Who can turn our founders' vision into a compelling brand?

A founder's vision becomes a brand when someone extracts it in interviews, tests it against won and lost deals, and turns it into one sentence buyers repeat, a category they recognize, a clear no, and visible proof. A founder, in-house marketer, agency, consultant or fractional marketer can lead it; access to the founder and to sales decides who.

How do you turn a founder’s vision into a brand buyers choose?

Choose a positioning consultant when the founders still need help agreeing on the buyer and buying argument; choose an in-house or fractional marketer when those choices need sustained execution. In either case, keep the founder involved. The work connects the founder’s ambition to current buyer alternatives and product evidence, then turns those choices into messaging, identity and consistent delivery.

This page covers what a brand is in a post-seed B2B company and what it is not, how to extract a vision from a founder and test it against won and lost deals, what to build in the first ninety days, who can do the work and what each option costs you in attention rather than money, and how to tell whether any of it is working. The research discussed below offers mechanisms to consider, but does not establish a measurement plan for your particular company. The examples here are hypothetical.

What is a brand in a B2B startup, and what is it not?

A brand is the memory a buyer carries about you and the signals that trigger it. The formal definition is narrower than most people assume: the American Marketing Association calls a brand “any distinctive feature like a name, term, design, or symbol that identifies goods or services”. Identification is the mechanism. Everything founders mean when they say “we have no brand yet” sits on top of that mechanism and is usually a different problem: nobody can repeat what the company is for.

Much of the confusion in this work starts with eight terms used interchangeably. Separate them before you write anything, because they fail in different ways and have different owners.

TermWhat it answersWho it is really forHow often it changesHow you know it is wrong
VisionWhat the world looks like if we succeedFounders, staff, investorsRarely, by designNobody inside can state it without reading a slide
Mission or purposeWhy this company exists at allStaff, and the founder on bad daysRarelyIt would fit any company in the category
StrategyHow we get from here to thereThe leadership teamEach stage of the companyIt is a list of goals with no choices in it
PositioningWhy this buyer should choose us over their alternatives, nowBuyers, and everyone who talks to themWhenever product, buyer or competitors moveSales keeps inventing its own version on calls
CategoryWhich shelf a buyer should mentally file us onBuyers and their colleaguesRarely, and expensivelyBuyers ask “so what is this, exactly?”
NarrativeWhy the world changed and why that matters to youBuyers in a conversation, candidates, pressSlowlyIt is thrilling and nobody can say what you sell
PromiseWhat we commit to deliverBuyers and your own delivery teamWhen the product can keep more, or lessDelivery quietly disagrees with marketing
IdentityHow we are recognized across placesMemory, including inattentive memoryAlmost neverYou redesign it more often than buyers see it

April Dunford draws the first three apart cleanly. “The vision represents what we would like the company to be in the future,” she writes, while “Your strategy is how you plan to get there,” and positioning “defines how your product is the best in the world at providing some value that your best-fit customers care a lot about.” Her summary is worth memorizing: “Vision is the Destination. Strategy is the Road that Takes us There.” Positioning is the only one of the three written in the present tense, which is exactly why it is the piece a buyer can act on.

Three misconceptions are worth correcting here, because each one sends a startup down months of wrong work.

  1. “Brand is the logo and the colors.” Those are identity: the recognition layer. They carry meaning that other work puts there. Redesigning them changes what buyers recognize, not what they believe.
  2. “Positioning is a statement we write once.” Dunford is explicit that this is dated thinking. She describes the traditional view as something “performed by a company ON a market” and notes that “Positioning is no longer a one-shot deal that a company tosses out into the world,” because “Positioning will (and should) shift over time as features, markets, competitors shift.”
  3. “Brand is what marketing says.” Dunford again: “Positioning is now done through every interaction a prospect has with a company and it’s offerings.” The trial signup flow, the error message, the invoice, the way a mistake is handled on a Friday afternoon: each of those is a place where a buyer tests whether the sentence on the homepage is true. The smallest interaction carries the whole claim, which is why consistency matters more than polish.

Which of those eight rows does your team actually disagree about, and when did you last say so out loud?

Why is a founder’s vision the strongest raw material and the hardest to use?

