the social pages, the email tool and its list, the design source files — created by the agency, in the agency's name, on the agency's credentials. The business that wants to leave finds it must start again, or pay to stay. The rule is absolute and belongs in the contract before the first invoice: every account, property, domain, list, credential and source file is created in the business's name, with the business holding the owner or administrator role, and the supplier granted access that can be revoked. Add to it: your tracking on your pages, so that the numbers do not leave with the supplier; deliverables in editable formats; an exit clause with a handover period. Book H1 (Contracts and Commercial Agreements) gives the clauses; the checklist in Section 12.6 gives the questions.

12.6 Briefing and Managing External Help

12.6.1 The brief

External help is only as good as the brief, and a good brief is mostly documents you already have. It contains: the ICP (Chapter 3) and the persona cards (Chapter 4) — who the work is for; the positioning statement and messaging hierarchy (Chapter 5) — what it must say; the brand guide and voice guide (Chapter 7) — how it must look and sound; the task and its verification criterion (Chapter 11) — what "done" means; the budget and the deadline; the metrics by which the work will be judged — your numbers: leads created in your base with the source, cost per customer, transition rates (Chapters 2, 8); and the rules — consent, claims with files, disclosure, French where required (Chapter 10). A supplier who receives this brief produces work that fits the system; a supplier who receives "make us a campaign" produces a campaign for a generic business and invoices for the strategy they had to invent.

METHOD The one-page brief (one hour, reused every time) When to use it: before engaging any freelancer, agency or fractional executive; before each new project with an existing supplier. Deliverable: one page plus attachments. Section 1 — Context (5 lines). The business, the position (the statement of Chapter 5), the situation that prompts the work. Section 2 — Audience (3 lines + attachments). The ICP segment and the persona(s) this work addresses; attach the cards. Section 3 — The task and the definition of done (5 lines). What is to be produced, by when, and the verification criterion in Chapter 11's format; the stage of the journey it serves and the call to action it uses. Section 4 — Message (attachments). The hierarchy: which pillar leads for this persona, which proofs are available (attach the files), the voice guide. Section 5 — Measures (3 lines). The numbers the work will be judged on, where they are read (your funnel, your log), and when (weekly report, monthly review). Section 6 — Rules (5 lines). Ownership: every account, asset and file in the business's name; compliance: consent, claims with files, disclosures, French; approval: who signs off, and what needs approval before publication (claims, prices, anything legal). Section 7 — Budget, dates, contacts (3 lines). Pitfalls: briefing the channel instead of the task ("run our social media"); omitting the criterion; forgetting Section 6 until the contract is signed; attaching nothing — the brief is the cover page of the system's documents.

12.6.2 Managing the relationship

Four habits keep a supplier relationship productive. Weekly numbers, monthly review: the supplier reports the metrics of Section 5 weekly in your format, and attends (or sends notes to) the monthly review, where the decision to continue, adjust or stop is taken on the numbers. One point of contact on your side, with the authority to approve within the brief and the discipline to escalate what commits the company (claims, prices, positioning). Feedback in the brief's terms: "this does not use the pillar that leads for this persona" is actionable; "I don't like it" is not. An exit that is not a crisis: the ownership rules of Section 12.5.2 mean that a supplier can be changed in a week without losing accounts, data or history — which also keeps the relationship honest, because the supplier knows you can.

12.7 What La Machine and AI Take Over

12.7.1 The machine's share of the task table

Read against Table 11.1, La Machine takes over — fully or in part — the tasks whose automation level is Auto or Semi-auto: acknowledgment and routing, review requests, exclusion enforcement, consent expiry, the "not now" nurture, the newsletter's assembly and distribution, the weekly count's data, the funnel's transitions, the log, the queue. These are the repetitive, rule-based, time-critical tasks — roughly half the hours of the recurring ring in a solo business (Chapter 11, Do the Math). The AI assistant and the agents of the platform (→ see Book N11, The AI Assistant, Agents and Orchestration) extend the machine's share into drafting: a first version of the newsletter from the month's case, a summary of the fortnight's buyer quotes, a draft reply in the voice guide's terms, a proposed segment from a description of the ICP — each reviewed and approved by a person before it acts on a customer.

