What Is the Sean Ellis Test? The 40% PMF Score Explained
The Sean Ellis test is a one-question survey that estimates product-market fit. You ask active users how they'd feel if they could no longer use your product, and you count the share who answer "very disappointed." If at least 40% say that, you likely have product-market fit.
Quick Answer: The Sean Ellis test asks users, "How would you feel if you could no longer use [product]?" with four options: very disappointed, somewhat disappointed, not disappointed, and not applicable. The score is the percentage choosing "very disappointed." A widely cited benchmark, popularized by Sean Ellis, holds that 40% or higher signals product-market fit. Below 40%, fit is likely still weak.
The appeal of the Sean Ellis test is its economy: one question, sent to the right people, produces a number you can watch over time. This piece explains the exact wording, where the 40% benchmark comes from, how to read your own score, and when it earns its keep.
How the Sean Ellis Test Works: One Question, Four Answers
The test is built around a single question with four fixed answer choices. You ask each respondent: "How would you feel if you could no longer use [product]?" The options are deliberately limited so the results stay clean and comparable.
The four choices are:
- Very disappointed — the only answer that counts toward your score.
- Somewhat disappointed — a signal of mild attachment, not fit.
- Not disappointed — the product is replaceable or unimportant to them.
- N/A (I no longer use it) — used to screen out churned or inactive users.
Only "very disappointed" counts. Your Sean Ellis score is the percentage of qualified respondents who pick that top option. "Somewhat disappointed" feels encouraging, but it does not move the number — the test is intentionally strict, because lukewarm attachment does not predict retention or word-of-mouth.
Who you survey decides whether the number means anything. Send it to recent, activated users — people who have experienced the core value of the product in the last week or two. Surveying everyone on your list, including trials that never activated and dormant accounts, dilutes the signal and usually understates true fit. This is why the Sean Ellis test lives alongside a broader PMF survey practice rather than replacing it.
Why 40%? The Origin and Limits of the Benchmark
The 40% threshold traces back to Sean Ellis, who noticed a pattern across the startups he advised. Companies where roughly 40% or more of users said they'd be very disappointed to lose the product tended to have found sustainable growth; those below that line usually struggled to grow no matter how hard they pushed on marketing. He turned that observation into a rule of thumb, and it stuck.
Rahul Vohra later refined the method at Superhuman. Rather than treating 40% as a single pass/fail gate, his team segmented on the "very disappointed" users — studying who they were, what they had in common, and which benefit they cared about most — then reshaped the product to convert more fence-sitters into that group. The test became an engine for improving fit, not just measuring it.
Treat 40% as a heuristic, not a law. It is a real, attributable benchmark, but it is a directional signal rather than a guarantee. A young product with a tiny, unrepresentative sample can clear 40% and still lack durable fit; a strong product surveyed too broadly can dip below it. Read the number as one input among several — alongside retention curves and revenue — not as a verdict on its own. For the wider set of gauges, see product-market fit metrics.
How to Read Your Sean Ellis Score
The score falls into three broad bands, and each implies a different next move. The table below is qualitative — it describes how to interpret where your percentage lands, not any promised outcome.
| Score band | What it suggests | What to do next |
|---|---|---|
| Well above 40% | Strong pull; users would feel a real gap without you | Study the "very disappointed" segment and double down on what they value |
| Around 40% | Borderline fit; promising but fragile | Segment responses, sharpen the core benefit, retest before scaling |
| Clearly below 40% | Fit is likely still missing | Learn from the disappointed minority and the reasons others feel indifferent |
Takeaway: The 40% line is a starting point for a conversation, not the end of one. Whatever your band, the highest-value analysis is reading why your most attached users care — that "must-have" reason is the raw material for improving the score.
Worked Example: Turning Responses Into a Score
Scoring is simple arithmetic once you have qualified responses. Imagine you survey recent, activated users and the counts below come back — these figures are illustrative, meant to show the method, not a benchmark.
- Very disappointed: 48
- Somewhat disappointed: 40
- Not disappointed: 22
- N/A / no longer use it: 10
First, remove the N/A responses. Those users are outside your active base, so you set aside the 10 and work from the 110 qualified answers that remain.