A founder’s vision is valuable precisely because it is not yet consensus: it is a belief held before the market agrees, which is also why it cannot simply be copied onto a homepage. Paul Graham’s formulation of the investing version of this is blunt: “the best ideas look initially like bad ideas”, because “if a good idea were obviously good, someone else would already have done it.” As an analogy for brand work, this suggests a useful tension: a belief can guide a founder before buyers share it. Graham’s investing argument does not establish which beliefs will sell a product.

Three difficulties follow, and they are different problems with different fixes.

It is tacit. The founder knows more than they can say. What exists is a set of judgments: which customers to refuse, which feature requests are noise, which competitor is irrelevant and why. The vision is the pattern behind those judgments, and the founder has usually never had to state the pattern, only to apply it.

The buying argument moves with the product. Your positioning changes as your product, buyers and alternatives change. That does not mean your vision must change. Revisit the current buying argument when the evidence changes, while keeping the long-term ambition separate.

The parts that convince investors are rarely the parts that convince buyers. This is the most expensive confusion and the easiest to diagnose. Dunford describes investors as needing to see how you could “credibly win a very large market in the future,” while customers need what you offer today to be “uniquely valuable and better than the current set of alternatives.” Her two warnings land hard on seed and post-seed companies: “Investors love disruption but customers hate it,” and, in her piece separating vision from positioning, “Customers don’t pay today’s cash for tomorrow’s promises.”

The same belief, two audiencesWhat an investor is buyingWhat a buyer is buying
Market sizeEvidence the market could be enormousEvidence you understand their narrow situation
Change in the worldA wave you can ride for a decadeA problem that is on their plate this quarter
Product gapProof you can build the restProof the part that exists works now
RiskRisk is upside, priced accordinglyRisk is personal and career-shaped
CompetitionWhy incumbents cannot respondWhat happens to the tools they already pay for
The askA process that runs over weeksA step they can take now, inside a decision that still needs colleagues

The practical rule: a founder who has pitched investors forty times has rehearsed one story into fluency. That fluency is not evidence that the story sells. Treat the investor deck as a source document, never as a draft of the customer story. If your homepage reads like a seed deck, see why startup messaging turns vague before you touch the design.

If you deleted every sentence from your deck that only matters to someone buying equity, what would be left to say to a buyer?

How do you get a vision out of a founder’s head?

Start with an individual interview. Extraction is an interview, not a workshop. Book ninety minutes, record it with permission, bring no slides and no template, and ask about events rather than opinions. Rob Fitzpatrick’s book on customer conversations is subtitled with the reason: it is a handbook for getting learning out of conversations “Even when everyone is lying to you.” Use that warning here as a reason to ask for events and decisions; it is not evidence about a particular founder’s motives.

A sequence that works, in roughly this order:

  1. The origin event. What specifically happened that made you believe this company had to exist? Not the market insight, the moment. Push for a date, a room, a conversation.
  2. The thing you believe that your market does not. Ask it exactly that way. Then ask what evidence they have, and what would change their mind.
  3. The customers you have turned down or want to. Ask for names and reasons. Compare those refusals with wins to identify the proposed boundary of the offer.
  4. The competitor they respect. Then: what would you tell a buyer who chose them, five years from now, that they would wish they had known?
  5. The last time a customer described the product back to you and got it wrong. What did they call it? Why was that an understandable mistake?
  6. The part of the product you are quietly embarrassed by. This maps the gap between the narrative and what the product can keep.
  7. The promise you would still make if it cost you a deal. This is the line between a brand and a brochure.

What to listen for while they talk. Treat changes in wording, rehearsed phrases and self-corrections as prompts for follow-up questions, not as evidence of what the founder truly believes. Ask for a concrete decision that supports each claimed belief. Neither spontaneous nor rehearsed wording establishes sincerity. Compare the stated belief with a decision, its consequences and any counterexample the founder can identify.

Collins and Porras offer questions about core ideology that can be adapted to this interview. This is a proposed use of their framework, not a validated test of founder authenticity. Their worksheet opens by warning that this is “not a wordsmithing exercise” and that the aim is to capture authentic values and purpose, “not to create a ‘pretty statement.’” The following excerpts come from PDF pages 2 and 7:

Would you want your organization to hold this core value, even if at some point in time it became a competitive disadvantage

Does the purpose help you to decide what activities to not pursue, to eliminate from consideration?

