Quick Answer (TL;DR)
Product-market fit (PMF) means you have built something a specific group of people wants badly enough to pay for, keep using, and tell others about. The clearest test: survey your users and ask "How would you feel if you could no longer use this product?" If 40% or more say "very disappointed," you have PMF. Everything before PMF is search. Everything after is execution.
What Product-Market Fit Actually Means
The term "product-market fit" was coined by Andy Rachleff (co-founder of Benchmark Capital and Wealthfront) in the mid-2000s and popularized by Marc Andreessen in his 2007 blog post "The Only Thing That Matters." Andreessen defined PMF as "being in a good market with a product that can satisfy that market."
That definition is directionally correct but hard to act on. A more useful framing: PMF is the point where your product solves a real problem well enough that customers choose it, retain on it, and refer others to it without heavy sales or marketing pressure.
Before PMF, everything feels like pushing a boulder uphill. After PMF, the boulder rolls on its own. Users pull the product into their workflows. Support requests shift from "how do I use this?" to "can you add this feature?" Revenue grows even when you are not actively selling.
PMF Is Not Binary
PMF is not a light switch. It exists on a spectrum, and it varies by customer segment. You might have strong PMF with 50-person engineering teams but zero PMF with enterprise IT departments. Superhuman had PMF with power email users long before it worked for casual email users.
This matters because it changes how you measure and pursue PMF. You do not need the whole market to love your product. You need one well-defined segment to find it indispensable.
The History: From Andreessen to Sean Ellis
Marc Andreessen framed PMF as the single most important thing for a startup. But his description was qualitative. "You can always feel when product/market fit is not happening," he wrote. "Customers are not getting value, word of mouth is not spreading, usage is not growing that fast."
Sean Ellis made the concept measurable. In 2010, he introduced what is now called the Sean Ellis test (or the PMF survey). The core question:
"How would you feel if you could no longer use [product]?"
The response options are: Very disappointed, Somewhat disappointed, Not disappointed, N/A (I no longer use it).
Ellis found that products where 40% or more of users answered "very disappointed" almost always went on to achieve sustainable growth. Products below 40% struggled. This threshold has held up across hundreds of startups since.
The 40% benchmark is not a law of physics. It is an empirical pattern. But it gives teams something concrete to optimize against, which is far more useful than "you will know it when you see it."
How to Measure Product-Market Fit
No single metric captures PMF perfectly. Use multiple signals together.
1. The Sean Ellis Test (PMF Survey)
Send the "how would you feel" survey to users who have experienced your product's core value at least twice. Do not survey users who signed up yesterday. Do not survey users who have never completed your core workflow. You want input from people who have given the product a fair shot.
Target: 40%+ "very disappointed" responses.
Sample size: Aim for at least 40-50 responses to get a statistically meaningful signal. With fewer than 30 responses, a single user switching from "very disappointed" to "somewhat disappointed" swings the score by 3+ percentage points.
Survey timing: For SaaS products, send the survey after users have completed your core workflow at least twice. For consumer products, wait until the user has been active for at least two weeks. The goal is to survey people who have had a real chance to experience the value.
Superhuman ran this survey continuously and used the score as their north star. They started at 22%, iterated relentlessly on feedback from the "very disappointed" segment, and eventually crossed 58%. The key insight: focus on the users who already love you and make the product even better for them. Do not try to please everyone.
2. Retention Curves
Retention is the most honest signal. If users come back, your product is solving a real problem. If they do not, no amount of acquisition spending will save you.
Plot your cohort retention curve. A product with PMF shows a curve that flattens rather than declining to zero. For B2B SaaS, a month-2 retention rate above 80% and a month-12 rate above 60% are strong PMF signals. For B2C products, the bar is lower (month-1 retention above 40% is solid).
The activation rate matters here too. If users are signing up but not reaching your "aha moment," you have an activation problem, not a PMF problem. Fix the onboarding before concluding that the market does not want your product.
3. Organic Growth and Word of Mouth
Products with PMF grow organically. Users tell colleagues, share on social media, or invite teammates without being prompted. Track your ratio of organic to paid acquisition. If paid channels account for more than 70% of new users after 18 months, PMF is weak.
Other organic signals: inbound demo requests, unsolicited mentions on social media or community forums, and a growing waitlist. Slack grew from 15,000 daily active users at launch to 500,000 in less than a year, almost entirely through word of mouth.
4. Net Revenue Retention
For B2B SaaS, net revenue retention (NRR) is the clearest financial signal of PMF. NRR above 100% means your existing customers are spending more over time (through upgrades, expansion, or additional seats) faster than others are churning. Best-in-class SaaS companies run NRR above 120%.
5. Qualitative Signals
Pay attention to how customers talk about your product. PMF signals include:
- Users describing your product as "essential" or "can't live without it"
- Customers building workflows around your product
- Angry reactions when you consider removing a feature
- Unsolicited testimonials and case studies
- Customers defending your product in online discussions
A Step-by-Step Process to Find PMF
Finding PMF is not linear. It involves loops of building, measuring, learning, and iterating. But the following sequence gives you a reliable path.
