How do I know I've reached product/market fit?
Short answer: You can't measure P/MF directly, but you can see its clues. Customers activate and keep coming back without being pushed. Your unit economics work, even if the company isn't profitable yet. A large share of active users would be very disappointed to lose you. And new customers arrive through referrals and word of mouth you didn't pay for. When those signals show up together, in one clearly defined group of customers, the market is telling you it wants what you built.
Every founder asks this at some point. Usually right after a good month.
The honest answer starts with what P/MF actually is. Then we can look for the evidence.
Start with the definition
Marc Andreessen put it simply: product/market fit means "being in a good market with a product that can satisfy that market."
Two halves. A good market. A product that satisfies it. You need both, and each one has clues of its own.
What makes a good market
A good market:
- Has the problem, and feels it often and badly enough to act.
- Can pay. A market that loves you but can't pay isn't a business.
- Is big enough, or growing. Broad, expanding, with room beyond your first customers.
- Is reachable. You can find them, and they can find you.
That last one gets ignored. If your customers aren't online, don't use social networks, never see ads, and you have no direct network into them, that's not a good market. Not for you, anyway.
P/MF is subjective
Here's the part founders don't like to hear: product/market fit depends on who's building.
I could build the exact same product as someone else and not have P/MF, because I don't have their acquisition channel. I don't have their access to the ICP. Same product, same market, different founder, different fit.
So when you ask "is there P/MF here?", the real question is "is there P/MF here, for us?"
What "a product that can satisfy that market" means
Satisfying the market means the product solves the problem, removes it, or does it for them, and does it cost-effectively.
Think in terms of the transformation. Before your product, they couldn't do X. Now it's easy. Or they don't need to do X anymore. Or it's done for them. The bigger and clearer that before-and-after, the stronger the fit.
The clues to look for (before $1M ARR)
P/MF can't be measured directly. You measure proxies, and you look for them together.
1. Activation and retention. People get to the value, and they come back without you nudging them. Look at retention by cohort: if each new cohort keeps a stable group of active users instead of decaying towards zero, the product is needed.
2. Unit economics that work. You don't have to be profitable on the bottom line yet. But each customer should bring in more than it costs to win and serve them, so you can reinvest what you make into growth. If every new customer loses you money, more customers won't save you.
3. "Very disappointed." Ask active users how they'd feel if they could no longer use your product. Sean Ellis's benchmark is 40% saying "very disappointed". The follow-up answers matter even more than the score: here's a P/MF survey template with the questions that matter.
4. Organic pull. They love it, they use it, they keep coming back, and they tell their friends. That shows up as referrals, word of mouth and inbound you didn't pay for. A quick test: turn off paid acquisition for a few weeks and watch what's left.
In B2B, two more checks
- Sales yield. As your team learns to sell, each salesperson should close more with the same effort. Leslie and Holloway call this the sales learning curve. If deals keep needing heroics, you're still selling something the market doesn't pull yet.
- Pull the plug. Ask your customers what would happen if you switched the product off tomorrow. If they'd shrug, you don't have P/MF. If they'd scream, you're close.
What doesn't count
- Sign-ups without usage. Interest isn't value.
- Compliments. "Great demo!" is not a signal. A renewal is.
- Revenue you bought. An aggressive marketing budget can fake growth for a while. It can't fake retention.
- Funding. Investors betting on you is not the market choosing you.
Common mistakes
- Averaging across segments. One group may retain beautifully while another bleeds. Your overall numbers hide both. Split by ICP before you decide anything, and if you're not sure who that is, here's how to define your ICP.
- Calling it on one signal. Great retention with broken unit economics isn't P/MF. Neither is a viral launch nobody returns to.
- Declaring victory once. P/MF moves with the market. Check the signals again whenever your ICP, offer or product changes.
What to do next
- Pull retention by cohort for your active customers, split by segment.
- Work out what one customer brings in versus what it costs to win and serve them.
- Run the Sean Ellis survey on active users only.
- Count how many of last month's new customers came from referrals or word of mouth.
If most of the answers point the same way, you have your answer. If they don't, they'll show you which variable is broken. Here's how to tell a P/MF problem from a go-to-market problem.
Not sure what your numbers are telling you? That's what a P/MF Sanity Check is for.