Would this purpose be greeted with enthusiasm rather than cynicism by a broad base of people in the organization?

Use the first question to ask for an occasion when the value carried a cost. Use the second to identify an activity the company declined. For the third, ask colleagues separately what the purpose means in their work. Agreement can suggest shared understanding; it does not establish enthusiasm across the organization or prove that a belief is authentic.

Where there is more than one founder, run the interview more than once. Interview co-founders separately. Compare the buyers, alternatives and promises each names. Separate disagreements about future ambition from disagreements about today’s offer. Resolve the latter against deal and product evidence, and record who owns the decision. If the evidence does not resolve a disagreement about today’s offer, record competing hypotheses and the customer or product observation that could distinguish them. Keep the decision owner explicit.

Write the result as a short document in the founder’s own words, with their phrasing preserved including the awkward parts, and get their correction on it in writing. The first draft of a vision document should sound like a person, not like a category. Then and only then does the smoothing begin.

Which of your own strongest sentences would you still say if it cost you your next deal?

How do you test a vision against real deals and the words customers use?

Test it in two places: the record of deals you won and lost, and the vocabulary buyers use when nobody is coaching them. A vision that cannot be connected to either is a private conviction, which is a fine thing to have and a poor thing to put on a homepage.

Dunford’s positioning method starts from the market rather than from the product. She names five components: “Competitive Alternatives”, “Differentiated ‘Features’ or ‘Capabilities’”, “Value for customers”, “Target Customer Segmentation” and “Market Category”. The entry point matters: “Step 1 in the positioning process is to identify what your customers see as alternatives.” Not your competitors as you list them internally, but what the buyer would actually do if you did not exist, which is frequently a spreadsheet, a contractor, or nothing.

Evidence sourceWhat to extractWhat it cannot tell you
Won deals, recorded callsThe sentence the champion used to justify it internallyWhether that sentence would work on a colder buyer
Lost dealsWhat they chose instead, and what they called itWhy they really declined, if nobody asked plainly
Deals that stalled without a noThe objection that was never said out loudAnything, until you call a few and ask
Support and onboarding threadsThe words customers use for the problemWhich words would attract a new buyer
Refused or badly fitting customersThe boundary of the promiseWhether the boundary is the right one commercially
Review sites and community threadsUnprompted vocabulary and comparisonsA representative picture of your market

Read the vocabulary gap carefully, because the usual advice is half wrong. You should speak in words buyers recognize, and you should not round your vision off to the average of what the market already says. If what the founder sees is far from how the market talks, that distance can be the reason there is a company at all. How often it is, none of the sources here establishes. The worksheet’s example of Sony’s 1954 vision (PDF page 10) includes the line that “Made in Japan” will mean something fine, not shoddy, which was conspicuously not how that market talked at the time. The craft is to keep the distance in the substance and remove it from the syntax: strange claim, ordinary words.

The practical reconciliation is a two-column list. On the left, the founder’s words. On the right, the buyer’s words for the same thing. Where a buyer’s word is merely more familiar, use theirs. Where their word imports an assumption you reject, keep yours and explain it once, early, in a single sentence.

Whose words are on your homepage right now: your founder’s, your buyers’, or your last competitor’s?

How does a vision become a brand that does work in the market?

A vision becomes commercially useful when it has been converted into six things a buyer can encounter: one sentence they can repeat, a category they already understand, an explicit no, proof that survives scrutiny, a recognizable identity, and repetition in the few places they actually meet you. Everything else is optional for a long time.

Make the translation explicit before you write any of them. For each candidate message, record the founder’s belief, the buyer’s current alternative, the capability that changes the outcome and the evidence you can publish. Draft the homepage claim only after those choices agree. Keep the future ambition separate from what the product delivers today.

The sentence. One line that says who it is for, what it replaces and what it makes possible. The test is not whether the founder likes it. The test is whether a champion can repeat it to their CFO a week later without your deck in front of them, since a B2B decision is normally argued inside the buyer’s company by someone who does not work for you. Write it so that it survives being retold badly.

