The observation
Ask a founder whether their product has found product-market fit, and watch what happens. Most shrug, or hedge, or reach for a metric they aren't quite sure proves the point. That hesitation is itself the most reliable signal in this whole subject — not because the founder is evasive, but because product-market fit is usually described as a feeling before it is described as a number.
Every practitioner in this wiki who has actually reached it agrees on roughly the same description: it is what happens when a product's value proposition lands squarely enough on a genuine struggling moment that users return, tell others, and resist leaving. Not a milestone on a roadmap. A state a product arrives in, or doesn't.
That distinction matters because a roadmap milestone is something you schedule — a launch date, a quarter, a target on a slide. A discovered state cannot be scheduled the same way. It can only be searched for, and recognised once it turns up.
The counterintuitive core
Eric Ries, author of The Lean Startup, offers the heuristic that cuts through every methodology built to detect this state: if you are asking whether you have product-market fit, you do not have it. Product-market fit is unmistakable when it arrives — the very fact that a founder still has the time and the doubt to ask the question is itself the answer.
This is the counterintuitive part. It reframes product-market fit from a target you engineer towards — set a metric, hit the number, tick the box — into a state you discover, usually later and more suddenly than planned. You do not schedule the moment users start coming back unprompted. You notice, after the fact, that they already are. The job changes accordingly: not 'build until we hit the date', but 'keep searching until the doubt itself disappears'.
What it looks like
The trouble with 'you'll know it' is that it gives a founder nothing to check against while they're still searching. Todd Jackson of First Round Capital supplies the corrective: fit is not a single on/off switch but a spectrum, and his PMF Levels framework describes four stations on it.
- L1 — Nascent. A few customers, significant churn or stagnant usage, and a team still searching for the right ideal customer.
- L2 — Developing. Consistent first use and some retention; customers see value but would not call the product essential, and word of mouth is still rare. Most B2B companies stall here — 60% never pass L2 — usually because they have mistaken an important problem for an urgent one, or a 'nice to have' for a 'must have'.
- L3 — Strong. High retention, a strong net promoter score, customers who turn into advocates, and a sales cycle that shortens as category awareness grows.
- L4 — Extreme. Customers embed the product in critical workflows and would actively fight to keep it; competitors reference you to explain what they do.
Progress across the levels is assessed on three dimensions at once — satisfaction (would customers be 'very disappointed' without it?), demand (does it pull customers in, or does it need constant pushing?), and efficiency (can it be acquired and delivered repeatably, not just heroically?). All three must be strong together for a level to count. To pressure-test the demand dimension in particular, Jackson offers the $100 vending machine: imagine a machine in the customer's office that dispenses your product for $100 a use — no relationship, no sales process, no service agreement. Would your target customers pay, and how often? The question forces past 'is this useful?' — almost everything is useful when it is free — to 'is this essential at a cost?', and it surfaces who the real buyer is: the person willing to spend $100 from their own budget, not just the one willing to click through a free trial.
One further nuance worth holding onto: the level is a property of a product for a given customer segment, not of a company as a whole. A company can sit at L3 with the segment it understands well and back at L1 with a new segment it is only starting to learn — which is exactly why 'do we have product-market fit?' so often gets a hedge for an answer rather than a clean yes or no.
The honest limit
'You'll know it' is a heuristic, not a metric — and it is worth naming that limit plainly rather than dressing the feeling up as a measurement. Ries's rule tells a founder what the destination feels like from the inside. It does not, by itself, give anyone a number to track on the way there.
Uri Levine, co-founder of Waze, puts a harder edge on the same idea: 'Product market fit have one metric... Retention. That's really simple. If you create value, they will come back.' That is the proxy this series will make measurable — a later lesson turns the feeling into an actual number worth tracking, rather than leaving it as something you recognise only once it has already arrived.
Go deeper
The single best source for the idea in this lesson is Eric Ries on Lenny's Podcast — watch it here, or read the episode.