Step 1: Pick a Narrow Segment
Do not try to serve "small businesses" or "product managers." Get specific. "Series A B2B SaaS companies with 5-15 person product teams who use Linear for project management." The narrower your initial segment, the easier it is to build something they love.
Use the TAM Calculator to make sure the segment is large enough to build a business on, even if you start narrow. A segment of 10,000 potential customers at $5,000 ACV is a $50M addressable market. That is plenty.
Step 2: Validate the Problem
Talk to 20-30 people in your target segment. Do not pitch your solution. Ask about their problems, current workarounds, and what they have tried. Use the Jobs to Be Done framework to understand the job they are trying to accomplish, not just the feature they think they want.
If fewer than half of your interviewees describe the problem as urgent or important, you are solving the wrong problem. Go back and pick a different one.
Step 3: Build Your MVP
Your minimum viable product should solve the core problem for your target segment and nothing else. Strip away every feature that is not directly related to the primary job-to-be-done. For detailed guidance on scoping and shipping your first version, see the MVP guide.
The goal of the MVP is not to impress people. It is to learn whether your approach to solving the problem works. Ship fast, measure retention, and talk to every early user.
Step 4: Measure and Iterate
Run the Sean Ellis test after users have had enough time to experience your core value (usually 2-4 weeks of active use). If you are below 40%, dig into the qualitative responses. The "somewhat disappointed" group tells you what is missing. The "not disappointed" group tells you who is not your target customer.
Iterate in tight cycles. Change one thing at a time so you can attribute improvements to specific changes. Track your Sean Ellis score weekly or biweekly. Superhuman improved from 22% to 58% over several months of focused iteration.
Step 5: Double Down on What Works
When you cross 40% on the Sean Ellis test and see retention curves flattening, you have early PMF. Now double down. Do not pivot to a new segment. Do not add features for a different audience. Make the product even better for the users who already love it.
This is the hardest discipline for founders. The temptation is to expand. Resist it until your core segment is locked in with high retention and organic referrals.
Real-World Examples
Slack: PMF Through Team Virality
Slack started as an internal tool at Tiny Speck (a game studio). The team realized their communication tool was more valuable than their game. They launched a preview in August 2013 and had 8,000 sign-ups on day one. Within 24 hours, it was 15,000.
The PMF signal was unmistakable: teams that tried Slack did not go back to email or HipChat. Retention was extraordinarily high. Growth was organic. Users invited their entire company without being asked. By February 2015, Slack had 500,000 daily active users and a $2.8 billion valuation.
What made Slack's PMF so strong? It solved a real problem (scattered team communication), the product was delightful to use (custom emoji, integrations, threading), and it had a built-in viral loop (every new team member made it more useful for everyone).
Superhuman: Engineered PMF
Superhuman is the textbook case of systematically engineering PMF using the Sean Ellis test. CEO Rahul Vohra shared the process in a widely-cited 2018 article.
When Superhuman first measured their Sean Ellis score, only 22% of users said they would be "very disappointed" without the product. Below the 40% threshold. Instead of panicking, the team:
- Segmented responses by user persona
- Focused only on the personas who scored above 40%
- Read every word of qualitative feedback from those users
- Built a roadmap based exclusively on what the "very disappointed" users wanted more of
- Ignored feedback from users who were "not disappointed" (wrong target segment)
Over several months, the score climbed to 58%. Growth followed naturally.
Airbnb: Persistence Through the Trough
Airbnb famously struggled before finding PMF. In 2008, the founders were selling cereal boxes to pay rent. The platform had listings but almost no bookings. They tried multiple iterations: air mattresses, rooms, whole apartments.
The breakthrough came when they visited their New York hosts in person and realized the listings had terrible photos. They hired a professional photographer, re-shot the listings, and bookings tripled. That single insight revealed PMF for a specific segment (budget-conscious travelers in New York wanting unique accommodations).
Airbnb's lesson: PMF sometimes requires doing things that do not scale. The photo project was manual and expensive per listing. But it proved the hypothesis and revealed the retention and growth signals that confirmed PMF.
Notion: PMF Through Flexibility
Notion launched in 2016 and struggled initially. The product was powerful but confusing. Users did not know if it was a note-taking app, a project management tool, or a wiki. Early retention was weak because the product tried to serve too many use cases at once.
The turning point came when Notion leaned into a specific segment: small startup teams who wanted one tool instead of five. By focusing on templates and use-case-specific guides, they helped this segment see how to use Notion's flexibility for their workflows. Retention jumped. The product spread through team referrals, and by 2020 Notion had over 4 million users.
The lesson: sometimes PMF is not about changing the product. It is about changing who you target and how you communicate the value.
Common Mistakes in the PMF Search
Mistake 1: Scaling Before PMF
The most expensive mistake. Founders raise a round, hire a sales team, and start running paid acquisition before retention proves the product works. CB Insights data shows that premature scaling is the number one reason startups fail. If your month-3 retention is below 30%, do not hire more salespeople. Fix the product.
For a deeper look at the pivot-or-persevere decision, which is closely related to PMF, see our breakdown of the signals that tell you which path to take.
