Ask ten founders if they have product-market fit and most will point at a revenue chart. Revenue is real, but it's a lagging, noisy signal, you can buy it with ads long before you have fit, and you can have fit for months before it shows up. The metrics that actually prove PMF are less obvious and more honest.
No single one is definitive. You triangulate fit from several. Here are the six that matter, ordered from strongest leading indicator to supporting evidence.
| Metric | What it measures | Type | The bar |
|---|---|---|---|
| PMF survey score | Share very disappointed to lose you | Leading | 40%+ |
| Retention curve | Whether cohorts flatten or decay | Leading | Flattens above zero |
| Engagement ratio | How often users return (e.g. DAU/MAU) | Leading | Model-dependent |
| Organic / referral growth | Growth you did not pay for | Lagging | Rising share |
| Churn | Rate users stop paying / using | Lagging | Low and falling |
| NPS | Willingness to recommend | Lagging | Positive, but secondary |
1. The PMF survey score (the leading indicator)
The single best early read on fit is the Sean Ellis survey: the percentage of engaged users who would be very disappointed to lose your product. The benchmark is 40%. It's leading because it captures how essential you are before that essentialness shows up in revenue or retention. It's the number PMFtracker is built around, and the one you should track over time.
2. The retention curve (does it flatten?)
Plot the share of each cohort still active over time. A curve that decays toward zero means no fit; a curve that flattens above zero means a group of users found lasting value, the clearest behavioral proof of PMF. The shape matters more than any single retention percentage. See retention curve vs the survey.
3. Engagement ratio
How often do users come back? A ratio like DAU/MAU (or weekly-active over monthly-active) captures habit. What counts as good is entirely model-dependent, a daily tool and a quarterly tool live at different numbers, but a rising ratio within your own product is a healthy sign that value is compounding.
4. Organic and referral growth
Growth you didn't pay for is growth that fit produced. If a rising share of new users arrive through word of mouth and referrals, people are recommending you unprompted, which only happens when the product genuinely matters to them.
5. Churn
High churn alongside a low PMF score is the classic no-market-need pattern. As fit improves, churn falls and the retention curve flattens. Watch the trend, not a single month.
6. NPS (useful, but secondary)
NPS measures whether people would recommend you; the PMF survey measures how essential you are. They answer different questions, and for fit, essentialness wins. Use NPS as supporting color, not the headline. See NPS vs the PMF survey.
Start with the one metric that leads the rest
Retention and revenue confirm fit after the fact. The Sean Ellis survey score reveals it first. Measure yours free against the 40% benchmark.
Measure your PMF score free → 14-day free trial · No credit cardHow to read them together
No single metric is proof. Real fit looks like a set of signals agreeing: a survey score above 40%, a retention curve that flattens, growing organic acquisition, and falling churn. When the leading indicators (survey score, retention) move first and the lagging ones (churn, revenue) follow, you're watching fit arrive. When only the lagging ones move, you're probably watching paid growth mask its absence.
Track all your leading PMF signals in one place
PMFtracker calculates your Sean Ellis score, tracks the trend, and segments your users, so the leading indicators of fit stop living in six different dashboards.
Start Tracking PMF → Set up in 5 minutes · No credit card required
