Product-market fit: the signals that matter (and the ones that fool you)

The moment that changes everything

Most startups don’t fail because they build the wrong thing. They fail because they think they’ve found product-market fit when they haven’t — and they scale prematurely based on that false signal. The result is burning through runway on growth efforts that can’t stick, hiring ahead of real demand, and watching metrics that looked promising collapse under scrutiny.

The tricky part: product-market fit isn’t a single metric you hit or a line you cross. It’s a state your product enters — and one you can lose. Understanding how to recognize genuine PMF (and distinguish it from lookalikes) is one of the most consequential skills a product manager can develop.

What product-market fit actually means

Marc Andreessen coined the term in 2007, describing it as “being in a good market with a product that can satisfy that market.” His description of what it feels like remains one of the best:

“You can always feel when product-market fit isn’t happening. The customers aren’t quite getting value out of the product, word of mouth isn’t spreading, usage isn’t growing that fast… And you can always feel product-market fit when it’s happening. The customers are buying the product just as fast as you can make it.”

But this visceral definition, while useful, doesn’t help you measure it. And if you can’t measure it, you can’t know where you stand — or whether you’re making progress toward it.

What PMF is not

Before we get to measurement, let’s clear out some common misconceptions:

• PMF is not having users. You can have thousands of signups and still be nowhere close. Free products especially can attract users who never become customers or advocates.
• PMF is not positive feedback. Early adopters are often enthusiastic by nature. They’ll tell you they love something while churning six weeks later.
• PMF is not initial traction. A spike from a Product Hunt launch or press coverage isn’t PMF — it’s attention. PMF is what happens after the spike.
• PMF is not revenue. You can generate revenue through aggressive sales tactics or heavy discounting without having a product the market actually wants.

The common thread: all of these can exist without retention. And retention is where PMF lives.

The Sean Ellis test: the 40% benchmark

Sean Ellis, who led growth at Dropbox and Eventbrite, developed the most widely-used quantitative test for product-market fit. It’s elegantly simple:

Ask your users: “How would you feel if you could no longer use [product]?”

Give them three options:

1. Very disappointed
2. Somewhat disappointed
3. Not disappointed

Ellis found that if 40% or more of your users say they’d be “very disappointed,” you’ve likely found product-market fit. Below 40%, you’re still searching.

This benchmark came from analyzing hundreds of startups. Companies that eventually scaled successfully almost always hit that 40% threshold first. Those that didn’t struggled to grow sustainably — no matter how much they spent on acquisition.

How Superhuman operationalized the 40% rule

Rahul Vohra, founder of Superhuman, took the Sean Ellis test and built an entire PMF measurement system around it. His approach, documented in his widely-read First Round Review article, has become a template for early-stage companies.

Superhuman’s process:

1. Survey continuously. They sent the Sean Ellis question to users who had used the product at least twice in the last two weeks — ensuring they were measuring people with enough exposure to have a real opinion.
2. Segment the responses. Rather than treating all users equally, they identified which user segments had the highest “very disappointed” scores. This told them who their product was really built for.
3. Double down on the right users. Instead of trying to please everyone, they focused on users who already loved the product — understanding what those users valued most.
4. Build what believers want. They asked “very disappointed” users what they’d most like to see improved, then prioritized that feedback over requests from lukewarm users.

Superhuman tracked their PMF score over time like a North Star metric. They went from 22% to over 58% by systematically applying this process. [INTERNAL_LINK: North Star metrics]

Retention-based signals: where PMF shows up in data

Lenny Rachitsky, former Airbnb PM, conducted extensive research on how successful companies measure product-market fit. His findings, drawn from interviews with leaders at companies like Slack, Figma, Notion, and Stripe, reveal that retention curves are the most reliable PMF indicator.

The flattening curve

When you plot user retention over time, pre-PMF products show a curve that keeps declining — eventually approaching zero. Post-PMF products show a curve that flattens out at some meaningful percentage.

The absolute number where it flattens varies by product category:

• Consumer social: 25%+ retention at day 30 is strong
• Consumer transactional (e-commerce, marketplaces): 20-30% monthly retention
• SaaS: 60-80% monthly retention, with best-in-class above 90%
• B2B with contracts: Net revenue retention above 100%

The shape matters more than the specific number. A curve that keeps dropping means you’re leaking users faster than the value you’re creating. A curve that flattens means you’ve found a core group who can’t imagine going back.

Cohort consistency

One of Lenny’s key findings: PMF shows up when your retention curve looks similar across successive cohorts. If your May users retain at roughly the same rate as your April users, you’ve found something repeatable. If each cohort looks wildly different, you’re still chasing signal through noise.

Organic growth indicators

When product-market fit is real, you’ll see growth that you didn’t pay for:

• Word of mouth: Users mention you without prompts. “How did you hear about us?” increasingly gets answered with “A friend told me.”
• Organic search growth: People start searching for your brand name, not just the problem you solve.
• Inbound interest: Press, investors, and potential hires start reaching out to you.