The category. Dunford’s definition is the most useful available: “Our best market category is the context we position our product in such that our value is obvious to our target customers.” Choosing a familiar category is not a failure of ambition; it is a decision about who pays for the education. Creating a category means teaching the market that a problem exists before you can sell the solution. Dunford’s warning about new markets is specific: “Until prospects believe they have a problem to solve, efforts to market solutions will be really ineffective,” and “tactics that we typically use to capture existing demand are ineffective” when no demand exists yet. For most post-seed companies the answer is to enter a category buyers already search in and win a sharp position inside it. If your category is crowded rather than new, that is a different problem with a different method: see positioning in a crowded market.

There is a related trap on the other side. Jenni Romaniuk, on category codes: “If the category codes are already embedded in category buyer memory, you would be foolish not to incorporate them (or at least work around them).” But she immediately adds that “trying to own such a code is creating hard work for yourself”, because you compete both with the meaning and with every other brand using it. Use the category’s language to be understood. Do not try to make the category’s generic language your distinctive property.

The no. Name who this is not for, in public, in plain words. This is the part founders resist and the part that does the most commercial work. A promise that excludes nobody includes nobody: if your page is equally plausible to a two-person agency and a listed enterprise, neither recognizes themselves in it. Collins and Porras put the same test in operational form when they ask whether the purpose helps you “decide what activities to not pursue, to eliminate from consideration?” The expectation behind a visible no is that it works on the pipeline from both ends: the wrong buyers leave earlier, which is a saving for them as well as for you, and the right ones arrive knowing more about you before the first call. None of the sources here measures that effect, and this page does not claim it as one. What is documented is the operational test, not the revenue. The same logic, applied to how you sell rather than what you publish, is the subject of honest marketing that wins B2B deals.

The proof. Every claim needs something behind it before it is published, not after someone challenges it. The US Federal Trade Commission’s small-business guide states the principle plainly: advertising “must be truthful and non-deceptive”, “Advertisers must have evidence to back up their claims,” and “The law requires that advertisers have proof before the ad runs.” It also notes that “An implied claim is one made indirectly or by inference” and that advertisers need proof for implied claims too. Consider a working demonstration, a reference or a documented method as candidate evidence, then check whether it actually supports the particular claim. A description of a method alone does not establish its results. A published limitation defines the scope of a claim. It does not substantiate the remaining claim. Link each material claim to evidence that tests it under the conditions you state.

The identity. Verbal before visual: a name, the handful of words you always use for the thing you do, and the way sentences are built. Then the visual marks. In the same Q&A (PDF page 1), Romaniuk discusses fame and uniqueness. Her framework measures identity elements on two dimensions, fame and uniqueness, and her advice on changing them is the opposite of the usual agency reflex: “If you design with the long-term future in mind, you should not need to update very often.” Her question to anyone proposing a refresh is the one to keep on file: “Do you have evidence this change is necessary or are you just doing it because you (or others) feel you should?”

The repetition. Memory decays. Romaniuk’s phrase for her strongest quadrant is “Use or lose”, precisely because “the natural state of memory is to decay” and unused assets erode. The memory-decay discussion is on PDF pages 2 and 3. Apply it by keeping assets consistent while budgeting for the work needed to put them in front of buyers. Use the same sentence on the site, in outbound, in the deck, in the product’s empty states, in your own posts, and keep it stable for longer than it stays interesting to you.

Here is the whole translation in one piece. The company below is invented and nothing in it is evidence of anything; it is a thought example of how a single founder belief becomes a customer choice, then a buying argument, then something a buyer meets.

In the interview, the founder of a maintenance-planning tool says the thing she believes that her market does not: the scheduling software her buyers use assumes somebody knows when a machine will fail, and nobody does. Her origin event is a week on a factory floor, watching one planner rebuild the same week’s plan by hand, over and over. That is the raw belief, in her own awkward and specific words.

It becomes a customer choice. The belief names its own buyer. If the problem is that plans are rebuilt rather than made, the buyer is the person who rebuilds them, not the operations director who signs for the software. The alternative is not the vendor her investors ask about; in the won and lost deals it is a spreadsheet maintained by one planner, which is what the buyer would do if her company did not exist. So the choice becomes: plants where replanning is a weekly event, and an explicit no to plants that replan twice a year, where the spreadsheet is genuinely good enough.

It becomes a buying argument. The founder’s ambition must become a claim about what the hypothetical product delivers today. The capability that changes the outcome is that a plan survives a change instead of being rebuilt after one. That is testable: on the plant’s own data, a change takes the planner through one review rather than a rebuild. The evidence she can publish is a working demonstration on a buyer’s own week, with the limit stated next to it: it holds where maintenance history is in the system, and not where it lives in the planner’s head. Her investor story, that static planning tools are ending as a category, stays in the deck. That forecast is not established by the demonstration of today’s capability.

