A positioning rewrite that gave a B2B SaaS one sentence to sell with
Twenty-four customer interviews narrowed three half-served audiences to one ideal customer, produced a positioning line a sales call could open with, and planned a version launch the whole company described the same way.
The brief, in specifics
Professional Services
9 weeks
Fixed price, phased
Client name withheld under NDA. Engagement details are shown to the extent our agreement permits.
The question we were asked
A compliance-workflow product sold to accountancy practices had three audiences on its website and a forty-minute explanation on every sales call. The team could build. What it couldn't do was say, in a sentence, who the product was for or why.
A compliance-workflow product with a strong engineering team and a website written for three audiences at once. Growth had come from referrals, which meant nobody had ever had to explain the product to a stranger in a sentence, and the sales calls showed it. The engagement was nine weeks, most of it research, because the team asked for a positioning line and didn't yet have the evidence to write one from.
Marketing Strategy
The decision, first
- 01
A homepage for three audiences is a decision nobody made, and prospects can tell.
Three audiences on one homepage isn't flexibility. It's the absence of a decision, and every prospect can feel it.
- 02
The people who said no were the most useful interviews in the corpus.
The lost deals and the churned accounts were the most useful interviews: they had already worked out what the product wasn't.
- 03
Positioning is a hypothesis; the win/loss loop is what keeps it honest.
A positioning line is a hypothesis, and the win/loss loop is the experiment. Without it, the line drifts back toward the feature list within a year.
What the numbers couldn't answer
- 01
The homepage spoke to sole practitioners, mid-size firms and enterprise compliance teams at once, so it convinced none of them. Demos began with the product's history instead of the buyer's problem, and a third-party comparison site described the product more clearly than the company did.
The clearest symptom was on the calls. A first demo opened with the company's founding story and a tour of every module, and reached the prospect's actual problem around minute forty. By then a third of them had dropped. The website mirrored it: three hero variants had been tried, each written for a different audience, and the one that stayed was the compromise that offended nobody and convinced nobody.
We interviewed recent buyers, lost deals and churned accounts, scored the segments against fit and reachability, chose one, and rewrote the positioning from that segment's own words. The version-three launch was then planned around that line (site, deck, onboarding and a win/loss loop) so every surface said the same thing.
How we worked it through
- 01
Interviewed twenty-four buyers, lost deals and churned accounts in three weeks
Buyers, lost deals and churned accounts were interviewed in equal numbers, because the people who said no explain a positioning gap better than the people who said yes.
- 02
Scored the three candidate segments on fit, urgency, reachability and deal size
Scoring the segments on shared axes turned a two-year argument into a table. The enterprise segment lost on sales-cycle length, not on revenue.
- 03
Wrote the positioning and a three-tier messaging hierarchy from the chosen segment's own language
The messaging hierarchy was built from transcript phrases so that every claim on the site could be traced to a buyer who had said it.
- 04
Planned the version launch as one runbook across site, sales and product
Launching everything against one line on one day is what made the positioning measurable afterward; a rolling launch would have blurred the before and after.
- 05
Set up a win/loss loop so the positioning keeps being tested after we left
The win/loss loop was set up before we left, so the positioning gets retested every quarter instead of simply trusted.
Phase by phase
Phase 1: Listen
Twenty-four conversations
Interviewed buyers, lost deals and churned accounts in equal numbers, coded against four shared questions.
- Interview corpus
- Coding framework
Phase 2: Choose
One segment, on evidence
Scored the three candidate segments on fit, urgency, reachability and deal size, and chose one.
- Segment scoring board
- ICP definition
Phase 3: Say it
The line, and what sits under it
Wrote and call-tested the positioning, then built the three-pillar messaging hierarchy beneath it.
- Positioning statement
- Messaging hierarchy
- Site and deck copy
Phase 4: Launch
Everywhere at once
Planned and ran the version launch as one runbook across site, sales, onboarding and partners, with a win/loss loop behind it.
- Launch runbook
- Win/loss dashboard
The interview corpus: twenty-four conversations in three weeks (eight recent buyers, eight lost deals, eight churned accounts) coded on the same four questions, with the founders' eight marked and the split by segment the scoring started from.
What it changed
+31%
Demo to proposal
40 min → 9 min
Time spent explaining on a first call
2.1×
Qualified pipeline, quarter after launch
Demo-to-proposal compares the quarter after launch with the quarter before, on the chosen segment only. Explanation time is measured from call recordings: minutes before the prospect's own problem is discussed. Pipeline is qualified opportunities created in the quarter after launch against the quarter before; the segment was narrowed at the same time, so part of the lift is the qualification bar moving.
Client name withheld under NDA. Figures are approximate, drawn from the engagement’s own reporting.
One ideal customer, named
Three half-served audiences scored against each other, and one chosen on evidence instead of hope.
Every interview was coded against the same four questions (how urgent the problem was, how the firm found and evaluated tools, who signed, and what they'd pay) so the three segments could be scored on one scale instead of argued about. Practices of twenty to two hundred staff with a named compliance lead scored highest on every axis but deal size, and the deal-size gap was smaller than the sales-cycle gap that came with enterprise. The choice was a table, not a meeting.
