Cited from real sources 7 min read Written by Arun Agrahri Updated October 2026

A product-market-fit test by Sean Ellis

Sean Ellis Product-Market Fit Test: 40% Survey for Founders

The Sean Ellis product-market fit test uses one survey question and a 40% threshold to turn fit from a feeling into a number. Ellis coined the term growth hacking. He ran growth at Dropbox and Eventbrite before building this. Ask your active users one question. How would you feel if you could no longer use this product? The share who answer very disappointed is the signal. At 40% or more you have a leading indicator of fit. It shows up weeks before retention curves confirm it. It is the first of eight tests in the PMF Detection Kit.

Best for: Founders with active users who need a number that says whether they have product-market fit yet
Price: Free to read. To apply it to your own situation, Gavel has a free plan; Plus is $19/mo.

The framework

One question, three answers, one number

The test is a single survey question put to people who have used your product. It asks: how would you feel if you could no longer use it. You give them three options. Very disappointed, somewhat disappointed, or not disappointed. Only the first answer matters. Ellis found that the percentage who say very disappointed predicts whether a product is a must-have, and he drew the line at 40%.

Why a survey instead of a retention chart? Retention is a lagging signal; by the time a cohort curve flattens, months have passed. The very-disappointed percentage moves earlier and tells you the same thing. On Lenny's Podcast, Ellis frames what the question is measuring:

It's a simple question that helps you figure out: does anyone consider your product a must have, ideally who and how many people consider it. Ultimately it's trying to figure out, is your product a must have, which could be equated to having product market fit.

Sean Ellis on what the test measures Watch at 03:17

So the test is not a satisfaction score. Satisfaction tells you people are content; the very-disappointed number tells you they are dependent. Dependence is what product-market fit feels like from the inside.

The threshold

Why 40%, and how firm is it?

The 40% line came from comparing startups. It was not in the first version of the survey. Ellis started using the question as a filter, so he could hear from the customers who cared about the product instead of weighing all feedback the same. He was working with YC companies at the time and shared the question with other Silicon Valley startups. Over time he saw that products with a high enough share of very disappointed users mostly did well, and products that fell too low tended to struggle.

I don't think it's that firm, you know. To me, I think the real power is having some kind of target for the team to be shooting for, that basically says we're not going to aggressively start to grow until we hit this target.

Sean Ellis on how firm the 40% is Watch at 17:45

Ellis treats 40% as the point where the team agrees to step on the gas. A 37% would not make him call growth impossible. A 70% guarantees nothing if he has no idea how to grow the business. When Lenny raised Nubank, which waits for 50%, Ellis agreed that culture moves the answers, since some people answer more optimistically than others.

Four very disappointed users out of ten still tell you something, but Ellis would not go to market on that sample. He set a floor of 30 responses, a number he says he made up and later heard was the usual minimum.

How to apply it

How do you run the Sean Ellis Test?

Seven moves. The trap is reading the number you want instead of the one the very-disappointed users give you.

1

Survey people who reached core value, not signups.

Only ask users who have experienced the must-have action of your product. Surveying people who signed up and never activated pollutes the result; they cannot miss what they never used.
2

Ask the one question with three options.

How would you feel if you could no longer use this product: very disappointed, somewhat disappointed, not disappointed. Add a not-applicable option for people who already stopped, and keep the wording exact.
3

Count only the very disappointed.

The headline metric is the share who say very disappointed. Forty percent is the line Ellis drew between a must-have and a nice-to-have. Below it, you do not yet have fit.
4

Ignore the somewhat disappointed.

This is the seductive trap. Somewhat disappointed users like you but will not miss you. Optimizing the product for their requests drags you toward the average and buries the must-have signal.
5

Segment your very-disappointed users.

Dig into who they are, what they used before, and what problem they solve. That cluster is your real market, and it tells you who to build for next.
6

If you are below 40%, sharpen the must-have value and re-survey.

Reposition around what the very-disappointed group loves and strip friction from the path to it. Ellis describes getting one company to 40% in two weeks by doing exactly this.
7

Ask new users how they found you.

A second question, how did you find this product, surfaces your best acquisition channel for free, from the people who already chose you.

Who considers it a must have, how are they using the product, what did they use before, what problem are they solving.

Sean Ellis on what to do after the survey Watch at 14:47

The survey is the start, not the answer. The number tells you whether you have fit; the segmentation tells you who with.