12.7.2 What remains human

Three things remain human, and the division is not temporary. The decisions — ICP, position, promise, channels, budget, the approval of any claim, price or message that commits the company — because responsibility cannot be delegated to a system, and because the regulators of Chapter 10 hold the enterprise, not its tools, accountable. The conversations — the interviews, the sales calls, the partner relationships, the complaint handled by a person — because their value is the human attention, and because the buyer knows the difference. The judgment about the machine's output — the weekly check of the logs, the review of the drafted newsletter, the decision that a blocked message was rightly blocked — because automation fails silently and drafting tools produce fluent errors. The owner's and the team's hours therefore move up the ladder, from repetition to judgment and relationship, which is where a small business's advantage has always been.

WHAT CHANGED Drafting, not deciding (2023–2026) Between 2023 and 2026, language-model assistants became competent at producing first drafts of marketing text — emails, posts, summaries, outlines — and at transforming existing material (a case into a post, a call transcript into notes, a long guide into a short one). For a small business this changed the economics of content production, not of marketing: the hours to produce a draft fell sharply; the hours to know what to say — the interviews, the hierarchy, the proof — did not, and the value of specific, first-hand, verifiable content rose as generic content flooded every channel (Chapter 1). The practical rules as of 2026: draft with the assistant from your documents (hierarchy, persona, voice guide, proof files), never from a blank prompt; verify every fact and figure before publication — an assistant does not hold your evidence files; keep a person's approval on anything that reaches a customer; and disclose nothing less than you would for any other tool — the message is yours, and so is the responsibility for it. → see Book K4 (Artificial Intelligence for Business) and Book N11.

IN LA MACHINE The assistant as a junior colleague Within the platform, the AI assistant works on the business's own data — contacts, segments, the log, the books of this library — and proposes rather than acts: a draft newsletter for approval, a suggested segment from a described ICP, a summary of the fortnight's quotes for the review, an answer drawn from this book when a team member asks how to write a verification criterion. Agents can run the approved repetitive steps of a workflow (Chapter 11) with the human-approval points you set (→ see Book N3, Book N11). What the assistant does not do is decide, sign or send without a person — and it does not hold your proof files, your authorizations or your legal judgment. Treat it as a fast junior colleague with an excellent memory and no accountability: brief it well, check its work, keep the decisions.

Costly Mistakes

1. Buying the strategy, keeping the execution. An agency decides whom to target and what to promise; the owner posts and answers the phone. Consequence: a generic strategy nobody in the company can defend, executed with the company's scarcest hours; the learning leaves with the supplier.
2. Hiring a channel instead of a system. A "social media specialist" or a "SEO person" hired before an ICP, a position or a task table exists. Consequence: activity in one channel, no funnel, no source tracking, and an expensive parting a year later.
3. The owner doing everything. Designing the identity, configuring the ads, writing the policy — at $150 an hour of displaced value. Consequence: marketing done last, done slowly, and the interviews and cases only the owner can do never done.
4. Accounts, data and files in the supplier's name. Consequence: a hostage situation at the end of the relationship — the advertising history, the analytics, the list, the source files — paid for twice or lost.
5. Briefing the channel. "Run our social media", "do our marketing". Consequence: the supplier invents the strategy, the criteria and the metrics — theirs, not yours — and the invoice reflects it.
6. Trusting drafts. Publishing assistant-generated text with an unverified statistic, a claim without a file, or a tone that is not the voice guide's. Consequence: the fluent error that a competitor, a customer or the Competition Bureau notices before you do.

