Product and engineering · Answered by Golden Section from more than 400 B2B software companies observed
Look at your best customer cohorts, not the blended churn figure. You have probably lost fit when customers who match your core profile retain worse than they used to, newer core cohorts adopt and expand less than older ones, and you lose deals to the same buyer you used to win. Many apparent losses of fit are something else: loose qualification bringing in customers who never fit, a drift away from your ideal customer, seasonal pauses logged as churn, or the roughly 20% of any cohort that is a poor fit. We underwrite gross revenue retention at 95% or better, and a core book holding there has not lost fit. Rebuild the last eight quarters of the ARR schedule by cohort and segment, then read retention with poor-fit and provisional accounts taken out.
Treat product-market fit as a property of your core segment, measured by cohort retention and expansion, not of the whole customer list. Rule out qualification, segment drift and accounting noise before concluding the product has stopped fitting the market.
| Metric | Value | What it means | Source |
|---|---|---|---|
| Gross revenue retention to underwrite | 95%+ | starting ARR retained after churn and contraction, before expansionGolden Section's underwriting standard; below 80% is treated as disqualifying | Golden Section, publishedInvesting in Software, 2026 addendum |
| Private B2B SaaS gross revenue retention, median | 84% (2025), down from 88% | gross revenue retention, annualBenchmarkit annual benchmarks, n of roughly 340 to 580; the fall appeared at every quartile | External benchmarkBenchmarkit B2B SaaS benchmarks, as cited in Investing in Software |
| Net revenue retention, what great looks like | 105%+ | ARR from existing customers a year later, including expansion, divided by starting ARRBalanced Path benchmark | Golden Section, publishedThe Balanced Path |
| Poor-fit share of a customer cohort | about 20% | customers in any cohort who do not fit the product wellraise their price or let them go; their churn is not evidence of lost fit | Golden Section operating viewGolden Section operating view |
Fit rarely disappears all at once, and it almost never shows first in the headline churn number. A blended figure mixes the customers who fit with the ones who never did, and it moves for reasons that have nothing to do with the product. So the first job is to split the book. Separate unproven accounts with the core and provisional segmentation and tag seasonal pauses. Then read retention by cohort on the ARR schedule. If the core holds, the problem is upstream in qualification or segment choice.
The most common false alarm is a qualification problem. When sellers say the product must change to land deals, loose qualification is usually the cause, and the fix is in the sales process rather than the roadmap. Real loss of fit looks different: core customers adopting less, expansion slowing in the segment that used to expand, and the churn identification work pointing to product or use-case fit rather than lost champions. In 2026 there is a second test. Where a product's value lived mostly in its interface, AI can erode fit quickly; where it lives in proprietary data, compliance and workflow depth, fit tends to hold.
The numbers are fictional. A vertical software company sees gross retention fall from 93% to 86% over a year, and the leadership team starts talking about a rebuild. Splitting the ARR schedule shows the core segment, mid-sized operators, still retains at 95%. The decline comes from small single-site customers that sales began closing eighteen months earlier, which retain at 70%. Lost-deal notes show reps asking for features those small customers wanted. The company tightens its qualification criteria, routes small accounts to a shorter provisional contract, and keeps the roadmap on the core segment. The rebuild conversation stops, because the product had not lost its market; the company had added a market the product never fit.
A real market shift, such as a regulation change or an AI-native competitor that does the core workflow at a fraction of the cost, can remove fit from a healthy core quickly, and then waiting for eight quarters of data is the wrong move. Very small books also make cohort reads noisy, so talk to churned and retained core customers directly.
From the Golden Section mistakes list, each paired with the play that prevents it.
Blending unproven customers into the core line makes a qualification problem look like lost fit.
Switching target customers too quickly is the usual response to a false alarm, and it costs the fit you had.
When fit really is slipping, the danger is holding the old position too long.
A seasonal pause logged as churn makes a healthy core look like it is losing fit.
In the order we would run them. Each is on its own page, most with a free Excel template.
Produces retention by cohort and segment so you can see where fit is or is not holding.
Keeps unproven accounts out of the core retention number you are reading.
Tests churn predictors to tell product fit apart from lost champions and budget changes.
Re-derives which customers are your highest quality if the core has drifted.
Tracks sentiment by segment over time, an early reading before retention moves.
Customer plays Retention, segmentation and adoption plays that show where fit holds and where it is slipping.
Customers in one industry run a daily workflow in the product, keep it year after year, and refer peers in a market where trust travels by word of mouth. In numbers, gross retention at or near 95%, net retention above 100%, and a sales cycle that shortens as references accumulate. Vertical fit is narrower and deeper than horizontal fit, which is why it is more durable.
No. Rising churn often comes from customers who were never a fit, from seasonal pauses counted as churn, or from unproven accounts mixed into the core revenue line. Split the book before you read the trend.
Usually not. In our experience that is the tell of loose qualification, the most common sales failure we see. Check whether the deals being lost are in your core segment before you change the roadmap.
We do not fund a company to find its fit again, and net retention below 85% is where we pass. Once the core segment retains, repositioning the product around it is the kind of structural change minority growth equity can fund.
Growth equity →Reviewed by Dougal Cameron, CEO & Co-Founder on 2026-10-22. Golden Section observations are labeled separately from external benchmarks and illustrative arithmetic.