Then divide the top box by the qualified total. Forty-eight "very disappointed" out of 110 qualified responses is about 44% — above the 40% line. In this illustration, that points toward product-market fit, and the follow-up is to profile those 48 people: what job were they hiring the product for, and what would they use instead if it vanished?
The "somewhat disappointed" group is your growth reserve. They are attached enough to care but not enough to count, so learning what would tip them into "very disappointed" is often the clearest roadmap the test can give you.
Sean Ellis Test vs. NPS: What Each One Measures
The Sean Ellis test and Net Promoter Score answer different questions, and confusing them leads to bad reads. The comparison below is qualitative — it contrasts what each metric captures and when to lean on it.
| Dimension | Sean Ellis test | Net Promoter Score |
|---|---|---|
| Core question | How disappointed would you be to lose this? | How likely are you to recommend this? |
| What it gauges | Product-market fit and must-have status | Loyalty and referral sentiment |
| Key threshold | 40% "very disappointed" | A relative score, benchmarked by industry |
| Best stage to use | Early validation, before scaling | Ongoing, once you have a customer base |
| Main blind spot | Small or biased samples mislead | Says little about whether fit exists yet |
Takeaway: Use the Sean Ellis test to decide whether you have something people would miss; use Net Promoter Score to track advocacy once fit is established. They complement each other rather than compete.
When to Run the Sean Ellis Test in Validation
Run it once enough users have genuinely experienced your product's core value — and then run it repeatedly. Too early, and you're surveying people who haven't reached the "aha" moment, so the score is just noise. The right window opens after users have completed the main workflow a few times.
Aim for a sample large enough to trust. Practitioners generally look for at least a few dozen qualified responses before reading the percentage seriously; a handful of answers swings the number too wildly to act on. In Lean Analytics, Alistair Croll and Benjamin Yoskovitz make the broader point that a metric only earns its place when it's tied to a decision — and the Sean Ellis test fits that test cleanly, because its result tells you whether to keep iterating on fit or start scaling.
Treat it as a repeated measurement, not a one-off. Because fit shifts as your product and audience evolve, the score is most useful as a trend line you revisit after meaningful changes. For the deeper mechanics of administering and acting on the survey, see the Sean Ellis test for product-market fit.
Key Takeaways
- The Sean Ellis test is one question: "How would you feel if you could no longer use [product]?" with four fixed answer options.
- Only "very disappointed" counts toward the score; the percentage choosing it is your result.
- 40% is the benchmark popularized by Sean Ellis — at or above it, you likely have product-market fit; below it, fit is probably still weak.
- Rahul Vohra refined the method at Superhuman by segmenting on the "very disappointed" users to systematically improve fit.
- Survey recent, activated users and screen out inactive ones with the N/A option, or the number loses meaning.
- Treat 40% as a heuristic, not a law — read it alongside retention and revenue, never in isolation.
- The richest insight is qualitative: understanding why your most attached users would miss the product is how you raise the score.
Frequently Asked Questions
What is a good Sean Ellis test score?
A score of 40% or higher — meaning at least 40% of qualified users say they'd be "very disappointed" to lose the product — is the widely cited benchmark for product-market fit. Popularized by Sean Ellis, it's a directional heuristic rather than a hard rule, so read it alongside retention and revenue rather than as a standalone pass or fail.
Who should you send the Sean Ellis survey to?
Send it to recent, activated users — people who have experienced your product's core value in the last week or two. Including trials that never activated or long-dormant accounts dilutes the signal and usually understates real fit. The N/A ("I no longer use it") option exists precisely to screen churned users out of the scored group.
Is the Sean Ellis test the same as NPS?
No. The Sean Ellis test measures whether users would miss your product, gauging product-market fit with a 40% threshold. Net Promoter Score measures how likely users are to recommend you, gauging loyalty and referral sentiment. Use the Sean Ellis test in early validation to confirm fit, and NPS later to track advocacy once you have a customer base.
How many responses does the Sean Ellis test need?
Aim for at least a few dozen qualified responses before treating the percentage seriously. With only a handful of answers, one or two responses swing the score wildly, making it unreliable. The exact number depends on your user base, but a larger, representative sample of activated users gives a score you can actually make decisions on.