Product-market fit is the most discussed and most misunderstood concept in startup land. Everyone claims they're "working toward PMF." Fewer people can articulate what it actually looks like, how to measure it, and — most importantly — what to do when you don't have it yet.
This guide cuts through the noise. It gives you the Sean Ellis framework, the key metrics, and a concrete 25-point checklist you can run against your product today.
TL;DR
- The 40% rule: If 40%+ of your users say they'd be "very disappointed" without your product, you likely have PMF — below that, you don't
- Retention is the ultimate test: If your week-4 retention curve flattens, you have PMF; if it keeps declining to zero, you don't
- Organic growth is the clearest signal: When users tell other people without being asked, something is working
- PMF is not binary: It exists on a spectrum, and you can have it in one segment before finding it in another
- Don't scale before PMF: Scaling a leaky bucket just empties your bank account faster
What Product-Market Fit Actually Means
Marc Andreessen coined the term in 2007, defining it as "being in a good market with a product that can satisfy that market." Simple to say, hard to achieve.
A more operational definition: Product-market fit is when your product solves a real problem so well that users want more people to have access to it. The behavior — not the sentiment — is what matters. Users who have PMF with your product recruit other users, resist churning, and get upset when you try to take features away.
The most memorable description came from Andreessen himself: "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 is happening. The customers are buying the product just as fast as you can make it — or usage is growing just as fast as you can add more servers."
That qualitative feeling is real. But you also need numbers.
The Sean Ellis PMF Survey: The 40% Rule
In 2010, Sean Ellis (the growth hacker who coined the term "growth hacking") developed the simplest and most reliable way to measure product-market fit: a single survey question.
The question: "How would you feel if you could no longer use [product]?"
Answer options:
- Very disappointed
- Somewhat disappointed
- Not disappointed (it isn't really that useful)
- N/A — I no longer use [product]
The benchmark: If 40% or more of respondents say "very disappointed," you likely have PMF. Below 40%, you need to improve before scaling.
Ellis developed this benchmark after testing with hundreds of startups. The 40% threshold is not arbitrary — it's empirically correlated with sustainable growth in his dataset. Companies above 40% were able to scale; companies below 40% that tried to scale burned through money without gaining traction.
How to Run the Ellis Survey
- Send it to users who have used your product at least 2 times in the past 2 weeks (active users only — non-users give you noise, not signal)
- Use a simple tool: Typeform, Google Forms, or Delighted
- Include a follow-up question: "What is the primary benefit you receive from [product]?" and "What type of person do you think would most benefit from [product]?"
- Run the survey when you have at least 30-40 respondents for statistical relevance
- Segment results by user type, company size, or use case — you may have PMF in one segment before another
What to Do When You're Below 40%
Ellis's insight: look at the respondents who said "very disappointed." These are your PMF segment.
- What do they have in common? (role, company size, use case, onboarding path)
- What primary benefit do they cite?
- What would they use instead?
The answers tell you what to double down on and what segment to target harder. Your job is not to make "somewhat disappointed" users love you — it's to find more of the users who already would be "very disappointed."
Net Promoter Score (NPS) as a PMF Signal
NPS measures a different dimension: willingness to recommend. The question is "How likely are you to recommend [product] to a friend or colleague?" on a 1-10 scale.
- Promoters: 9-10
- Passives: 7-8
- Detractors: 0-6
NPS = % Promoters − % Detractors
SaaS benchmarks (2025):
- Above 50: Excellent — strong PMF signal
- 30-50: Good — early signs of PMF
- 0-30: Fair — needs improvement
- Below 0: Poor — serious product issues
NPS is a lagging indicator of PMF, not a leading one. Use it alongside the Ellis survey and retention data, not as a standalone metric.
Limitation of NPS: It measures intention, not behavior. Someone who gives you a 9 might never actually refer anyone. Supplement NPS with actual referral tracking (how many users came from word-of-mouth).
Retention Curves: The Clearest PMF Signal
If there's one chart that predicts PMF more reliably than any other, it's the retention curve.
How to read a retention curve:
- X-axis: Time since signup (weeks or months)
- Y-axis: Percentage of users still active
The PMF pattern: The curve flattens after the initial drop. Some users churn in weeks 1-2 (normal), but the curve levels off and holds steady at week 4-8.
The non-PMF pattern: The curve keeps declining toward zero. This means even your most engaged users eventually stop using the product — there's no sustainable core of retained users.
Industry benchmarks for B2B SaaS:
- Week 1 retention: 40-60% (healthy)
- Week 4 retention: 25-40% (healthy)
- Month 3 retention: 20-35% (healthy — if the curve has flattened)
- Month 6 retention: 15-30% (acceptable if curve is flat)
If your month-6 retention is 15% but flat (not still declining), that's significantly better than a month-6 retention of 25% that's still dropping.
