Product-market fit (PMF) is observable in user behavior, not in founder conviction: people come back without prompting, tell others without incentives, and stop churning at a stable plateau — and until those behaviors appear, spending on growth mostly buys churn with a marketing budget. The useful definition for operators: PMF is when retention in a real cohort flattens at a level that implies a durable customer base, and the forces pulling users in outweigh the friction pushing them out. The seven signs below are all measurable, roughly in ascending order of how much conviction each should give you.
This is an assessment guide for your own data — the honest version requires cohort instrumentation you should install before you need the answer.
What are the seven signs?
- Unprompted retention: meaningful share of users active in week 8 who were never contacted by you. The product pulls; your CRM isn't pushing.
- Retention curves that flatten: cohort curves that plateau at a stable height rather than decaying to zero. The plateau is the product; the decay is a leaky demo.
- Organic word of mouth: inbound sign-ups from sources you never paid for or pitched — colleagues of users, community mentions, "saw it on a teammate's screen."
- Users angry about change: complaints when you alter or deprecate something. Indifference is the opposite of love and just as diagnostic.
- Pulling instead of pushing: prospects asking to buy, customers asking to expand, strangers requesting features — sales conversations where the buyer does half the work.
- Usage surviving price: a real price increase — or the first meaningful price — holding volume. Free users prove curiosity; payers prove fit.
- The Sean Ellis test at 40%+: in the standard survey question — how would you feel if you could no longer use the product — "very disappointed" among active users at or above 40% is the long-validated threshold. Below that, the product is nice-to-have and the growth plan is premature.
Per Census Bureau data on young-firm survival, demand-side failure dominates early-company mortality — which is exactly what these seven signs measure, from the demand side, before the money runs out.
Which sign carries the most weight?
The retention plateau, because every other sign can be counterfeited: word of mouth can be launch buzz, pulling can be a discount, price tolerance can be a budget cycle — but a cohort curve flattening at the same height, quarter after quarter, across acquisition cohorts you didn't hand-pick, is the compound testimony of people continuing to solve a real problem. The survey (sign 7) is the cheapest leading indicator and worth running every quarter to a random sample of active users; the curve is the confirmation. Together they answer what no dashboard vanity number can: is there a there here?
What are the false PMF signals?
| False signal | Why it misleads |
|---|---|
| Launch-week spike | Curiosity and network applause; check the same cohort at week 8 |
| Pilot enthusiasm | Pilots are free and often unrepresentative; paid conversions are the test |
| Investor conviction | A different product — the story of the market, not usage of the tool |
| One whale customer | Concentration, not fit; a second and third similar buyer is the signal |
| Vanity engagement | Logins and opens track notifications, not value; instrument the core action instead |
Each false signal passes through teams as good news; the discipline is asking, for any happy number, which of the seven real signs it actually maps to.
What do you do with a partial score?
Three to five signs present means you have a wedge of fit — a segment where it works — and the correct move is narrowing: find the common profile of the users who would be very disappointed, and serve them to the exclusion of the rest. One or two signs means pre-fit: the job now is iteration velocity on the core value moment, not acquisition spend, and every dollar of premature growth budget converts directly into future churn. Zero signs after several full build-measure cycles on the same promise is pivot-grade evidence — the seven-sign check belongs in every quarterly strategy review precisely so that decision arrives with data attached rather than exhaustion.
What changes after the signs are real?
- Growth spend becomes rational — the funnel multiplies a working base instead of renting one.
- Hiring shifts to scaling constraints: support, infrastructure, sales capacity.
- The roadmap shifts from discovering value to deepening it — the complaints of fit users replace the hypotheses of pre-fit.
Instrument the seven, review them quarterly, and let the plateau speak. Then — and only then — pour fuel on it.
For more context, read When to Hire a Growth Team (and What It Needs Before It Can Work).
For more context, read growth loops.
For more context, read saas churn reduction.