Build a positioning document from customer evidence to live validation

01 Start with your best customers, not your largest.

Before research, before interviews, before anything: pull a list of 5–10 customers who understood quickly, bought quickly, stayed, and told others. These are not your biggest contracts. They are the buyers the product is most right for, right now.

The distinction matters. Your largest customer may have required heavy customization, a long proof-of-concept, and executive sponsors to push it through. That is expensive fit, not natural fit.

Look instead for buyers who found the product, saw the point immediately, signed without a lengthy procurement cycle, and still recommend it. They are your signal. They represent a buyer type you can find more of.

If you're pre-revenue, use the prospects who had the strongest "aha" moment during evaluation. You're looking for the same pattern at an earlier stage: recognition, not persuasion.

Write down what these customers have in common. Industry, team size, tech stack, trigger that started the search. That commonality is the first outline of your ICP, before you've asked them a single question. State it as the question this whole exercise must answer, then check the finished document against it. Diagnose before executing, refuse the playbook ask

02 Run JTBD switch interviews. Bring up the product last.

Interview 5–10 buyers from your best-customer list. Cover recent switchers from the past 90 days if you can. The goal is not feature feedback. The goal is language: the words buyers use to describe the problem before they knew your product existed.

Jobs-to-be-Done switch interviews trace a timeline backward from the purchase decision. You ask: what happened right before you started looking? What did you try first? What almost made you not buy? What did you tell colleagues when you recommended it? Each answer peels back a layer of rationalization and gets closer to the real trigger.

"JTBD interviews help you define a customer's language — what they actually mean by 'easy' — the root cause of why they switched, and how their story connects, all of which directly impact your product marketing."

· Bob Moesta, Intercom podcast, 2020 · “JTBD interviews surface the…”

The trigger language is operationally valuable. "We had a critical demo and the call quality dropped" is not just an anecdote. It tells you what channel to run ads on, what copy to write, and what objection to preempt. The decision language ("we just needed something that worked without us thinking about it") tells you what the headline should promise.

Cover five areas in each interview: background and role, what they were trying to accomplish, what triggered the search, how the decision unfolded, and how they describe the product to others now. Do not show the product or discuss your solution in the first 30 minutes. Buyers narrate the problem more honestly before they're defending a purchase decision.

After 5–10 interviews, highlight every phrase that appears across multiple buyers without being prompted. Those phrases are your message candidates. Keep them verbatim. Do not let an AI summarizer smooth them into clean themes, because the exact phrase is the hook and the clean theme is just a description. AI synthesis smooths over verbatim buyer phrases: mine Reddit and review sites directly, mirror the words, do not paraphrase

"Resonance comes from mirroring, not paraphrase."

· Nicolas Cole, Build and Launch a VOC Landing Page, 2026-05-07 · “AI synthesis smooths over…”

03 Map the full alternative set. Include doing nothing.

For every group of competitors or alternatives, ask: what would our best customers do if this product didn't exist? The answers cluster into four types: direct competitors, adjacent tools that partially solve it, manual processes (spreadsheets, hiring, building internally), and doing nothing.

Most positioning exercises stop at direct competitors. That is the wrong place to stop. The real distribution of how you lose deals is radically different from what the CRM shows.

"When you survey buyers who didn't buy, 40–60% say they made no purchase decision at all."

· April Dunford, Lenny's Podcast, 2026-04-28 · “40–60% of B2B buyers…”

When 40-60% of buyers who passed on you made no decision at all, your real positioning problem is not "why us over them." It is "why do anything at all." Status quo / no-decision is the real competitor

The pitch must first establish that the cost of the status quo is real and specific. Only then does arguing why your solution is better than alternatives become useful. Skip this step and you build messaging that wins competitive head-to-head deals while losing to inertia at twice the rate. The remedy for a status-quo loss is narrative and problem framing. The remedy for a competitive loss is a comparison and a differentiator. Treating one as the other sends the work in the wrong direction. No-decision-as-competitor vs. battle-card-driven competitive workflows

Cluster your alternatives into 3–5 groups. Each cluster becomes a lens for isolating your unique attributes in the next step. Include the status quo as an explicit cluster with its own description. What does "doing nothing" actually look like for this buyer? What's the daily cost? What breaks?

04 Build buyer personas from interview data, not from job titles.

Synthesize your interview findings into persona profiles. Goals, pain points, functional day-to-day, purchase path, emotional drivers. Every entry should trace to something a real buyer said or did. If you cannot cite a source interview, label it "hypothesis" and flag it for testing.