Tools to measure this: Amplitude, Mixpanel, ChartMogul (for revenue retention), or a simple cohort analysis in your own database.
Organic Growth Signals
Retention tells you if users are staying. Organic growth signals tell you if they're talking.
Before you have PMF, you drag every new user through manual outreach, paid ads, or cold email. After you have PMF, users bring users. The ratio of organic to paid acquisition shifts noticeably.
Organic PMF signals to track:
- Viral coefficient (K-factor): For every user you acquire, how many additional users do they invite? K > 1 = viral growth. K > 0.5 = meaningful organic lift.
- Referral source data: What percentage of signups say "I heard about this from a friend/colleague"?
- Unsolicited social mentions: People tweeting about your product without being asked
- Support ticket → feature request ratio: Pre-PMF teams get bug reports. Post-PMF teams get "can you add X so I can use this for Y" requests.
- Organic search growth: Rising search volume for your brand name is a PMF signal
The Product-Market Fit Checklist: 25 Items
Use this checklist every month. You're looking for movement in the right direction, not an overnight shift.
Customer Behavior
- [ ] 40%+ of active users would be "very disappointed" without your product (Ellis survey)
- [ ] Your retention curve has flattened at week 4 or later — it's no longer declining
- [ ] Users are returning more frequently over time, not less
- [ ] Users are expanding usage — using more features, inviting teammates, connecting integrations
- [ ] Users push back when you try to remove or change core features — this is one of the clearest signals of genuine dependency
- [ ] Users recommend your product without being asked — you hear about it through support tickets ("my colleague told me to sign up")
- [ ] Session depth is increasing — users are spending more time in the product per session, not less
Sales and Growth Signals
- [ ] Sales cycles are getting shorter — early adopters took 2 weeks to close; now similar profiles close in days
- [ ] Inbound leads are growing without proportional increase in marketing spend
- [ ] Your close rate is above 20% for qualified leads (B2B benchmark)
- [ ] Expansion revenue exists — existing customers are upgrading, not just staying on starter plans
- [ ] Net Revenue Retention (NRR) is above 100% — you're making more from existing customers than you're losing from churn
- [ ] CAC payback period is under 18 months for B2B SaaS (under 12 months = strong)
- [ ] Organic channels contribute 30%+ of new signups
Product and Feedback Signals
- [ ] User feedback is specific and feature-focused, not "it's confusing" or "I don't get it" (specificity = engagement)
- [ ] Power users emerge — 10-15% of your users use the product dramatically more than others
- [ ] Support volume hasn't grown proportionally with user growth — the product is getting easier to use without you
- [ ] Users are building workflows around your product — it's not a standalone tool anymore, it's part of their stack
- [ ] NPS is above 30 and trending upward quarter over quarter
- [ ] Users can articulate your value prop better than you can — their language for what your product does is cleaner than your own marketing copy
Qualitative Signals
- [ ] Press and media are covering you without you pitching — journalists are finding you through user word-of-mouth
- [ ] Competitive mentions increase — customers tell you "we evaluated [competitor] but chose you because..."
- [ ] Enterprise customers are asking to sign multi-year deals without being pushed
- [ ] You feel "pull" from the market — you're prioritizing the roadmap based on user demand, not founder intuition
- [ ] Your team is excited again — this is a soft signal but a real one. When PMF clicks, the energy in a company changes noticeably
How to Find PMF Faster: The Iteration Framework
The median time to PMF for B2B SaaS is 12-24 months. But teams that find it faster share a common pattern: they talk to users weekly, not monthly or quarterly.
The Weekly PMF Loop
- Monday: Review last week's retention and usage data. Identify 3 users who churned and 3 who expanded.
- Tuesday-Wednesday: Call or message the churned users (5-10 minute conversation: "What made you stop?"). Message the expanded users ("What made you come back / upgrade?").
- Thursday: Share what you learned with the full team. Identify the one change that would most impact the gap between your current Ellis score and 40%.
- Friday: Ship the change or create the task with a specific owner and deadline.
- Repeat.
Teams that do this consistently reach PMF measurably faster. A study by First Round Capital found that B2B founders who had weekly user conversations reached PMF in an average of 9 months vs. 22 months for founders who talked to users monthly.
The PMF Sprint
When you're far from PMF (Ellis score below 20%), consider a structured 6-week sprint:
Week 1-2: Survey all active users with the Ellis question. Identify your "very disappointed" segment.
Week 3: Conduct 10 qualitative interviews — 5 with "very disappointed" users and 5 with "not disappointed" users. Map the difference.
Week 4: Write a crisp hypothesis: "PMF exists for [specific persona] using the product for [specific use case]. We'll validate this by [specific change]."
Week 5-6: Ship the change. Re-survey. Measure movement.