Key Takeaways

• Seven skill families — strategy, content, data, channels, automation, design, compliance — are rated at three levels: aware, capable, expert. A small business is aware in all, capable in strategy, data and compliance, and buys the rest until weekly use justifies hiring.
• Own the strategy and the customer knowledge; standardize execution; buy the rest. The reverse — buying strategy, keeping execution — is the most common and most expensive inversion.
• A solo owner spends four to six hours a week — one of strategy and knowledge, two of content in your voice, one of review — with the six workflows automated and specialists bought by the hour for design, paid setup, automation and the policy. The owner's hour is the scarcest resource; do not spend it on what an expert does in a fraction of the time.
• The first hire is a generalist, made when the recurring ring exceeds ten hours a week, leads outrun the follow-up, the owner bottlenecks content — and the system exists in writing. Write the job from Table 11.1; verify wage data before setting a range; get the status (employee or contractor) and the language rules right.
• An established business has a marketing manager who owns the system, the numbers, the suppliers and the compliance program, with specialists in-house or bought; the owner still decides whom to serve.
• Freelancers, agencies and fractional executives each fit a situation; compare them on deliverables, fees (dated), control and risk — and never let a supplier own an account, a property, a list, a credential or a source file.
• Brief from your documents — ICP, personas, hierarchy, brand and voice guide, the task with its criterion, your metrics, the rules — on one page; manage with weekly numbers, a monthly review, one point of contact and an exit that is not a crisis.
• La Machine and AI take over the repetition and the drafting; the decisions, the conversations and the judgment about the machine's output remain human — and the responsibility with them.

Your Action Plan

The action plan below presents the seven steps that turn this chapter into work, in the order in which each step consumes the output of the one before. For each step, the table gives the deliverable that must exist at the end, the timing on a small business's calendar, and the verification criterion by which anyone — a co-founder, a mentor, the owner's own future self — can tell that the step was done. The criterion column is the one to read first: a step whose criterion you cannot picture is a step you do not yet understand.

In La Machine

Roles and suppliers are managed by people; La Machine holds the structure that makes delegation safe.

• Roles and access (→ see Book N12, Administration, Roles, Compliance and Backups) let you give a marketer, a seller, a freelancer or an agency exactly the access their tasks require — and revoke it in a minute when the relationship ends — while the accounts, the contact base and the log remain the business's. This is the platform-side half of the ownership rule of Section 12.5.2.
• Plan tasks and Runs (/taches, /executions) give each recurring task an owner and its verification criterion; the owner is chosen from a short list fixed in the code, which offers one human actor, so a supplier's tasks name the accountable person at the start of their objective, appear in the same list as the team's, with the same definition of done, and their runs show whether the work was done on time.
• The marketing log, responses and sales records (business/marketing/log, business/marketing/forms/responses) are where a supplier's results are read in your numbers — leads by source, transitions, cost per customer — rather than in a platform's dashboard.
• Templates, CTAs, branding and the voice (business/marketing/templates, ctas, business/branding) are the brief made operational: a freelancer producing content inside the platform produces it from your hierarchy and identity.
• The AI assistant and agents (→ see Book N11) draft from your data and run approved repetitive steps with the human-approval points you set; books and courses (business/assets/books) — including this one — are the training material a new hire or a supplier reads to learn the system in the terms you use.

What La Machine does not do: it does not recruit, evaluate or manage people; it does not determine a worker's status or a supplier's contract terms; and it does not decide which tasks to delegate — Table 11.1, the skills grid and the owner do. It makes the delegation safe and visible, which is what turns a hire or a supplier from a risk into capacity.

Self-Check

1. Name the seven skill families and the three that must stay in the house — and why.
2. What is the division of labor this chapter recommends for a small business, in one sentence?
3. What four signals, together, justify a first marketing hire, and what should the job description be written from?
4. Compare a freelancer, an agency and a fractional CMO in one line each, and state the ownership rule that applies to all three.
5. What does La Machine's assistant do with marketing content, and what must remain human?

