How to Measure Product-Market Fit: Signals, Metrics, and Frameworks

Marc Andreessen described product-market fit as "being in a good market with a product that can satisfy that market." He said you can always feel when it is happening: customers are buying as fast as you can ship, usage is growing, and you are scrambling to keep up.

That description is vivid but not useful for measurement. How do you know if you are at 30% PMF or 80%? How do you track progress toward it? And how do you know when you are losing it?

This guide covers the concrete metrics, surveys, and frameworks that make product-market fit measurable and actionable.

Table of Contents

1. PMF Is a Gradient, Not a Switch
2. The Sean Ellis Survey: The 40% Threshold
3. Retention Curves: The Most Reliable Signal
4. Organic Growth Signals
5. Engagement Metrics
6. PMF for B2B vs B2C
7. The PMF Dashboard: What to Track
8. Common Measurement Mistakes
9. Key Takeaways

PMF Is a Gradient, Not a Switch

The biggest misconception about product-market fit is that it is binary. You either have it or you do not. In reality, PMF exists on a spectrum.

The PMF Gradient

Most early-stage companies aim for "PMF for a niche" first, then expand. Trying to achieve broad PMF before nailing a niche is a common trap that dilutes focus.

Why Gradients Matter

If PMF were binary, you could not improve it. You would just wait for it to click. Because it is a gradient, you can measure your current position, identify which segments are closest to PMF, and focus product investment on strengthening fit in those segments first.

Superhuman is the canonical example. Rahul Vohra tracked PMF scores by user segment, found that freelance designers had 22% (not very disappointed), while startup founders had 48%. He focused product development on making the founder segment's score even higher, rather than trying to improve scores for segments with weak fit.

The Sean Ellis Survey: The 40% Threshold

The Sean Ellis survey is the most widely used PMF measurement. Sean Ellis (founder of GrowthHackers) tested multiple survey questions and found one that best predicted growth:

"How would you feel if you could no longer use [product]?"

>

- Very disappointed

- Somewhat disappointed

- Not disappointed

- N/A. I no longer use [product]

The 40% Rule

If 40% or more of your users answer "Very disappointed," you have PMF. Ellis found this threshold by benchmarking companies that went on to achieve strong growth. Nearly all of them scored above 40%.

How to Run the Survey

1. Who to survey: Users who have experienced the core value of your product. Not people who signed up yesterday and never came back. Target users who have completed your activation milestone (whatever that is).

1. When to survey: After the user has had at least 2 weeks of active use. Surveying too early captures first impressions, not fit.

1. Sample size: Minimum 30-40 responses for statistical relevance. 100+ is better.

1. Follow-up questions (critical for action):

- "What type of people do you think would benefit most from [product]?" (Identifies your target segment in users' own words)

- "What is the main benefit you receive from [product]?" (Reveals your positioning from the user's perspective)

- "How can we improve [product] for you?" (Directs product investment)

Segment Your Results

The aggregate score is less useful than segmented scores. Break results by:

• User type (role, industry, company size): Which segments have the highest "very disappointed" rates?
• Acquisition channel: Do organic users score higher than paid users? (Usually yes. Organic users self-selected for fit)
• Usage frequency: Do daily users score higher than weekly users? (Always)

Use the PMF calculator to analyze your survey data.

Retention Curves: The Most Reliable Signal

If you can only track one metric for PMF, track retention. Retention curves reveal whether users stick around after the initial novelty wears off.

How to Read a Retention Curve

Plot the percentage of users who return in each period (day, week, or month) after sign-up.

100% ─┐
      │\
      │ \
      │  \
      │   \_______________  ← Flattening = PMF
      │
      │\
      │ \
      │  \
      │   \
      │    \
      │     \___________\  ← Declining = No PMF
      │
  0% ─┴─────────────────
      D1  D7  D14  D30  D60

What Good Looks Like

The Flattening Test

The most important signal is whether the curve flattens. A curve that keeps declining means users are continuously leaving. A curve that flattens means a cohort of users has found lasting value. They are not going anywhere.

Example. Slack's retention curve in 2014-2015: Teams that sent 2,000+ messages had 93% retention at day 30. Teams that sent fewer than 200 messages had 45% retention. The product had strong PMF for high-volume communication teams and weak PMF for casual users. This insight drove Slack's focus on team activation (getting teams to 2,000 messages as fast as possible).

Cohort Analysis

Do not just look at overall retention. Break it into weekly or monthly cohort analysis:

• Improving cohorts: If more recent cohorts retain better than older ones, your product is getting better. This is a strong signal.
• Declining cohorts: If newer cohorts retain worse, something is breaking. Often because paid acquisition brings in less-qualified users.
• Stable cohorts: The product is neither improving nor declining. PMF is steady but not growing.

Organic Growth Signals

If your product has PMF, growth starts happening without you pushing it. These are the organic signals to watch.

Word-of-Mouth Ratio

What percentage of new users come from word-of-mouth, referrals, or direct traffic (vs paid acquisition)?

Calendly is a useful reference: by 2020, over 70% of their growth was organic. Every time someone sent a Calendly link, it was a product demo to a potential new user.

NPS as a Growth Indicator

Net Promoter Score measures willingness to recommend. While NPS has limitations (it measures intent, not action), scores above 50 correlate with organic growth.

Use the NPS calculator to compute your score.