Boundary conditions

When it works, when it fails

Works best when

You have enough active users who have reached core value to get at least 30 responses
You want an early read on fit, before retention cohorts have had time to mature
You can act on the result by segmenting the very-disappointed and sharpening their value
The product has a clear must-have moment a user can recall and miss

Fails when

You survey signups who never reached the core value and cannot miss the product
You treat somewhat disappointed as success and chase the nice-to-have crowd
You read the number as a vanity metric and never segment or act on it
Your category has a multi-year path to fit where no fast survey signal exists yet

The most common mistake is counting the wrong group. On Lenny's Podcast, Ellis is blunt about it:

Just ignore the people who say they'd be somewhat disappointed. They're telling you it's a nice to have.

Sean Ellis on the somewhat-disappointed trap Watch at 00:20

The honest caveat is category. The 40% survey is fast and cheap once you have a base of activated users. In long-arc markets where fit takes years, the very-disappointed number can stay low without meaning the bet is wrong. Pair it with the rest of the PMF Detection Kit before you scale or quit on one survey.

The sources

Where Ellis discusses this

02:53

Timestamp

Lenny's Podcast · The original growth hacker reveals his secrets

The test and the 40% threshold

The Sean Ellis test is explained in detail, including the specific question and response options. A 40% threshold for very disappointed users is presented as a leading indicator of product-market fit.

05:41

Timestamp

Lenny's Podcast · The original growth hacker reveals his secrets

Case study: repositioning Lookout to 40%

A case study of Lookout, a mobile security company, illustrates how the test identified a focus on antivirus, and repositioning plus streamlined onboarding quickly improved the product-market fit score.

07:41

Timestamp

Lenny's Podcast · The original growth hacker reveals his secrets

Build the flywheel around the must-have value

Sean outlines a growth strategy prioritizing activation and engagement before acquisition, emphasizing understanding the must-have value for customers and improving the initial user experience and time-to-value.

10:03

Timestamp

Lenny's Podcast · The original growth hacker reveals his secrets

Ask the right questions, like how did you find us

Direct customer feedback, even simple questions like how did you find us, is crucial for uncovering effective growth strategies and the channels that actually work.

Where experts disagree

Where operators disagree: ask people how they feel, or watch what they did?

Sean Ellis

trusts a single survey question put to active users, how would you feel if you could no longer use this, and reads 40 percent answering very disappointed as the fit signal. One number you can run this week and compare against the last time you ran it.

Bob Moesta

distrusts what people say in a survey and reconstructs what they actually did, interviewing recent switchers to find the push, pull, anxiety and habit that moved them. Stated preference is not behaviour.

Stated preference against revealed behaviour is a real methodological split, not a style preference. Ellis tells you whether you have fit; Moesta tells you why, which is the part you need before you can do anything about a bad number.

Useful? Pass it to a founder still guessing whether they have product-market fit.

Want the full playbook?

Get 108 product & growth frameworks.

35 frameworks 20 rules 45 heuristics & principles 51 operators

From Hamilton Helmer, Bill Carr, Anuj Rathi, and 48 more. Drop one .md into Claude, Cursor, or ChatGPT. Your AI cites practitioners, not guesses.

See the pack

Instant .md download · One-time purchase · No subscription

Related frameworks

Research

The Minimum Detectable Effect Rule

Ronny Kohavi

Pick the smallest lift worth catching, 5% or less, and size the A/B test for it before launch. A win from a low-power test is usually an exaggerated one.

Research

The Sample Ratio Mismatch Check

Ronny Kohavi

Check the traffic split before you read any metric. A 50.2/49.8 split on a million users means the test is broken.

Research

The Opportunity Solution Tree

Teresa Torres

One outcome at the root, unmet needs below it, solutions and assumption tests below those. Prune the rest.

Research

The Reference Customer Method

Christian Idiodi

Product-market fit is not a score. It is a count of people who will put their reputation behind you.

One question, three answers, one number

Why 40%, and how firm is it?

How do you run the Sean Ellis Test?

Survey people who reached core value, not signups.

Ask the one question with three options.

Count only the very disappointed.

Ignore the somewhat disappointed.

Segment your very-disappointed users.

If you are below 40%, sharpen the must-have value and re-survey.

Ask new users how they found you.

When it works, when it fails

Works best when

Fails when

Where Ellis discusses this

Where operators disagree: ask people how they feel, or watch what they did?

Get 108 product & growth frameworks.

Know when the next expert lands.