Answers. (1) Strategy, content and writing, data and measurement, channels, automation and tools, design and production, compliance and ethics. Strategy (whom to serve and what to promise commits the business), customer knowledge (the persona becomes a stranger's document otherwise) and compliance ownership (a legal responsibility of the enterprise under P-39.1 s. 3.1) stay in-house (Sections 12.1.1–12.1.2). (2) Own the strategy and the customer knowledge; standardize execution into tasks with verification criteria; buy the rest from people who do it weekly (Section 12.1.2). (3) The recurring ring exceeds ten hours a week and is skipped; leads outrun the follow-up; the owner bottlenecks content; and the ICP, positioning, hierarchy and task table exist in writing — the job is written from the recurring tasks and criteria of Table 11.1 (Sections 12.3.1–12.3.2). (4) A freelancer sells one or two skills by the hour or project — fast, cheap, dependent on your brief; an agency sells a bundle on a retainer — breadth and process, higher cost, higher risk of owning your accounts; a fractional CMO sells one or two days a week of senior judgment — strategy and supplier management without a full-time salary. For all three: every account, property, list, credential and source file in the business's name, with revocable access for the supplier (Section 12.5). (5) It drafts from the business's own documents and data — newsletters, summaries, replies, segments — and proposes rather than acts; the decisions, the conversations and the judgment about its output (verification of facts, approval before anything reaches a customer) remain human, and so does the responsibility (Section 12.7).

Chapter References

Act respecting labour standards, CQLR c N-1.1. LégisQuébec. Retrieved September 7, 2026, from https://www.legisquebec.gouv.qc.ca/en/document/cs/N-1.1

Act respecting the protection of personal information in the private sector, CQLR c P-39.1, s. 3.1. LégisQuébec. Retrieved September 7, 2026, from https://www.legisquebec.gouv.qc.ca/en/document/cs/P-39.1

Charter of the French language, CQLR c C-11. LégisQuébec. Retrieved September 7, 2026, from https://www.legisquebec.gouv.qc.ca/en/document/cs/C-11

Canada Revenue Agency. (n.d.). Employee or self-employed? (RC4110). Government of Canada. Retrieved September 7, 2026, from https://www.canada.ca/en/revenue-agency/services/forms-publications/publications/rc4110.html

Government of Canada. (2026). Job Bank: Explore careers — wages. Retrieved September 7, 2026, from https://www.jobbank.gc.ca/explorecareers

Chapter 13 — Tools: Functions and Popular Tools

Most small businesses do not have a marketing tool problem; they have twelve marketing tools, each with its own list, none talking to the others, and an owner who spends the four available hours reconciling spreadsheets. This chapter is about the functions a marketing stack must perform — which are stable — and then about the tools that perform them — which are not.

What You'll Be Able to Do

• Name the twelve functions of a small-business marketing stack and say what each must do, independently of any tool.
• Apply three selection rules — free and already-paid first, connect everything, one source of truth — before buying anything.
• Read a function-by-function tool table and choose among popular tools by your situation rather than by fashion.
• Draw your integration map and locate the breaks where data is retyped or lost.
• Compute the total cost of your stack in dollars and hours, and cut what does not earn its place.
• Place La Machine in the stack: what it covers, what it does not, and how it connects to the rest.

Why It Matters

Tools are where the marketing system becomes real — or fails silently. A stack with one source of truth, connected end to end, executes the workflows of Chapter 11 at a marginal cost near zero and produces the counts of Chapter 2 without anyone typing. A stack of disconnected tools produces inconsistent data, unlawful sends (a list without consent in tool number seven), unmeasurable channels and a recurring cost that grows each time someone adds a subscription. For the owner, this chapter is a buying discipline that saves money and, more importantly, hours; for the team, it is the map of where each piece of data lives and which tool is authoritative — the end of "which list is the right one?"