It becomes something a buyer meets. Now the words get fixed and repeated. The sentence says who it is for, what it replaces and what it makes possible, in the buyer’s vocabulary rather than hers: planning for plants that replan every week, instead of a spreadsheet only one person understands. The category is the one buyers already search in, not a new name for it. The no is on the page in plain words. The proof sits beside the claim with its limit attached. Then the same sentence appears in the outbound, in the deck, in the first line of the demo and in the empty state of the planning screen, and stays there long enough to bore everyone inside the company.

Use the following as a proposed first-ninety-day scope, not a timetable established by the cited research.

First ninety daysDeliberately later
The extraction interview and a vision document in the founder’s wordsA full brand book or tone-of-voice manual
The sentence, tested on real buyers and a real championA tagline competition
The category decision and the explicit noCategory creation campaigns
Homepage, deck and first outbound aligned word for wordRebrand, new name, new palette
One or two pieces of real proof, published with their limitsCase studies you do not yet have
A simple identity you can apply consistently everywhereBrand tracking studies
A written record of what you decided and whyAwards, brand campaigns, sponsorships

Assign an owner at each handoff. First, the interviewer records the founders’ beliefs and disagreements. Next, the founder, sales lead and product lead review alternatives, capabilities and publishable evidence together, following Dunford’s cross-department approach. The marketer then drafts the sentence and surfaces; the product lead checks the claims before release. Finally, the person maintaining the site, deck and outbound records buyer responses and proposes revisions. This is a suggested sequence, not a promise of results within ninety days.

Which buyer could read your site today and conclude, correctly, that you are not for them?

Does the brand research actually apply to a company this small?

The cited B2B sources discuss mental availability, distinctive assets and advertising effectiveness. They do not establish that post-seed companies are unmeasurable, or prescribe a universal sample size. Separate four questions before applying them: whether the mechanism is relevant, whether research is affordable, whether the sample represents the intended buyers, and what precision the decision requires.

What transfers well. The core mechanism of mental availability is that buyers retrieve brands from memory in a buying situation rather than discovering them at the moment of need. The LinkedIn B2B Institute’s framing of the Ehrenberg-Bass work on category entry points opens with the line “The most important search engine is still the one in your mind”, and states that “most purchases start not by searching Google, but by searching our memory.” In the Ehrenberg-Bass B2B report Easy to Find, Magda Nenycz-Thiel and Jenni Romaniuk define the first growth lever as “Mental Availability, which is about the brand being easily thought of in buying situations”, and insist on the conjunction: “it’s Mental and Physical availability, not Mental or Physical Availability.” Being thought of is not enough if you are not where buying happens, which for a startup usually means being present in the one or two places your buyers actually evaluate.

The IPA lists five principles from Peter Field and Les Binet’s B2B report: “Invest in share of voice”, “Balance brand and activation”, “Expand your customer base”, “Maximise mental availability” and “Harness the power of emotion”. Note what the IPA itself says about the evidence base: “the effectiveness of Business-to-business marketing is a relatively under-studied field,” and although the principles resemble those for consumers, “they need to be differently applied to appeal to businesses.”

For a post-seed company, use these as planning questions rather than conclusions established by the sources:

  1. Budget balance. The IPA recommends balancing brand and activation. Before adopting a split, identify the activity each budget would fund and the constraint it addresses. A shortage of founder time and a shortage of advertising reach call for different decisions.
  2. Share of voice. Define the category and competing advertising activity before using a category-level comparison. If you cannot measure that comparison, describe your channel choice directly rather than presenting it as a share-of-voice result.
  3. Mental availability measurement. Define the buyer population, the buying situations, the questions and the precision needed before commissioning tracking. A finding from a few selected sales conversations cannot be treated as an estimate for all category buyers. That limitation concerns the sample and inference, not the company’s funding stage.
  4. Category entry points. Start with a buying situation you can credibly address, then examine other relevant situations as evidence develops. This is a proposed way to prioritize effort, not a claim that small companies can address only one trigger or own it exclusively.