Natural Virality

Some products have built-in viral loops:

• Collaboration tools (Figma, Notion, Slack): Every user invites others to use the product
• Network effects (LinkedIn, Marketplace apps): The product gets better with more users
• Content sharing (Canva, Loom): Output is shared publicly, driving awareness

If your product has a natural viral mechanism and growth is still slow, it is a PMF signal, not a marketing problem.

Engagement Metrics

Engagement metrics tell you whether users are getting value from the product on an ongoing basis.

DAU/MAU Ratio

The ratio of daily active users to monthly active users. Higher ratios indicate daily habit formation.

For reference: Facebook's DAU/MAU has historically been around 65%. Slack's was around 40% for active teams. A B2B analytics tool at 20% is doing well. See the DAU/MAU glossary entry.

Feature Adoption Rate

What percentage of active users use the core features?

If you have a project management tool and only 15% of users create tasks (the core feature), something is wrong with activation, not PMF. If 80% create tasks but only 5% use reporting, reporting may not be a PMF issue. It may just not be valuable. Track feature adoption for your core value proposition, not for every feature.

Depth of Use

Beyond frequency, measure how deeply users engage:

• Actions per session: How many meaningful actions does a user take per visit?
• Session duration: How long do users spend? (Be careful. Longer is not always better. For a tool, faster task completion = better product.)
• Feature breadth: How many different features does the typical user use? Broader usage often indicates deeper integration into the user's workflow.

PMF for B2B vs B2C

PMF looks different depending on whether you sell to businesses or consumers.

B2B PMF Signals

B2B PMF example. Linear (2020-2022): Linear knew they had PMF when engineering teams at high-growth startups started migrating from Jira unprompted. The switching cost was significant (years of data, workflow habits, integrations), but teams chose to pay it because Linear's speed and design were that much better for their workflow.

B2C PMF Signals

B2C PMF example. Duolingo (2013-2015): Duolingo's PMF was visible in retention. Users who completed their first lesson had 55% day-7 retention. Extraordinarily high for a consumer app. The streak mechanic created a daily habit loop that drove engagement and word-of-mouth growth.

Key Differences

The PMF Dashboard: What to Track

Build a dashboard with these metrics and review it weekly:

Leading Indicators (Early Signals)

1. Sean Ellis "very disappointed" %: Survey quarterly or after major changes. Target: >40%.
2. Activation rate: % of signups who reach the core value moment. Track weekly.
3. Organic acquisition %: % of new users from non-paid channels. Track monthly.

Lagging Indicators (Confirmation)

1. Retention curve: Weekly cohort retention, plotted over 30/60/90 days. Look for flattening.
2. Net revenue retention: For B2B. >100% means existing customers grow. Track monthly.
3. NPS: Survey quarterly. Target: >50 for strong PMF.

Engagement Metrics

1. DAU/MAU ratio: Weekly trend. Is engagement deepening or declining?
2. Core feature adoption: % of active users engaging with the primary value feature.
3. Actions per session: Is usage deepening over time?

Segment All Metrics

Every metric above should be segmented by:

• User type (role, industry, company size)
• Acquisition channel
• Account tier (free, paid, enterprise)
• Geography

Aggregate numbers hide the truth. You might have 35% Sean Ellis overall but 55% among startup founders and 15% among enterprise users. That is not "close to PMF". It is PMF in one segment and no PMF in another.

Common Measurement Mistakes

Mistake 1: Surveying the Wrong Users

Asking new signups about PMF. They have not experienced your product long enough to know. Survey users who have been active for at least 2 weeks and have reached your activation milestone.

Mistake 2: Confusing Satisfaction with Fit

Users can be satisfied (NPS 40) without having PMF. Satisfaction means they like it. Fit means they depend on it. The Sean Ellis question distinguishes between the two by asking about disappointment, not satisfaction.

Mistake 3: Ignoring Segments

"Our overall retention is 35%" might mean 60% retention for power users and 10% for casual users. Segment your metrics by user type, acquisition channel, and plan tier.

Mistake 4: Measuring Too Early

PMF metrics are meaningless with 50 users. You need enough volume for statistical significance and enough time for retention patterns to emerge. For most products, 3-4 months of data with 200+ users is the minimum for reliable measurement.

Mistake 5: Only Measuring Once

PMF is not permanent. Markets shift, competitors emerge, and customer expectations evolve. Monitor PMF metrics continuously, not just during the search phase.

Mistake 6: Treating PMF as an Excuse to Stop Improving

Even with strong PMF, product investment is required to maintain it. The market moves. Features that were delighters become table stakes. New competitors chip away at your advantages. PMF requires ongoing investment.

Key Takeaways

1. PMF is a gradient, not a switch. Measure where you are on the spectrum and focus on improving, not on declaring "we have PMF."

1. The Sean Ellis survey is the fastest PMF indicator. If 40%+ of users say they would be "very disappointed" without your product, you have a strong signal. Segment results by user type.

1. Retention curves are the most reliable confirmation. A curve that flattens = PMF. A curve that keeps declining = no PMF. Track weekly cohorts.

1. Organic growth is the ultimate confirmation. When users bring in more users without being asked (or paid), PMF is real.

1. Segment everything. Aggregate metrics hide the truth. You may have strong PMF in one segment and none in another. Find the strongest segment and double down.

1. B2B and B2C have different PMF signals. B2B PMF shows up in sales velocity, willingness to pay, and net revenue retention. B2C PMF shows up in viral coefficient, DAU/MAU, and day-7 retention.

1. PMF is not permanent. Monitor it continuously. Markets change, competitors emerge, and what was delightful last year becomes expected this year.