What is product-market fit?
Product-market fit is the point where a market pulls your product out of you faster than you can supply it. It shows up as retention that flattens instead of decaying, users who complain when it breaks, and sales that get easier. It is a state you observe, not a milestone you declare.
What it actually feels like
Marc Andreessen's original description is still the best one: you can always feel when it isn't happening — customers aren't getting value, word of mouth isn't spreading, sales cycles take forever. And you can always feel when it is: usage grows faster than you can add servers, money piles up, you're hiring as fast as you can.
The reason this matters is that fit is not subtle. Teams debating whether they have it almost never do.
Measure it, don't sense it
The retention curve. Plot each cohort's usage over months. Three shapes exist: decay to zero (no fit), decay then flatten (fit within a segment), and the smile curve where usage climbs back (strong fit and expanding value). A flattening curve means a stable set of people for whom the product is now infrastructure.
The 40% test. Ask users how they'd feel if they could no longer use the product. Superhuman built an entire product process around this question — segmenting to the users who said "very disappointed", ignoring the rest, and building exclusively for the segment that already loved it. That is the useful move: fit is found in a segment first, not in an average.
Why teams fake it
Because the alternative is admitting a year was wrong. Quibi launched with $1.75 billion, a full content slate and a fully-built product, and no evidence that anyone wanted premium short-form video on a phone-only app. Fit was assumed at the funding stage and tested at the launch stage, which is the most expensive possible ordering.
Slack, by contrast, found fit sideways — the internal chat tool built for a failing game company turned out to be the product. That only happens if you're watching what people actually use rather than what you planned to sell.
Seen in practice
Case studies where this shows up as a real decision, not a definition.
Related questions
How do you measure product-market fit?
Two common methods. The retention curve test: plot cohort retention over time and look for it flattening rather than decaying to zero — a flat line means a durable set of users. The Sean Ellis survey: ask users how they would feel if they could no longer use the product; above roughly 40% answering 'very disappointed' is the usual threshold.
Can you lose product-market fit?
Yes, and it happens more often than teams expect. The market moves, a competitor resets expectations, or a platform shift changes how people work. BlackBerry had extraordinary fit with enterprise email and lost it in about three years without changing its product at all.
Is product-market fit binary?
In practice it's closer to a gradient, but treating it as binary is more useful. Teams that describe themselves as having 'some' fit are usually rationalising weak retention, and the honest question is whether a specific segment loves it enough to be upset if it disappeared.
More on metrics
Last reviewed 2026-09-07