Step-by-Step Guide

How to Calculate Your PMF Score

Five steps to run the Sean Ellis product-market fit survey and interpret the result. Methodology, formula, and tips for product managers.

Last updated: April 2026

1
Identify your active users

The PMF score is only meaningful when answered by people who have used your product enough to form an opinion. Surveying anonymous signups or one-time visitors gives you noise.

Define active before you send anything.
Common rule: 2+ sessions in the last 14 days, most recent in the last 30 days.
Pull the matching list from your analytics or product database.
Filter out internal accounts, test users, and unconverted trial users.

Formula

Active user list = users with 2+ sessions in last 14 days, most recent within 30 days

Pro tip: If you have fewer than 100 active users, your PMF score will be too noisy to act on. Wait until you have a real cohort before running the survey.

2
Send the Sean Ellis survey question

Sean Ellis published the methodology in 2009 after benchmarking nearly 100 startups. The exact question matters. Don't paraphrase.

Lead with: "How would you feel if you could no longer use [product]?"
Provide four answers: Very disappointed, Somewhat disappointed, Not disappointed, N/A.
Add one open follow-up: "What would you use as an alternative?"
Keep the survey to two questions to maximize completion rate.

Formula

Survey response set = answers from your active user cohort

Pro tip: Aim for at least 40-100 responses. Below that, the percentage will swing too much for the result to be reliable.

3
Segment your respondents

The score only counts people who answered the core question. Drop the N/A group because they're not really users anymore.

Tag each response: very disappointed, somewhat, not disappointed, or N/A.
Remove N/A respondents from the denominator.
Look at the very-disappointed group separately for qualitative insight.
Read every very-disappointed follow-up answer.

Formula

Qualifying respondents = total responses - N/A

Pro tip: The patterns in very-disappointed users’ follow-up answers are the marketing copy that will resonate with future users.

4
Compute the percentage of very-disappointed users

The PMF score is the share of qualifying respondents who said they'd be very disappointed without your product. The formula is straightforward.

Count the very-disappointed responses.
Divide by qualifying respondents from step 3.
Multiply by 100 and round to the nearest whole percentage.
Example: 32 very-disappointed out of 80 qualifying = 40% PMF score.

Formula

PMF Score = (Very Disappointed responses / Qualifying responses) x 100%

Pro tip: Track the score by user segment. You may have strong fit with one persona and weak fit with another. The blended score can hide both signals.

5
Interpret against the 40% threshold

Sean Ellis's benchmark from his startup survey work: products at 40% or above tend to grow comparatively easily. Below 40%, you typically have to push hard on growth to compensate.

Above 40% indicates strong product-market fit.
Between 25% and 40% is a "getting closer" zone.
Below 25% suggests a positioning or product gap that growth tactics won’t fix.
Use segmented scores to decide where to focus next.

Formula

Status: above 40% (fit) | 25-40% (close) | below 25% (re-examine)

Pro tip: 40% is a benchmark, not a guarantee. Pair the survey result with retention curves and word-of-mouth data before drawing conclusions.

Score Your Product-Market Fit

Skip the spreadsheet math. Use our free PMF calculator with the Sean Ellis question, segmentation, and instant interpretation against the 40% threshold.

Open Free PMF Calculator

Frequently Asked Questions

Who created the PMF score and where did the 40% benchmark come from?

Sean Ellis, founder of GrowthHackers, developed the methodology and published it in a 2009 blog post. He arrived at 40% by benchmarking nearly 100 startups and noticing that companies above that threshold consistently saw easier growth and stronger word-of-mouth than those below it.

Why does the survey ask about disappointment instead of satisfaction?

Asking "would you recommend?" or "do you like our product?" invites polite over-reporting. Asking how disappointed someone would be if the product disappeared forces them to think about whether anything else could replace it, which is a much stronger test of fit.

How often should I run the PMF survey?

Once a quarter is a reasonable baseline. Run it more often if you've shipped a major change or pivoted positioning. Watch the trend across runs more than any single number.

Should I survey free users or only paying customers?

Survey active users regardless of whether they pay. The point is to measure whether the product is creating real value. If free users would be very disappointed but won’t pay, you have a pricing or packaging problem, not a fit problem.

What if I run the survey and only get 30 responses?

Don't compute the percentage. Treat the qualitative responses as the value of the run, read the open-ended answers, and try again next quarter when you have a larger cohort.