Brian Balfour, former VP Growth at HubSpot, calls this the difference between “pushing” and “pulling.” Pre-PMF, you push your product into the market. Post-PMF, the market starts pulling it from you.

Qualitative signals: what PMF sounds like

Numbers alone don’t tell the whole story. Some of the strongest product-market fit signals are things you hear, not things you measure.

Unprompted evangelism

When users recommend your product without being asked — in Slack channels, on Twitter, in casual conversation — that’s PMF speaking. Notion’s early growth was driven almost entirely by users sharing templates and screenshots, effectively doing the company’s marketing for them.

Users hacking around limitations

When people build workarounds to use your product for things it wasn’t designed to do, you’re onto something. Early Airtable users were building CRMs, project trackers, and content calendars on a platform that was technically just a “spreadsheet database.” That creative abuse signaled a core value proposition that exceeded the product’s current feature set.

The “magic moment” becomes obvious

You start hearing users describe the same aha moment in similar language. For Slack, it was the first time a team conversation happened there instead of email. For Dropbox, it was the first time a file synced across devices. When multiple users independently describe the same experience as transformative, you’ve found your value proposition’s center of gravity. [INTERNAL_LINK: user onboarding]

Complaints shift from “this doesn’t work” to “I wish this did more”

Pre-PMF, user feedback tends to be about basic functionality: bugs, confusion, unmet core needs. Post-PMF, feedback shifts to expansion: “I love this, but I wish it also did X.” The baseline expectation has moved from “prove you’re useful” to “you’re already useful, now grow with me.”

What to do before you have PMF

If you’re honest with yourself and recognize you’re not there yet, your priorities should be ruthlessly simple:

Talk to users obsessively

Not surveys. Conversations. Teresa Torres’s continuous discovery framework recommends talking to users every single week — not occasionally, not quarterly, weekly. At the pre-PMF stage, you need to understand not just what users do but why they do it, what alternatives they’ve tried, and what would have to be true for them to fully commit to your product. [INTERNAL_LINK: continuous discovery]

Narrow your focus

Superhuman’s key insight wasn’t “improve the product for everyone.” It was “find the users who already love us, and build for them.” Early-stage products often try to serve too many segments at once, ending up mediocre for all of them. It’s better to be essential to 100 users than nice-to-have for 10,000.

Don’t scale anything

This is the hardest discipline. Every instinct says “grow.” But spending on acquisition before PMF is like pouring water into a leaky bucket. Paul Graham’s famous advice: “Do things that don’t scale.” At the pre-PMF stage, that means doing things manually, learning intensively, and resisting the temptation to automate or delegate your way out of the hard work of understanding users.

Kill your darlings

If the data says a feature or direction isn’t working, let it go. Marty Cagan often notes that most product ideas — even from experienced teams — fail to move the metrics they’re intended to move. Pre-PMF, you need to iterate fast enough that you can try many bets and kill the losers quickly.

Common mistakes teams make when chasing PMF

Mistaking early adopter enthusiasm for PMF

Early adopters are optimists by definition. They’re excited by novelty and potential. Their enthusiasm can mask the fact that your product doesn’t yet work for the pragmatic majority who’ll eventually make or break your business. Watch what early adopters do, not just what they say.

Optimizing for the wrong metric

Teams often track acquisition (signups, downloads, trials) when they should be tracking activation and retention. A hockey-stick signup curve means nothing if the retention curve is a cliff. [INTERNAL_LINK: pirate metrics]

Declaring PMF too early

One good month isn’t PMF. One viral moment isn’t PMF. PMF is a sustained state, validated across multiple cohorts and time periods. Companies that declare victory too early often scale prematurely — then have to painfully contract when the numbers don’t hold.

Chasing growth before the foundation is set

This is the single most expensive mistake startups make. Pre-PMF growth spending doesn’t just waste money; it wastes time, creates organizational debt, and can mask the underlying problem until it’s too late to fix. Sequoia’s research suggests that premature scaling is the leading cause of startup failure.

Trying to boil the ocean

Adding features to please different user segments often delays PMF rather than accelerating it. Each new feature adds complexity, creates new expectations, and dilutes focus. The path to PMF is usually subtraction, not addition — removing friction, narrowing scope, deepening value for a specific use case.

What to do after you find it

Finding product-market fit isn’t the end of the journey — it’s the beginning of a new one. Once you have it:

• Document what’s working. Capture the user segment, the value proposition, the channels that work, and the onboarding experience that activates users. This becomes your playbook for scaling.
• Protect what you have. PMF can be lost. Changing the product, changing the market, or changing the competition can all erode your fit. Keep measuring, keep talking to users.
• Now you can scale. With retention working, acquisition spending becomes an investment rather than a burn. Growth efforts now compound rather than leak.

The next time someone asks if you’ve found product-market fit, don’t answer with fundraising news or user counts. Answer with your Sean Ellis score, your retention curves, and the words your most passionate users use to describe you. Those are the signals that actually matter.

Frequently asked questions