- Twenty-four interviews coded on four shared questions
- Segments scored on fit, urgency, reachability and deal size
- The choice recorded as a table the team can re-score
The scoring board: three segments scored on fit, urgency, reachability and deal size, the 20–200-staff practices chosen, enterprise set aside on sales-cycle length (deal size wasn't the reason), and the twenty-four interviews tallied beneath by type.
A positioning line that survives a sales call
The messaging hierarchy: the kept line beside the three candidates read at the top of live demos, then three pillars and nine proof points, each proof traced to the interview quote that produced it.
The line was written from buyers' own words and tested in calls before it went on the website.
The interview transcripts were mined for the phrases buyers used for the problem, and the positioning line was assembled from those, with the product's feature list set aside. Three candidate lines were then read aloud at the top of live demos for two weeks, and the one that made prospects nod instead of ask a question was kept. Under it sit three pillars and nine proof points, each traced to the quotes that produced it, so a future rewrite can see what evidence it's discarding.
- Written from transcript phrases, not the feature list
- Three candidates tested on live calls before the site
- Every pillar traced back to the interviews behind it
A launch the whole company said the same way
Site, sales deck, onboarding and partner emails planned as one runbook against one line.
The version-three release was treated as the moment to make the positioning true everywhere at once. One runbook held every surface that would say something about the product (homepage, pricing page, deck, first-run onboarding, the partner announcement) with an owner, a due week and a review against the messaging hierarchy. Nothing shipped saying it a different way, which is the only reason the mention rate in the win/loss calls afterward was measurable at all.
- One timeline across marketing, sales and product
- Every surface reviewed against the hierarchy before release
- Partner and customer announcements on the same day
The launch runbook in release week: site, deck, call script, onboarding, customer email and partner announcement on one timeline, each with an owner, a review against the hierarchy and a status, all shipping on the same day, with the first win/loss review after.
How the engagement ran
A team of 3 worked on a fixed price basis over 9 weeks in three movements (research, positioning, launch), covering Customer research, Positioning & messaging, Launch planning. The founders sat in on a third of the interviews, and the launch runbook was run by their marketing lead with a weekly review, so the positioning was theirs before the engagement ended.
Nine weeks, fixed price, in three movements: research, positioning, launch. The founders sat in on a third of the interviews, which mattered more than any deliverable. A founder who has heard four customers describe the problem the same way stops defending the old homepage. The launch runbook was run by their marketing lead with us reviewing weekly, so it was theirs before we left.
Forty minutes to nine
A first call that reaches the buyer’s problem before the buyer leaves
Switch between the old demo and the new one to see where the minutes went. Below it, re-score the three segments one axis at a time to see why the line was written for one buyer. Use the arrow keys once a tab is focused.
The same first call, before and after the launch
How the call opens“Let me tell you how we started…”
Why one segment, scored one axis at a time
- Sole practitionersOne to five staff, owner does compliance2 / 5
- Practices of 20–200 staffWith a named compliance lead5 / 5
- Enterprise compliance teamsRisk function, procurement signs3 / 5
From what buyers said to what every surface says
Research, a decision recorded as a table, words traced to the people who said them, one launch, and a loop that keeps testing the result after the engagement ends.
- 01 · SourceTwenty-four interviewsRecent buyers, lost deals and churned accounts in equal numbers, because the people who said no explain a positioning gap best.
- 02 · ScoreThree segments, four axesEvery interview coded on the same four questions, so fit, urgency, reachability and deal size sit on one scale, kept as a table to re-score.
- 03 · WriteLine and hierarchyBuilt from transcript phrases and call-tested for two weeks; three pillars and nine proof points, each traced to the quotes behind it.
- 04 · ShipOne runbook, one dayEvery surface with an owner, a due week and a review against the hierarchy; partner and customer announcements on the same day.
- 05 · RetestWin/loss loopReasons won and lost, and the line's mention rate per call, reviewed every quarter so the positioning keeps getting retested.
So choosing one buyer isn’t a bet
A narrower market, chosen on evidence and kept honest after launch
Narrowed on a table, not a meeting
Three segments scored on the same four axes from twenty-four coded interviews. Enterprise lost on sales-cycle length, not on revenue, and the table stays with the team to re-score.
A hypothesis with an experiment behind it
The win/loss loop was set up before we left and retests the line every quarter, so the positioning is checked against real calls instead of trusted to hold.
Figures labeled for what they are
Demo to proposal is the chosen segment only. Explanation time is from call recordings. The segment narrowed while pipeline was measured, so part of the 2.1× is the qualification bar moving.
Could everyone on your team open a first call with the same sentence? Scope your build in 3 minutes.
Scope your buildNearby engagements
AI & AutomationA private legal assistant grounded in verified precedents
A private knowledge assistant that searches internal case files and precedents, providing cited answers legal teams can verify in seconds.
Legal & Law Firms · 14 weeks
Product DesignAn onboarding flow that guides trial users to value
A redesigned SaaS trial onboarding experience with progressive checklists, sample data, and inline guidance that turns signups into active subscribers.
Professional Services · 10 weeks
Product DesignA design system that brought speed and consistency to 4 product teams
A token-based design system in Figma and React that eliminated component duplication across 4 product squads and cut the time from design handoff to merged frontend.
Professional Services · 14 weeks
Let's talk
Running a large platform, shaping a first MVP, or getting a product ready for a funding round? Tell us where you are. We'll shape the process around it, and stay with you after launch.