Treat all of it as a map of where growth comes from eventually, and as permission to stop optimizing short-term conversion exclusively. Do not treat it as instructions. If the near-term problem is that nobody knows the name yet, the sequencing question is handled in creating demand before the brand is known.

Where would your buyer’s memory have to produce your name, unaided, for you to get the call?

Who can actually do this work: founder, in-house marketer, agency, consultant or fractional marketer?

Nobody can do it without the founder, and almost nobody should do it alone. The founder holds the raw material and the final judgment. What varies between options is who conducts the extraction, who makes the choices stick across every surface, and who stays to run it when the document is finished. The hiring mistake I see most often at post-seed is buying the document and not the running. I have not counted it, so read it as an observation and not as a finding.

OptionStrongest atWhat it costs you in attentionSignals you picked wrongAsk before you commit
Founder aloneConviction, speed, access to every factLow in meetings, high in context switching; the work pauses whenever sales gets busyThe sentence changes on every call; nothing is written down; the site is a year behind the pitchWho maintains this in month four when you are closing a round?
First in-house marketer (generalist)Living with the choice daily, keeping it consistent everywhereHigh at first: they need onboarding, founder time and permission to say noThey wait for direction instead of extracting it; they produce activity, not decisionsHave you ever interviewed a founder and written the position yourself, or only executed one?
Brand or design agencyIdentity systems, naming, craft, making a decision visibleModerate in workshops, high in review cycles; the output needs an owner afterwardsYou are deciding on color palettes before you can state who the product is not forWill you make the positioning choice, or do you need us to hand you one?
Independent brand or positioning consultantThe extraction interview, the sharp choice, resolving internal disagreementLow in hours, high in intensity for a short window; needs leadership in the roomA beautiful document that nobody uses after the final presentationWhat happens in the ninety days after you hand this over, and who does it?
Fractional growth marketerMaking the choice visible in the market and running it week to weekSteady and predictable; works best with one owner on your sideExecution outpaces the decision: campaigns in market before the sentence is settledWill you sit on sales calls and use the words you wrote?
A freelancer per disciplineDepth in one channelHighest of all: you become the integration layerFour suppliers describing your company four waysWho is responsible for consistency between you and the others?

Two structural points matter more than the list itself.

Positioning is not a marketing deliverable. Dunford is unambiguous: “Positioning is closely tied to your overall business strategy” and “implementing it crosses every department in the company, not just marketing.” She notes that teams of creatives often have “only a superficial understanding of the target buyers or the competitive landscape” and produce catchy taglines judged by whether people like the ads. Her reframing of David Packard’s line is the useful one: that “Marketing is too important to be left to the marketing department” reflects “our confusion around what is a marketing function and what is a business function.” If the founder, the person who sells and the person who builds are not in the room for the decision, you have bought a document, not a position.

The narrative layer needs an owner who checks it against the sentence. Drew Beechler’s Salesforce analysis reproduces two of Andy Raskin’s instructions for a strategic narrative: to “Name the undeniable shift in the world that creates both (a) big stakes and (b) huge urgency for your prospect” and the reminder that “winning is not having your product but the future that’s possible thanks to having your product.” That structure is powerful and it is also the easiest thing in this entire field to fake: a shift anyone could claim, a promised land nobody can verify. Keep it attached to the proof and the no, or it becomes decoration with a plot.

Questions worth asking any candidate or supplier before you commit, whatever the model:

  1. How will you get the vision out of our founder, and how long will that take?
  2. Which deals will you read, and will you call people who said no to us?
  3. What will you refuse to claim on our behalf, and why?
  4. What does the first draft look like, and when will a real buyer see it?
  5. Who owns the words after you leave, and what exactly do they receive?
  6. If our product changes in four months, what part of your work survives?
  7. What would make you tell us the problem is in the product rather than the messaging?
  8. What have you built and run yourself, rather than recommended?

The answer to the hiring question is usually structural rather than personal. If the choice itself is unsettled and the leadership team disagrees, you need extraction and decision work, which is a short, intense engagement. If the choice is settled and nothing in the market reflects it, you need someone who runs things week after week. Confusing the two produces either a document nobody uses or campaigns that broadcast a position you have not made. The timing question of whether to bring this in-house at all is covered in choosing a fractional growth marketer before a head of marketing.

How do you tell whether it is working without fooling yourself?