Mistake 2: Confusing Initial Excitement with PMF
A successful launch on Product Hunt or a viral tweet can create a spike of sign-ups. That is not PMF. PMF is about retention, not acquisition. If those sign-up spikes do not convert into retained users after 30-60 days, you have awareness without fit.
Check your day-7 and day-30 retention numbers before celebrating launch metrics. A product with 10,000 sign-ups and 2% month-1 retention is in worse shape than a product with 200 sign-ups and 60% month-1 retention. The second product has PMF. The first has a marketing win.
Mistake 3: Building for Everyone
Trying to please all users dilutes your product. You end up with a tool that is acceptable to many people and essential to none. Narrow your focus. It is better to have 100 users who cannot live without your product than 10,000 who think it is "nice to have."
Mistake 4: Ignoring Distribution
PMF is not just about the product. It includes how customers find, evaluate, and adopt your solution. You can have a great product with no PMF if your distribution model does not work. A product that requires a 6-month enterprise sales cycle but is priced at $50/month does not have PMF, regardless of how good the product is. The product-led growth model works precisely because it aligns distribution with the product experience.
Mistake 5: Not Talking to Churned Users
Churned users are your best source of PMF feedback. They tried your product, experienced it, and decided it was not worth continuing. Exit surveys and churn interviews reveal the gap between what your product delivers and what the market needs.
Set up an automated exit survey that triggers 48 hours after cancellation. Ask two questions: "What was the primary reason you cancelled?" and "What would need to change for you to come back?" The answers cluster into patterns that point directly at what is missing from your PMF equation.
Mistake 6: Optimizing the Wrong Metric
Vanity metrics (page views, sign-ups, social followers) feel good but tell you nothing about PMF. Teams that optimize for top-of-funnel numbers often neglect the retention and engagement metrics that actually indicate fit. Focus on the metrics that prove users are getting value: activation rate, weekly active usage, and the Sean Ellis score. If those numbers are strong, the top-of-funnel metrics will follow.
Maintaining PMF Over Time
PMF is not a milestone you hit once and forget. It requires ongoing attention.
Monitor Leading Indicators
Track the Sean Ellis score quarterly. Watch retention cohorts monthly. Monitor NRR and expansion revenue. If any of these metrics start declining, investigate immediately. Small drops compound quickly.
Stay Close to Customers
The teams that lose PMF are the ones that stop talking to customers. As you scale, it is easy to rely on dashboards instead of conversations. Resist this. Run monthly customer interviews. Read every support ticket. Use the Founder Fit Assessment to pressure-test whether your product still aligns with what the market needs.
Watch for Market Shifts
New competitors, platform changes, regulatory shifts, and technology disruptions can all erode PMF. Notion had strong PMF as a team wiki, then faced competition from AI-native tools. Companies that survived disruptions (Netflix transitioning from DVD to streaming, for example) did so by detecting the shift early and iterating before retention cratered.
Build a quarterly competitive review into your planning process. Track three things: new entrants in your space, pricing changes from existing competitors, and shifts in how your customers talk about their problems. If the language customers use to describe their pain points is changing, your positioning may need to change with it.
Expand Deliberately
Once PMF is strong in your core segment, you can expand to adjacent segments. But treat each new segment as a new PMF search. What works for 50-person startups may not work for 5,000-person enterprises. The Product Strategy Handbook covers how to sequence expansion without losing your core.
PMF and the Broader Product Strategy
Product-market fit does not exist in isolation. It connects to everything else in your product strategy.
Before PMF: Your job is discovery. Run customer interviews, prototype rapidly, measure activation and retention, and iterate. Spend as little as possible on everything that is not learning. Use frameworks like Jobs to Be Done to stay grounded in real customer needs rather than assumptions.
At PMF: Document what is working and why. Capture your ideal customer profile, your core value proposition, the activation sequence, and the retention drivers. This becomes the foundation for everything that follows.
After PMF: Shift from search to execution. Build the go-to-market engine, hire the team, and scale acquisition. But keep measuring. PMF can erode faster than you expect.
PMF Checklist: Signals to Track
Use this checklist to assess where you stand. You do not need all of these signals to be positive. But if fewer than half are, you are still in the search phase.
- ☐ Sean Ellis score above 40% (surveyed users who completed core workflow)
- ☐ Month-1 retention above 40% (B2C) or month-2 retention above 80% (B2B SaaS)
- ☐ Organic acquisition accounts for 30%+ of new users
- ☐ Users complete the core workflow without hand-holding
- ☐ NPS above 40 or NRR above 100% (B2B SaaS)
- ☐ Support requests focus on feature requests rather than confusion
- ☐ Users describe the product as essential in their own words
- ☐ Revenue grows month-over-month without proportional marketing spend increases
- ☐ Churned user interviews reveal fixable gaps rather than fundamental misalignment
- ☐ At least one customer segment shows all of the above signals
The difference between startups that succeed and those that fail is rarely the idea. It is the discipline to keep searching for PMF until the signals are undeniable, and then the restraint to not break what is working as they scale.