Product-Market Fit Score

What product-market fit actually means

Product-market fit is the point at which a defined group of customers reliably gets enough value from your product that they keep using it, keep paying for it, and tell other people about it without being asked. It is a property of a pairing, not a product. The same software can have fit with one segment and none with another, which is why narrowing usually raises the score faster than building.

The most quoted proxy is Sean Ellis's survey question: how would you feel if you could no longer use this product? When more than 40% of active users answer 'very disappointed', companies historically found it much easier to grow. It is a proxy, not a law, and it is only meaningful with a decent sample of genuinely active users.

The second signal is the shape of the retention curve. Fit looks like a curve that declines and then flattens on a horizontal asymptote above zero. No fit looks like a curve that keeps sliding towards the floor, no matter how many users you pour in at the top.

Why this score weights retention over growth

Growth can be bought. Retention cannot. A team with a healthy paid channel and a leaking bucket will show impressive top-line numbers for two quarters and then stall, usually just after they have hired against those numbers.

When I was building Wocal, the number that carried the £2.7 million pre-money valuation conversation was not venue signups. It was how many of the venues onboarded in a given month were still active twelve months later. Investors who have seen a few of these know which chart to ask for.

So this calculator treats retention and activation as the spine of the score and treats demand quality as the evidence that the retention is commercially real rather than a free-tier habit.

The four inputs, and how to measure each one honestly

• Sean Ellis score: survey users who have used the product at least twice in the last two weeks. Anything under 40 responses is anecdote, not data.
• Retention flattening: cohort your users by signup month, plot the percentage still active each subsequent month, and look for the plateau. If there is no plateau, there is no fit.
• Activation rate: define one core value moment (for Wocal it was a venue publishing its first live listing) and measure the percentage of signups who reach it within seven days.
• Demand quality: does anyone pay list price without a discount, and does anyone arrive that you did not personally chase? Both are proxies for whether the pull is real.

What to do with a low score

A low score is not a verdict on the idea. In almost every case I have seen, a score under 35 means one of three things: the segment is too broad, the value moment is buried behind onboarding friction, or the problem being solved is real but not expensive enough for anyone to change their behaviour.

All three are diagnosable in about two weeks of proper discovery, which is far cheaper than another quarter of building. The mistake is treating a low PMF score as a signal to add features. Feature count and fit are close to uncorrelated.

If you want the structured version of that fortnight, the Discover and Validate stages of The Crossley Method lay out the interviews, the evidence bar and the kill-or-continue decision in order.