How to Measure Product-Market Fit
Product-market fit is evidenced by demand that pulls you forward — retention, organic pull, and customers who would be very disappointed without the product — not by a vanity launch spike. Sean Ellis popularized a survey threshold (roughly 40% “very disappointed” if the product disappeared) as one leading indicator. Pair survey results with retention cohorts and qualitative signals; do not treat any single number as a universal law or invent benchmark tables you cannot cite.
Updated August 8, 2026.
Sean Ellis survey (the “very disappointed” test)
Ask users who have experienced the core product: “How would you feel if you could no longer use [product]?” Options typically include very disappointed, somewhat disappointed, and not disappointed (plus N/A). In Ellis’s framing, strong product-market fit often correlates with about 40% or more answering “very disappointed.” Treat 40% as a heuristic popularized by Ellis — not a regulated standard — and segment results by ICP so non-target users do not dilute the signal.
- Survey engaged users, not one-time tire-kickers.
- Read verbatim answers to “what is the primary benefit?” and “how can we improve?”
- Re-run after major positioning or ICP changes.
Retention cohorts
Cohort retention answers whether people come back after the novelty wears off. Define the activity that represents real value (not a trivial login), then chart retention by signup week or month. Improving or stabilizing retention in your ICP cohorts is stronger evidence than a high signup week with a steep drop.
- Pick one primary retention event tied to the job-to-be-done.
- Compare ICP vs non-ICP cohorts separately.
- Look at shape over time; avoid claiming “good retention” without showing the curve to your team.
Qualitative PMF signals
- Inbound demand from the same ICP without heavy paid spend.
- Users insist on paying or expanding seats without being pushed.
- Word-of-mouth intros that match your ICP.
- Support and sales conversations converge on the same value prop language.
- You feel pull (more demand than you can fulfill) more often than push (constant convincing).
What not to do
- Do not invent industry-wide retention or NPS “benchmarks” without a primary source.
- Do not declare PMF from launch-week traffic or press alone.
- Do not average survey results across unrelated segments.
- Do not confuse high usage by free users who will never pay with commercial PMF.
Put this into practice on Bowora
When retention and ICP pull are real, public proof (reviews, rankings) reinforces trust for the next wave of customers — it does not create PMF by itself.
Common questions
- What is the Sean Ellis product-market fit survey?
- A survey asking how disappointed users would be if they could no longer use the product. In Sean Ellis’s framing, having around 40% or more say “very disappointed” is a positive leading indicator of product-market fit among engaged users — interpret it with retention and ICP context.
- Is 40% “very disappointed” a hard rule?
- No. It is a widely cited heuristic from Ellis’s work, not a universal law. Segment by ICP, pair with retention, and do not chase the percentage while ignoring whether users pay and return.
- What retention proves product-market fit?
- There is no single published retention number that proves PMF for every business. Track cohorts on a value event that matches your product, compare ICP vs non-ICP, and look for retention that stabilizes as you improve the core job — without inventing fake industry benchmarks.
- Can I have PMF without revenue?
- Consumer or freemium products may show engagement PMF before monetization. For B2B, willingness to pay and expansion are usually part of the story. Be explicit which kind of fit you claim — usage fit is not the same as commercial fit.
Sources
Facts, frameworks, and program details were checked against these first-party references. Last content review: August 8, 2026.