How to Measure Product-Market Fit

Why PMF can't be captured by a single metric

PMF is a concept that Marc Andreessen popularized in his VC practice to mean "having a product that can satisfy a market," and from the outset it was never defined by a clear formula. Even so, in many teams you see PMF declared on the basis of a single number—"NPS passed a certain score" or "MRR grew by a certain percentage."

What makes this judgment dangerous is that the meaning of a metric changes with context. Healthy retention levels differ entirely between B2B SaaS and consumer apps, and in a seasonal business a short-term growth rate may merely reflect a temporary boom. The very posture of trying to reduce PMF to a single number invites "post-hoc justification"—picking a convenient metric after the fact to convince yourself.

Signal 1 — The Sean Ellis test and PMF surveys

The method proposed by Sean Ellis asks existing users, "How would you feel if you could no longer use this product starting tomorrow?" and has them answer with one of three choices: "very disappointed," "somewhat disappointed," or "not disappointed." In Hacking Growth, co-authored by Ellis and Morgan Brown, the idea is introduced that if roughly 40% or more respond "very disappointed," it serves as a rough benchmark for being well-positioned to pursue sustained growth. That said, this is only one rough benchmark, not an absolute standard that can be generalized while ignoring differences in industry or survey population.

Care is also needed in how you choose the survey population. In the practice that Superhuman's head of product Rahul Vohra described in First Round Review, rather than simply surveying all users, he narrowed the question to "high-expectation users" who had actually experienced the product's core value, removing the noise from light users on the verge of churning. If you look only at the numbers without designing who you ask, the results tend to come out more optimistic than reality.

Signal 2 — Retention curves and the "smile curve"

Surveys rely on self-report, but retention curves reflect actual behavior. If you divide users into weekly or monthly cohorts starting from their sign-up date and track the share who remain over time, most products drop sharply in the first few weeks and then level off at some level. When this leveling-off gently turns into an upward slope, growth practitioners call it the "smile curve," and it is commonly referenced by multiple growth writers, including the essays of Reforge founder Brian Balfour.

For example, if a SaaS product's 30-day retention plateaus in the low teens and then gradually reverses upward from there, that is taken as a sign that a core user base is repeatedly coming back. Still, this "level at which it reverses" varies greatly by industry and business model, so importing the absolute values from another company's case directly as your own target is dangerous.

Signal 3 — Organic traction and referral chains

States in which new sign-ups accumulate without increasing ad spend, or in which the share of inflow via referrals from existing users keeps rising, are strong corroborating evidence of PMF. In his book The Cold Start Problem, Andrew Chen discusses in detail how products with network effects acquire self-reinforcing traction, and he points out that this kind of traction is the sort of thing ad spend cannot substitute for.

On the other hand, organic traction is a lagging indicator that takes time to observe. If you mistake a one-off spike from a temporary social-media buzz or press-release reaction for sustained traction, you run the risk of committing to expansion investment when demand has not actually taken hold. It is essential to confirm it as a trend spanning multiple quarters rather than weeks.

What to change before and after PMF

Before reaching PMF, this is the phase in which you should narrow your target segment and concentrate resources on rapidly iterating between user interviews and the product. If you rush into GTM (go-to-market) investment or channel expansion at this stage, you end up pouring people and money into a product that doesn't fit, and the cost of correcting course later soars.

After several PMF signals start to line up, you enter the phase of shifting your center of gravity toward diversifying acquisition channels, refining onboarding, and designing growth loops. What becomes important here is to redefine the metric you chase clearly as your "North Star metric," and to keep monitoring cohort-by-cohort retention as a fixed-point observation.

Read next

To continuously confirm PMF signals, your precision improves if you have the procedure for reading retention at the cohort level down. Practical measurement methods are covered in "How to Read Cohort Analysis and Retention." We have also organized how the center of gravity of decision-making shifts before and after PMF, from a broader context, in "The Product Lifecycle and Decision-Making at Each Stage."