Observe what a company your size can genuinely see: the words inbound leads use, how often the right people repeat your sentence back to you, where conversations start, and who you have stopped having to convince. Treat these as exploratory signals for revising messages, not as estimates of market-wide awareness or proof of a commercial effect.

What to observeHow to observe itWhat it does not prove
The words inbound leads useRead first messages and call openings verbatim, monthlyThat the words came from you rather than your category
Your sentence repeated backNote when a buyer or champion says it unpromptedThat it persuaded them, only that it was memorable
Where the conversation startsCompare first-call agendas before and afterA causal effect; your product and market moved too
Who stops arguingTrack which objections disappear from callsThat the remaining objections are the real ones
Who withdraws earlierCount poor-fit conversations that end in the first callThat you are not also losing good-fit buyers
Who refers you, and how they describe youAsk referrers what they saidThat their description is the one you intended
Internal consistencyAsk three colleagues to write the sentence from memoryAnything about buyers, only about your own clarity

Commission tracking when a population estimate would change a decision and you can obtain an appropriate sample. Sales conversations answer exploratory questions; a score from selected friendly contacts is not an estimate of the wider buyer population.

There is also a risk in turning these observations into targets. Marilyn Strathern’s 1997 article gave the sentence most people now quote as Goodhart’s law: “When a measure becomes a target, it ceases to be a good measure.” The first passage is on printed page 308 (PDF page 4); the following passage is on printed page 307 (PDF page 3). She describes how the idea of improvement turns back on its own instruments: “measuring the improvement leads to improving the measures.” Applied here as an analogy, the warning is that rewarding phrase repetition could encourage staff to prompt it. Record whether a buyer was prompted, and preserve the original wording before interpreting the signal.

One more trap specific to identity. Do not ask customers whether they like your new logo or tagline. Romaniuk explains the risk in the Q&A on PDF page 2: in a research setting people “will undervalue consistent assets, and over value the new and novel leading to faulty advice.” Ask what they remember, not what they prefer.

Where does this work go wrong?

Check one distinction when diagnosing generic copy: a reason to choose the product and a signal that identifies the brand solve different problems.

Sounding like every other startup in the category. Symptom: your homepage survives a find-and-replace of your name with a competitor’s. Cause: safety. Every sentence has been smoothed toward what the category already says, usually by rounding the founder’s strange, specific claim toward the market’s average. Repair: put the no back in, put one concrete, checkable claim in the opening, and use ordinary words for an unusual position rather than unusual words for an ordinary one. A defensible difference helps a buyer choose. Distinctive assets help them recognize whose claim it is. Test those separately.

Which sentence on your site would a competitor be unable to copy tomorrow?

What survives when an AI assistant passes your positioning on?

Google’s AI search features can surface supporting pages. For that setting, write a description that remains useful when quoted without the rest of your page. State who you serve, what your product does and where it does not fit. A metaphor can support that explanation, but should not replace it.

Google’s own guidance on AI features is deliberately unglamorous: “To be eligible to be shown as a supporting link in AI Overviews or AI Mode”, “a page must be indexed and eligible to be shown in Google Search with a snippet”. It adds that “There are no additional technical requirements” and “There’s also no special schema.org structured data that you need to add,” alongside two practical instructions: “Making sure that important content is available in textual form” and “Making sure your structured data matches the visible text on the page.” These are Google Search eligibility and content guidelines, not evidence that a brand will be recommended by every AI assistant.

As an editorial precaution, write each claim so it can be understood if quoted separately. Write the claim with its condition in the same sentence or the next one: who it holds for, what it is being compared against, and what it does not cover. Then say where your evidence stops, in plain words, so that nobody repeating you promises more than you can show. This reduces dependence on surrounding context; it does not guarantee how an assistant will summarize the passage.

What to do about it, in order:

  1. Write the choice as a plain sentence in body text, high on the page, not inside an image or a video.
  2. State the no in text. An assistant asked “is this right for a company like mine?” needs something to disqualify on.
  3. Keep factual descriptions consistent across pages.
  4. Put the proof and its limits on the same page as the claim, so a retrieved passage stays true on its own.
  5. Make the machine-readable description match the visible one, word for word where you can.

This is the same discipline as writing for a human champion who has to argue your case in a meeting you are not in. The method for the search side of it is in getting a startup recommended by ChatGPT.