The fastest way to keep a persona honest is to refuse abstraction. Name three specific customers with three specific frustrations before you write a single 2x2 or persona archetype. The truth lives one level below the framework, and the framework debate can run forever without naming a real person.

"The truth is one level down. Always."

· Shreyas Doshi, Get to the Core of the Thing, 2026-05-16 · “Three customer stories beat…”

The most common persona failure is defining the champion by job title alone. "Head of Revenue Operations" or "VP of Sales" is a job title, not a champion. Two people with the same title at different companies can have completely different buying situations: different stage, different trigger, different alternatives, different urgency.

Define the champion in three layers:

The situation layer is the one most messaging skips. "Reps are calling without notes" is a champion. "VP Sales" is an audience.

Read the emotional layer too. At small contract sizes the rational story is the whole story. Above the threshold where the decision becomes career-defining, the buyer is reading you for whether this choice makes them the hero or the person who got it wrong in the all-hands. B2B buying is more emotional than the rational-buyer myth says, large-contract decisions carry personal-career stakes

"If I go sign a contract for a million dollars with a company that puts cake all over my face in front of my boss, my job is gone. That is real emotion."

· Mark Storin, on Dave Gerhardt's Exit Five Ep. 349, 2026-04-23 · “B2B buying is more…”

PMM's job runs in both directions. You bring the product to market (positioning, messaging, launch) and you bring the market to the product (buyer signals, competitive intel, customer language, switch triggers). The persona is where both halves meet.

"Product marketing's job is to bring the product to market and the market to the product."

· Martina Lauchengco, Loved: How to Rethink Marketing for Tech Products, 2022 · “PMM owns both halves…”

05 Isolate unique attributes. Apply the three-check filter.

For each cluster of competitive alternatives from step 3, list every feature or capability that differentiates your product from that cluster. Stay at the capability level: what does the user actually do, not what the product technically has. Sell to the buyer's mindset, not to product features

Then apply Omojola's three-check filter to each attribute:

1. Is it actually different from what alternatives offer?
2. Is it actually better on a measurable axis?
3. Does it matter viscerally to the user's job?

"Being different is not enough. Being better is not enough. It has to be better in a way that matters to the end user. When someone says 'why bet on Venmo?' I'd say 'try and send me a dollar I can use now.'"

· Ayo Omojola, Lenny's Podcast, 2026-04-28 · “Differentiation requires three checks:…”

An attribute that passes checks 1 and 2 but not 3 is a feature, not a differentiator. Note it as a supporting proof point but do not lead with it. Differentiation requires three checks, different, better, matters

The visceral check is not theoretical. It is a test the buyer can perform in a single sentence. Cash App's differentiator was "send me a dollar I can use now." Yours should be just as concrete.

A second thing most teams miss: you can differentiate on what you deliberately do not have. If simplicity is a differentiator (because your competitor requires a 12-week implementation and a dedicated admin), the absence of that complexity is a valid capability to name.

List no more than 5–7 differentiators that pass all three checks. Those are your foundation.

06 Map attributes to value themes. Phrase them as user verbs.

Group the attributes that deliver similar value. Then phrase each cluster as an action verb, what the user can do, not an abstract benefit or a product feature.

Wrong: "Advanced pipeline analytics"

Right: "See why deals stall before they fall out of the pipeline"

Build a value matrix per persona:

The message column is the hardest. It must be written in buyer language from your interviews, not in the vocabulary your product team uses internally. If the same message works for two different personas without editing, it's too generic for either.

This matrix is the foundation for every downstream artifact. Every headline, every pillar in the messaging house, every CTA should trace to a row here.

07 Choose the market frame. This determines who the buyer compares you to.

The market frame is your positioning category. It is not a naming exercise. Choosing a frame determines which competitor the buyer mentally places you against, what evaluations they put you in, and what table-stakes they expect you to have. Get it wrong and you spend your entire sales cycle fighting the wrong comparison.

Five frames and when each works:

The instinct to dominate the broadest possible audience is usually the wrong one. When every brand in a category sells the same general appeal, no buyer notices any single brand. A sharp, specialist frame that alienates the buyers who don't fit will out-convert a safe, broad one. The internet removed the old penalty for going narrow.

"If no advertising has a point-of-difference, all advertising has is a point-of-sameness."

· Dave Trott, POINT-OF-SAMENESS, 2025-01-20 · “If no advertising has…”

"The internet is Tokyo. The internet allows you to be niche at scale."