Case Study: How AFFiNE Found Product-Market Fit
AFFiNE is an open-source knowledge management tool (docs, whiteboard, databases in one workspace) that grew from 0 to 60,000+ GitHub stars. Their PMF journey is instructive.
The early signal they almost missed: In the first 4 months, AFFiNE had thousands of GitHub stars but very low activation — people starred the repo but didn't use the product daily. Their Ellis score was in the low 20s.
The pivot insight: When the team conducted user interviews, they found a consistent pattern: the users who were "very disappointed" were all using AFFiNE for one specific use case — replacing Notion for structured docs with embedded whiteboard. This group was 15% of their user base but 80% of their "very disappointed" respondents.
The decision: Instead of trying to be everything to everyone, they doubled down on this specific use case. They improved the doc-to-whiteboard linking, improved the embedding experience, and made templates for this workflow.
The result: Within 8 weeks of shipping these changes, their Ellis score moved from 22% to 44%. GitHub organic traffic increased 3x as the "very disappointed" users shared the product more actively.
The lesson: PMF rarely comes from improving your average. It comes from finding the segment where the signal is already strong and serving them so well that they become your evangelists.
For growth tools and frameworks used by teams like AFFiNE, visit the growth tools directory.
What to Do Before PMF (And What NOT to Do)
Do Before PMF
- Talk to users weekly — the feedback loop is your most important product
- Narrow your ICP — serve fewer people better, not more people worse
- Reduce time-to-value — get users to their "aha moment" faster
- Remove friction from the core workflow — every click between signup and value is a leak
- Run the Ellis survey quarterly — track movement, not just score
- Find your power users and clone them — understand who they are and go find more of them
Don't Do Before PMF
- Don't scale paid acquisition — you'll spend money to acquire users who churn
- Don't hire a sales team — there's nothing to sell at scale yet
- Don't build for enterprise when your PMF is in SMB (or vice versa)
- Don't add features based on individual user requests before understanding the pattern behind those requests
- Don't rebrand or redesign — PMF is a product problem, not a marketing problem
The PMF Spectrum: Partial PMF Is Still Progress
PMF is not a binary switch. It exists on a spectrum, and partial PMF — strong signal in one segment or one use case — is a valid and valuable place to be.
Partial PMF patterns:
- Segment PMF: You have PMF with startups under 50 people but not enterprise
- Use case PMF: You have PMF for one specific workflow but not the broader platform vision
- Geographic PMF: You have PMF in the US market but not Europe (or vice versa)
In each case, the strategy is the same: go deep before you go wide. Serve your PMF segment so well that they become advocates who do your marketing for you. Then — and only then — expand to adjacent segments.
FAQ
What is the 40% rule for product-market fit?
Sean Ellis's rule: ask users "How would you feel if you could no longer use this product?" If 40%+ say "very disappointed," you have PMF. Below 40% means you need to improve before scaling. This benchmark was developed empirically from hundreds of startups and is the most widely used PMF measurement tool in the industry.
How do you know if you have product-market fit?
Key signals: 40%+ of active users would be very disappointed without your product, organic word-of-mouth growth, users complaining when you try to change core features, and retention curves that flatten after week 4. No single signal is definitive — PMF is confirmed by a cluster of signals moving in the same direction.
How long does it take to find product-market fit?
Median is 12-24 months for B2B SaaS. Some find it in 3 months (usually because the founder was living the problem and built the exact solution they needed), others take 4 years. The key metric is iteration speed — teams that talk to users weekly find PMF 2x faster than teams that talk to users monthly.
What comes before product-market fit?
Problem-solution fit: confirming the problem exists and your solution is directionally right. Validated by user interviews, not product usage. You need this before building anything significant. Problem-solution fit is confirmed when you can interview 10 people with the problem and 8 of them say "I would use this if it existed." PMF is confirmed when 8 of 10 active users say "I would be very disappointed if this went away."
Can you scale before product-market fit?
You can, but you shouldn't. Scaling before PMF accelerates burning money on leaky acquisition. The tell: if users churn before they get value, more users won't fix it — a better product will. The companies that scale before PMF and survive do so because they have enough runway to find PMF during the scale. Most don't.
The Bottom Line
Product-market fit is not a feeling. It's a measurable, observable state that shows up in your retention curves, your Ellis survey scores, your NPS, and — most viscerally — in how your users talk about your product to other people.
Use the checklist in this guide monthly. Run the Ellis survey quarterly. Talk to churned users. Talk to power users. Find the 20% of your user base where the signal is already strong, and serve them better than anyone else in the world could.
That's how you find PMF. Not by building more features. By finding the people for whom your product is already irreplaceable — and doubling down on them.
For more frameworks, tools, and templates to accelerate your SaaS growth journey, explore the complete growth tools directory.
Top comments (0)