Introduction
Product-market fit is the state in which a product satisfies a strong demand from a well-defined market: enough people want it, badly enough, to use it, pay for it, keep it, and tell others. It is the milestone every early product is chasing and one of the hardest to measure honestly, because the signals (growth, retention, love) arrive noisily and the temptation to declare fit early is enormous. This article covers what product-market fit means, the signals and survey methods used to gauge it, and the research that finds it before the metrics confirm it.
What is Product-Market Fit?
Product-market fit (PMF) is the condition in which a product serves a real, sizable need in a specific market well enough that customers adopt it, retain it, and pull it into use with little forcing from the company. The term was popularized by Marc Andreessen in a 2007 essay arguing that the only thing that matters for a startup is getting to it: before fit, nothing else (team, features, marketing) rescues the company; after fit, the market pulls the product forward and the company's job becomes keeping up. The definition is deliberately experiential rather than numerical: you can usually feel fit (customers buying as fast as you can serve them, usage growing, press calling) and its absence (deals that stall, users who churn, features nobody uses), which is why the measurement methods below are approximations of a state rather than the state itself.
The Signals
Retention. The most reliable: a cohort retention curve that flattens (some stable share of users keeps coming back) rather than decaying to zero. Products without fit leak everyone eventually; products with it keep a core, the reading behavioral analytics provides most directly.
Organic growth and referral. New users arriving through word of mouth, not just paid channels; a rising NPS among engaged users is a leading indicator, with all its caveats.
Usage intensity. Frequency and depth of use among retained users: the product is becoming a habit rather than a trial.
Willingness to pay and low sales friction. Customers who buy without heavy persuasion, renew without heavy nurturing, and expand.
The "very disappointed" survey. Sean Ellis's widely used test asks users how they would feel if they could no longer use the product; a threshold around 40% answering "very disappointed" has become a folk benchmark for fit, with the caveat that it is a rough heuristic drawn from his experience across startups, not a validated cutoff, and that it works only when asked of people who have actually used the product enough to miss it.
Researching Toward Fit
1. Define the market narrowly.
Fit is always with a specific segment; "small businesses" has no fit, "bookkeepers at agencies with 10 to 50 clients" might. Audience definition comes first and is revised as evidence arrives.
2. Find the job and the pain.
Discovery interviews and jobs-to-be-done work to locate a problem people already spend money, time, or workarounds on; fit is rarely found for problems nobody is currently trying to solve.
3. Test the proposition before the product.
Value proposition and concept tests on target users gauge whether the promise lands; recorded reactions from a screened audience (a Ballpark study) show the gap between polite interest and real pull.
4. Segment the survey.
Run the disappointment question and ask who the product is best for and what its main benefit is; the segment that answers "very disappointed" describes the market where fit exists, and the product should double down there rather than averaging across everyone.
5. Interview the retained and the churned.
The core users explain what the product is really for; the churned explain what's missing or who it isn't for. Both are cheaper than guessing from the retention curve.
The Hazards
Declaring fit from early growth driven by spend rather than pull. Mistaking enthusiasm from a tiny early-adopter cohort for a market. Averaging survey results across segments and finding no fit anywhere when strong fit exists in one. Treating fit as permanent: markets move, competitors arrive, and a product that had fit can lose it, which is why the signals are worth tracking as a series rather than checking once.
The Bottom Line
Product-market fit is the point at which a specific market pulls a product into use and keeps it, felt in retention that flattens, growth that comes unpaid, usage that deepens, and users who would miss it. Measure it with cohort retention, the disappointment survey read by segment, and interviews with the retained and the churned; find it by narrowing the market, locating a real job, and testing the promise before the build. And keep measuring, because fit is a state the market grants and can withdraw.
Further reading
For the concept and its measurement:
Articles:
1. The Only Thing That Matters - Marc Andreessen
The 2007 essay that popularized the term, on why fit dominates everything else in an early company.
2. The Startup Pyramid - Sean Ellis
The origin of the "very disappointed" survey and its use as a fit heuristic.