· Dave Trott, POINT-OF-SAMENESS, 2025-01-20 · “The internet is Tokyo,…”

One practical rule: don't name a new category before the market accepts it exists. The cost of educating buyers on what a category even is runs into millions of dollars of top-of-funnel investment. Most products that think they need a new category name actually need clearer positioning within an existing frame.

08 Write the unique belief. The worldview, not the value prop.

The unique belief is the philosophical ground your company stands on. It answers a different question than "what does your product do?" It answers: "what does this company believe about how the world should work?"

A value prop is "we help teams close more deals." A unique belief is "every revenue-critical conversation deserves an immediate, intelligent response, whether a human or an agent handles it." The first is a claim. The second is a worldview.

You are not trying to talk buyers out of what they think. You are finding the belief a slice of the market already holds and telling a story that rides it. The audience that already shares the worldview does the persuasion work for you.

"Godin's worldview concept holds that effective marketing begins by identifying people who share a specific worldview (a set of beliefs about how the world works) and then telling a story that resonates with that existing internal narrative, rather than trying to change minds through persuasion."

· Seth Godin, All Marketers Are Liars / Tell Stories, 2005 · “Don't try to change…”

The belief is what makes the company internally consistent. The product decisions, the sales narrative, the content, the brand voice: all connect back to it. It is also the check. Once written, it should make the sales deck, the homepage hero, and the product release notes obviously agree with each other, or reveal where they've drifted. Company story IS the strategy, name the shift, run the audit

"The company story is the company strategy."

· Andy Raskin, April 2026 · “Run a monthly narrative-drift…”

Optionally layer a trend for urgency: "We believe X, and right now Y is happening, which makes this the right moment to act." Trend layering works for new-way and new-category positioning. It is less useful when the product fits a well-established category where buyers are already searching.

09 Run the Onlyness test before you trust the differentiation.

Before you commit the differentiation, stress-test it with one sentence: "Our brand is the only ___ that ___." The first blank is the category that uniquely identifies what you are. The second is the value that uniquely identifies what you do. If you cannot fill both with something genuinely different and genuinely valuable, the problem is not the wording. The problem is the business. Onlyness is a company-viability test, not a positioning exercise, if you can't fill in the blank, the company is the problem

"If you cannot complete that sentence with something both different and compelling, the problem is not your positioning statement, it is your company."

· Marty Neumeier, ZAG, 2007 · “Onlyness is a company-viability…”

This is a structural diagnostic, not a copywriting exercise. If the sentence fails, do not hand it back to the marketing team to wordsmith until it "passes." Escalate it as a strategy question: narrow the category, deepen the value, or change the product. The cost of skipping this is a positioning document that reads well and sells nothing.

When the test passes, decide which differentiation type the evidence supports. Two types, chosen on evidence, not on what sounds dramatic in a board deck.

Binary: you are the opposite of something. DuckDuckGo vs. Chrome on privacy. Basecamp vs. project-management bloatware. Binary is sharp and memorable. It polarizes. The risk is that competitors reframe, claiming they also care about the same dimension.

By degree: you are measurably better on a specific axis. "3x more calls per hour." "Deploys in 20 minutes, not 12 weeks." By degree is safer because it ties to proof. The risk is that without the proof, it reads like marketing noise.

Most B2B software companies end up by-degree, because binary claims require architectural or philosophical choices that are hard to sustain. Pick whichever the evidence supports.

10 Write the positioning statement. All four questions, answered.

The positioning statement is the internal anchor for all messaging downstream. It's not customer-facing copy. It's the test for whether the positioning is complete.

Format:

For [champion in specific situation], [Product] is the [category] that [value]. Unlike [primary competitive alternative], [Product] [key differentiator].

Validate against Pierri's four questions. The statement must answer all four from a single reading:

1. What is it? (category: the frame you chose in step 7)
2. Who is it for? (champion in situation, not a job title)
3. What does it replace? (the primary alternative, could be the status quo)
4. Why is it better? (the differentiator that passed all three checks)

The same trinity governs the homepage hero you'll build in step 12: the specific use case, the alternative the buyer is using today, and the result, readable in roughly five seconds. B2B homepages must communicate use case, alternative, and result in five seconds

If any of the four questions is unanswerable from the statement alone, the positioning is incomplete. Do not move to the messaging house until the statement is tight. The most common failure at this step: the champion is too broad ("sales teams") and the differentiator is unsubstantiated ("best-in-class AI").

11 Build the messaging house. Three pillars, feature-level evidence under each.

The messaging house is the internal architecture. It's not customer-facing copy. It's what copy hangs off.