In This Chapter

• 13.1 The Marketing Stack by Function
• 13.2 Selection Rules
• 13.3 Function-by-Function Tables
• 13.4 The Integration Map
• 13.5 The Total Cost of the Stack
• 13.6 La Machine in the Stack

13.1 The Marketing Stack by Function

13.1.1 Twelve functions

A function is a job the stack must do for the system of Chapters 1 to 12 to run. Functions are stable — a business needed to store contacts, send messages and measure results in 2006 and will in 2036 — while the tools that perform them change names, prices and owners every few years. Twelve functions cover a small business's marketing:

1. Contact base and CRM keeps one record per person and company, with consent, source, stage, segment, history and tasks (Chapters 2, 3, 9, 10). It is the hub of Figure 13.1.
2. Email, SMS and marketing automation covers campaigns, sequences, templates, the outbound gate and the log (Chapters 8, 10, 11).
3. Website, landing pages and forms is the owned destination: pages, conversion points, consent capture with source (Chapters 5, 8).
4. Social publishing and listening covers scheduling, posting, replies and visuals across the platforms you chose (Chapter 8).
5. Search visibility (SEO and answer engines) covers technical health, content structure, and keyword and citation tracking (Chapter 8; Books C3, C5).
6. Analytics and reporting covers site behavior, funnel counts, sources, cost per customer, the weekly count and the dashboards (Chapters 2, 9).
7. Design and visual production covers the brand guide's assets, templates, images and layouts (Chapters 6, 7).
8. Video and audio production covers recording, editing, hosting and captions (Chapter 8; Book C6).
9. Advertising platforms cover the paid channels' buying and tracking (Chapter 8; Collection D).
10. Forms, scheduling and payments cover booking a diagnostic, taking a deposit and selling the entry rung (Chapter 6).
11. Reviews and reputation covers requesting, collecting, displaying and answering reviews (Chapters 5, 7).
12. AI assistants and workflow automation covers drafting from your documents, connecting tools and running the six workflows (Chapters 11, 12).

DEFINITION Marketing stack, function and integration The marketing stack is the set of software tools a business uses to run its marketing system. A function is a job the stack must perform — storing contacts, sending messages, measuring results — defined independently of any product. An integration is a connection through which two tools exchange data automatically (a form creating a contact, a purchase updating a consent date, a campaign writing to the log). A stack is judged not by how many tools it contains but by whether every function is covered, every tool is connected to the source of truth, and the total cost in dollars and hours is earned by the results it produces.

13.1.2 Functions first, tools second

The order matters because tools are sold by feature and bought by fashion, while functions are needed by the system. A business that starts from the tool ("we should get the platform everyone uses") buys features it will not use and misses the function it needed; a business that starts from the function ("we need one contact base with consent and source on every record, connected to the forms and the email tool") can evaluate any product in ten minutes against a short list. Each table of Section 13.3 therefore opens with what the function must do before naming who does it.

13.2 Selection Rules

13.2.1 Three rules

THE RULE Free and already-paid first; connect everything; one source of truth A rule of practice, not of law, and the one that saves the most money in small-business marketing. Free and already-paid first: before buying a tool, check whether a function is already covered by something you pay for (the CRM's email module, the website platform's forms, the productivity suite's video recorder, the platform's built-in scheduler) or by a free tier that meets the verification criteria of Chapter 11; buy only when the existing tool fails a criterion you can name. Connect everything: a tool that cannot exchange data with the contact base — automatically, not by export and import — creates a second list, a second truth and a compliance gap; if it cannot be connected, it is not adopted. One source of truth: exactly one system holds the authoritative record of each contact — consent, source, stage, segment — and every other tool reads from it and writes to it; lists and segments are derived from it, never maintained separately. A stack that follows the three rules has fewer tools, cleaner data and lawful sends; one that does not accumulates subscriptions and spreadsheets.