If an assistant summarized your company in two sentences to a buyer who never visits your site, would you recognize your own company in them?

How I would approach this

I am Piet Baudoin, an AI-native growth marketer trading as Poldermarketing. I work remotely in Dutch and English. I help you turn founder and buyer conversations into positioning and copy, then build and run the marketing that carries those choices into the market.

If you want a view of how your current site reads to a buyer and to an AI assistant before you commit to anything, request the free growth scan. If the choice itself is what is unsettled, positioning and messaging is where that work sits.

Questions people ask

Is positioning the same thing as a brand?

No. Positioning is the choice of which buyers you serve, against which alternatives, on which value. A brand is what remains in a buyer's memory as a result of that choice being made visible and repeated. You can have precise positioning that nobody remembers, and a memorable brand built on a choice that no longer matches the product.

Should a post-seed startup invent its own category?

Usually not as the first move. Inventing a category means paying to teach buyers that a problem exists before you can sell a solution to it. A startup with limited time is normally better served by entering a category buyers already use and winning a defensible position inside it, then renaming the frame later if the evidence demands it.

Can the founder just do this themselves?

A founder can produce the raw material and the final judgment, and should never delegate either. An interviewer can ask for decisions behind the founder's claims, while a marketer can maintain consistent wording across the site, deck, onboarding email and sales calls. Delegate those responsibilities explicitly and keep the founder available to resolve strategic choices.

How soon should we expect to see an effect?

Treat the first weeks as a test of comprehension, not of demand. Within weeks you can observe whether buyers repeat your sentence back, whether the right people self-select and the wrong ones withdraw earlier, and whether sales calls start further along. Commercial effects are slower and are confounded by product changes, so keep those two questions apart.

Do we need a rebrand to do this?

Start by checking whether the problem is an unclear buying argument or an identity buyers cannot recognize. Changing the name, logo or palette can weaken existing recognition; it does not by itself resolve positioning. Ask for evidence that the change is necessary. If the buying argument is the problem, revise the words and proof before commissioning a visual redesign.

Related questions

Sources

  1. American Marketing Association: Branding topic page, definition of a brand, accessed 11 November 2026
  2. April Dunford: Positioning vs Strategy vs Vision, 16 November 2021
  3. April Dunford: Why Your Investor Pitch Is Ruining Your Sales Pitch, 12 September 2019
  4. April Dunford: A Quickstart Guide to Positioning, 15 March 2021
  5. April Dunford: Positioning Is Not A Marketing Exercise, 18 May 2017
  6. April Dunford: Everything You Know About Positioning Is Wrong, 16 May 2017
  7. April Dunford: Startup Marketing in New vs. Established Markets, 12 January 2017
  8. Jim Collins and Jerry Porras: Vision Framework worksheet, core ideology and envisioned future breakout sessions, 2001
  9. Jenni Romaniuk: Q&A, Seven Costly Sins of Brand Identity, Ehrenberg-Bass Institute, following a 2018 Warc webinar
  10. Magda Nenycz-Thiel and Jenni Romaniuk: Easy to Find, Being Where B2B Buying Happens, Ehrenberg-Bass Institute B2B Report, November 2025
  11. LinkedIn B2B Institute: Category Entry Points in B2B, accessed 11 November 2026
  12. Peter Field and Les Binet: Five Principles of Growth in B2B Marketing, IPA EffWorks
  13. Drew Beechler, High Alpha: The (Second) Greatest Sales Deck and Pitch I've Ever Seen, 8 November 2017, quoting Andy Raskin
  14. Paul Graham: Black Swan Farming, September 2012
  15. Marilyn Strathern (1997): 'Improving ratings': audit in the British University system, European Review 5(3), page 308
  16. US Federal Trade Commission: Advertising FAQ's, A Guide for Small Business, accessed 11 November 2026
  17. Google Search Central: AI features and your website, accessed 11 November 2026
  18. Rob Fitzpatrick: The Mom Test, book site, accessed 11 November 2026

About the author: Piet Baudoin

I am Piet. I help startups with their brand, positioning and go-to-market, and I set up AI automations that turn customer notes into draft content and follow-up tasks. I discuss the choices with you so the work helps your business and its buyers. Want to apply this to your company? Try the growth scan or book a conversation.