Structure: one overall value proposition, supported by three pillars. Each pillar is a named benefit, not a feature. Under each pillar: the features that enable it, and at least one customer quote or data point as evidence.

Three pillars. Not four or five. If you have more value themes, choose the three that matter most to the champion. The rest live on persona-specific landing pages or vertical plays.

Each pillar needs: a pillar name phrased as a benefit, a one-line message for that pillar, the features that enable it, and at least one customer quote or data point underneath it. The evidence is load-bearing. Without it, the pillar is a claim, not a position.

The messaging house is an internal document. Sales uses it to stay consistent. Marketing uses it to brief content. When reps say different things about the same product on the same day, the messaging house is what's missing.

12 Translate to homepage. The homepage is the positioning artifact.

The homepage is the one artifact that is always one click away from every investor, customer, employee, and prospect simultaneously. It has to hold the positioning under scrutiny.

Eight-section structure:

Lead with capabilities: what someone actually does with the product, not aspirational outcomes. "Build pipeline faster" describes an aspiration. "Answer every inbound call in 2 seconds and log the conversation to your CRM automatically" describes a capability. The capability converts.

Buyers default to perceiving comparable features across competitors. Gartner names this "sameness." Before the homepage goes live, test the hero with 5 people outside your team. Ask them: what does this product do, who is it for, and what does it replace? If they cannot answer all three from the hero alone, rewrite it.

"Buyer skepticism and the perception of 'sameness' will limit differentiation efforts if product marketers do not test their differentiators with external audiences."

· Gartner, Market Guide for B2B Message Testing, 2025 · “Buyers see "sameness", test…”

13 Test in live pitches. Messaging is theory until a real buyer hears it.

Before declaring positioning done, sit in 3–5 first-meeting discovery calls. Not to validate. Observe. Watch for three signals:

• Where the buyer interrupts with "wait, what do you mean by..."
• Where the buyer's energy drops or they stop asking questions
• Where reps deviate from the script (they deviate because the script isn't working for them)

Those moments are where the positioning broke. Fix them. Then test again.

"A live sales pitch is the best way for B2B companies to test their positioning… sales is a back-and-forth with a prospect; sales can and should be asking questions, looking for feedback, and handling objections. Marketing doesn't fully understand the sales team's needs when it comes to a narrative, and marketing stories don't fully serve what a salesperson needs to do in a first-call pitch."

· April Dunford, Sales-First Storytelling, 2024 · “Test positioning in a…”

The sales team detects positioning failures months before any dashboard does. When reps say "I keep getting asked about X and I don't have a good answer" or "buyers keep saying we're just like Y," that is a positioning failure in real time. Frontline customer contact is the PMM substrate

"Your sales team knows months before anyone else when a position is failing."

· April Dunford, LinkedIn, April 2026 · “The sales team detects…”

Build the rep-input loop from day one: a weekly objection log, a competitive channel in Slack, shadowing calls monthly. Without that loop, positioning drift accumulates silently.

14 Enable the organization. Positioning in a folder is a hypothesis.

A positioning document that lives in a Google Drive folder and gets visited twice a year is not a strategy. It is a hypothesis. The work isn't done until the positioning is installed across the organization.

Enablement checklist:

• Sales narrative built from the positioning statement and messaging house
• Product team briefed on positioning implications for roadmap sequencing
• Marketing team has a copy-standards brief that traces every claim back to an insight card
• Homepage, sales deck, and comparison pages audited against the positioning
• Live-pitch test completed and revisions incorporated
• Monthly narrative-drift audit scheduled: compare sales deck, homepage hero, and most recent release notes. When they diverge, fix before the next campaign launches.

The monthly audit is not optional. The artifacts that face buyers drift from each other within a quarter if nobody owns the canonical story. The PMM lead or founder is the right owner. The audit is a 30-minute read across three documents. The cost of skipping it compounds.

A note on the AI you'll use across this pipeline. Models compress the production: drafting personas, clustering attributes, summarizing interviews. They do not compress the judgment about which buyer and which frame, and they will fabricate when you run them across multiple thin or contradictory inputs without a human checkpoint at each stage. Building costs collapsed; judgement didn't, the squeeze is on positioning, not production Running AI on multiple inputs simultaneously without structured validation checkpoints produces fabricated output, not analysis errors

"AI needs A LOT of help to do a good job (way more than the average person realizes)."

· Anthony Pierri, AI needs A LOT of help to do a good job, 2026-05-10 · “Running AI on multiple…”

Keep a human verifying every generated claim against the source interview before it enters the document. Verification, not execution, is the irreplaceable human job