13.2.2 Four more questions before buying

Beyond the three rules, four questions decide most purchases. Does it meet a verification criterion we have? If no task in Table 11.1 would be done better or faster, the tool is a curiosity. Who owns the data and the account? In the business's name, exportable at any time, deletable on request — the ownership rule of Chapter 12 applies to software as to suppliers. Where is the data stored, and under which law? Personal information stored outside Québec requires a privacy impact assessment under P-39.1 s. 17 (Chapter 10); the vendor's data-residency and privacy terms are part of the evaluation. What is the exit? Export formats, contract term, what happens to the data when you leave. A tool that fails the last three questions is a liability whatever its features.

KEY CONCEPT The cheapest tool is the one you already pay for and actually use Most small businesses pay for more marketing capability than they use: the CRM with an email module they never opened, the website platform with forms and a scheduler, the productivity suite with a video recorder and a survey tool. The first step of any stack review is not a purchase but an inventory of what is already paid for — and a check, against Table 11.1's criteria, of whether the unused capability would do the job. A function covered by an existing, connected tool at 80% of a specialist tool's capability is usually the right choice for a small business, because the 20% costs a second list, a second login and a second invoice.

13.3 Function-by-Function Tables

Each table names the function's must-do list first, then popular tools with their pricing model as observed in September 2026 — never an amount, which changes too often to print; verify on the vendor's pricing page before deciding. Tools are listed for information, not endorsement; La Machine appears in the functions it covers, alongside the alternatives. Where a tool is named in several tables, it covers several functions — often a reason to prefer it under the "free and already-paid first" rule.

13.3.1 Contact base and CRM

A contact base must keep one record per contact and per company; store the consent type, its date, its source and its expiry; record the source of the lead, the stage and the segment; keep the history of messages and interactions; carry tasks with an owner and a deadline; require consent on import; allow export at any time; accept the fields you define; and manage permissions by role. Table 13.1 lists popular tools for this function, with the function each covers, what it is best for, its pricing model as of September 2026 and a note.

Table 13.1 — Contact base and CRM (as of September 2026)

13.3.2 Email, SMS and marketing automation

An email, SMS and automation tool must provide templates that carry identification and unsubscribe; send campaigns to consented segments; run sequences triggered by events and dates; suppress excluded and expired contacts before every send; log every send and every block; send SMS with an opt-out; manage deliverability, including authentication of the sending domain; run A/B tests; and report by campaign and by source. Table 13.2 lists popular tools for this function, with the same four columns.

Table 13.2 — Email, SMS and marketing automation (as of September 2026)

13.3.3 Website, landing pages and forms

A website tool must publish pages that state the position and show proof and price; produce landing pages per campaign; offer forms that capture consent per purpose, with the wording version and the source; serve French versions where required; load fast on mobile; be editable by the team without a developer; run analytics and tracking under a consent banner; and live on a domain owned in the business's name. Table 13.3 lists popular tools for this function.

Table 13.3 — Website, landing pages and forms (as of September 2026)

13.3.4 Social publishing and listening

A social tool must schedule and publish to the chosen platforms from one place; hold a calendar; produce on-brand visuals; show replies and the inbox in one view; report the basics; route inquiries to the contact base; and require approval before publication where roles require it. Table 13.4 lists popular tools for this function.

Table 13.4 — Social publishing and listening (as of September 2026)

13.3.5 Search visibility: SEO and answer engines

A search-visibility tool must check the technical health of the site (indexation, speed, mobile); research keywords and questions in the persona's words; track rankings and citations by AI assistants; structure content so that engines and assistants can parse it; maintain the local business profile; and report. Table 13.5 lists popular tools for this function.

Table 13.5 — Search visibility (as of September 2026)

13.3.6 Analytics and reporting

An analytics tool must record site behavior by source under consent; count the funnel and its transitions; attribute leads and customers to their source; compute the cost per customer by channel; hold the weekly count and the monthly review's numbers; present dashboards readable in ten minutes; and export. Table 13.6 lists popular tools for this function.

Table 13.6 — Analytics and reporting (as of September 2026)