Product-Market Fit
Product-market fit is the point where a product satisfies real demand in a market strong enough to pull growth, not push it.
The two signals, side by side
What is Product-Market Fit?
Product-market fit is the point where a product satisfies real demand in a market strong enough to pull growth, not push it.
Marc Andreessen gave the term its most quoted definition in his 2007 essay "The Only Thing That Matters": "Product/market fit means being in a good market with a product that can satisfy that market." He argued it was the single variable that decided whether a startup lived or died, ahead of everything else on the standard startup checklist (pmarchive.com).
Andreessen popularized the phrase, but he didn't coin it. That credit goes to Andy Rachleff, cofounder of Benchmark Capital and later Wealthfront, who named a pattern he'd watched Sequoia's Don Valentine invest around for years. As Rachleff put it: "I put a name to that of product-market fit. Don didn't call it that. But it was really Don who came up with the concept." (unusual.vc). So the term has two camps, an originator and a popularizer, and Produck uses Andreessen's working definition because it's the one most founders already recognize.
What both men were describing is a state, not a milestone you check off once. A product can have fit in one market and lose it when that market moves. That instability is exactly why teams need a repeatable way to keep checking for it, not a one-time survey result.
Why it matters for product-market fit
Produck exists because product-market fit isn't a single verdict, it's an ongoing loop of Listen, Diagnose, Decide, Ship. You listen by collecting user feedback from every channel instead of guessing. You diagnose by sorting that feedback into patterns instead of reacting to whoever emailed loudest. You decide what to build next based on those patterns. You ship it, then you listen again. Fit shows up as an output of running that loop consistently, not as a box you tick after one big launch.
Two signals tell you the loop is working. One is qualitative: the Sean Ellis Test asks users how they'd feel if they could no longer use your product, and a strong "very disappointed" response tells you the market is pulling. The other is behavioral: retention curves that flatten instead of decaying to zero tell you people are actually staying. Neither signal is a one-time reading. Both drift as your market changes, which is the argument for building a loop around them instead of running a survey once and filing it away. We go deeper on why this distinction matters for early-stage teams in what product-market fit actually means and why startups obsess over it.
Most teams building toward fit are still working from a Minimum Viable Product, which is exactly why the loop matters early. An MVP is a hypothesis. Product-market fit is what happens when you've run enough Listen-Diagnose-Decide-Ship cycles that the hypothesis stops needing defending.
When it works, and when it doesn't
It works when:
- The team has a real feedback channel and actually routes what comes in to decisions, instead of letting it sit in a spreadsheet
- Retention and word of mouth are both trending up at the same time, so the signal isn't coming from one metric alone
It falls short when:
- The team declares fit off a single good survey result and stops watching, treating a market-level judgment as a permanent one
- The market itself shifts (a competitor ships, buyer priorities change) and nobody notices because the loop stopped running after the first "yes"
How to apply it
- Set up one place where feedback from support, sales calls, reviews, and in-app messages actually lands, so Listen isn't scattered across five inboxes.
- Run the Sean Ellis survey on your active users this month to get a baseline "very disappointed" percentage.
- Diagnose the feedback into themes weekly, not just when something breaks, so patterns surface before they become churn.
- Pick the one theme with the clearest overlap between frequency and revenue impact, and decide to build against it before adding anything else.
- Ship the change, then re-run the same survey question on the same cohort to see if the percentage moved.
- Repeat the loop monthly. Treat any single "we have fit" reading as a snapshot of that month, not a permanent status.
Frequently asked
How do you reach product-market fit?
You reach it by running a repeatable loop, not a single push: collect real user feedback, diagnose it into patterns, decide what to build based on those patterns, ship it, and repeat. Teams that treat fit as one launch to survive usually lose it as soon as the market shifts.
How do you know if you have product-market fit?
Run the Sean Ellis Test: if 40% or more of active users say they'd be "very disappointed" without your product, that's a strong signal. Pair it with retention data that flattens instead of decaying to zero, since one good survey result alone isn't proof the market is still pulling.
Sources
- The Only Thing That Matters, Marc Andreessen, pmarchive.com (2007)
- Andy Rachleff on coining the term product-market fit, Andy Rachleff, Unusual Ventures (2020)
- Using Product/Market Fit to Drive Sustainable Growth, Sean Ellis, GrowthHackers (Medium)