The Sean Ellis 40% test is the most practical single question for measuring product-market fit. Ask your users "How would you feel if you could no longer use [product]?" When 40% or more say "very disappointed," you have strong PMF signal.

One question. One number. And a segmentation strategy that reveals everything else.

What Is the Sean Ellis 40% Test?

The Sean Ellis 40% test is a single survey question that measures product-market fit: "How would you feel if you could no longer use [product]?" If 40% or more of active users answer "very disappointed," the product has achieved product-market fit. Sean Ellis defined that benchmark after comparing results across nearly 100 startups.

Who Sean Ellis Is (Creator of GrowthHackers)

Sean Ellis coined the term "growth hacking" and created the survey this article is about. He ran early growth at Dropbox, was on the founding team of LogMeIn, led marketing at Uproar through its IPO, was the first marketer at Lookout and Xobni, worked with Eventbrite, founded the customer-insights company Qualaroo, founded GrowthHackers, and co-authored Hacking Growth.

In 2010, Sean published his PMF survey methodology. He was frustrated that founders didn't have a reliable, quantitative way to measure whether they were on the right track. Gut feeling and vanity metrics weren't cutting it.

So he built a one-question test. It has been used widely since, and it remains the most practical single question for the job. It is one input, not a verdict.

The 40% Test Explained: Question + Scoring

Here's the exact question:

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

Response options:

Very disappointed
Somewhat disappointed
Not disappointed
N/A

That's the test.

Your PMF Score = (Very Disappointed Responses / Total Responses) × 100

If 40% or more of your respondents say "very disappointed," you have a product-market fit signal.

Why These Specific Response Options?

The three options aren't arbitrary. Each maps to a distinct user relationship with your product:

"Very disappointed" = emotional dependency. These users need your product. They've integrated it into their workflow. Losing it would create real pain.
"Somewhat disappointed" = interested but uncommitted. They see value, but something's missing. They could switch to an alternative without much friction.
"Not disappointed" = no meaningful attachment. They might use your product, but they don't depend on it. Their feedback will often lead you astray.

Why 40%?

Ellis arrived at 40% by comparing results across nearly 100 startups, and he was candid about how firm that line is. In his own words:

"Admittedly this threshold is a bit arbitrary, but I defined it after comparing results across nearly 100 startups. Those that struggle for traction are always under 40%, while most that gain strong traction exceed 40%."

Read that pattern carefully, because it is asymmetric. Struggling companies were always below 40%. Only most of the companies with strong traction were above it. So a low score is a reliable warning, while a high score is encouraging rather than conclusive.

Broadly, the bands he described look like this:

Above 40%: growth tended to come more easily, with word-of-mouth doing real work
Between 25 and 40%: growth was possible but effortful, and churn kept reappearing
Below 25%: growth was difficult at every level

Ellis never published the vertical or geography of the startups in that comparison, so treat the number as a target and a directional signal rather than a calibrated benchmark for your specific market.

How to Run the Sean Ellis Survey

Step 1: Define Your Survey Population

Who should you ask? Active users who have used the product enough to have an informed opinion.

Don't ask:

Brand new signups (they haven't experienced your product yet)
Users who never activated (they don't represent your product's value)
People who already churned (they've already made their decision)
Anyone who signed up less than 2 weeks ago

Do ask:

Users active in the last 30 days
Users who completed a key action (your "aha moment")
Users with at least 2-3 weeks of real usage

Step 2: Get Enough Responses

There is no response count that makes a score valid. Ellis says 30 responses makes the survey directionally useful and that he is much more confident at 100 or more. Vohra puts the directional threshold at around 40 respondents. Report the respondent count beside the percentage: 16 "very disappointed" answers out of 40 is a 40% score with a 95% interval running roughly from 26% to 55%, before any selection bias.

Response rate tips:

In-app surveys get 10-30% response rates (much better than email)
Keep the survey short: the core question plus 2 or 3 follow-ups at most
Send at the right time, after a user completes a meaningful action rather than at random
Don't incentivize, because you want honest answers, not people clicking for a reward

Step 3: Ask Follow-Up Questions

The core question gives you the score. Follow-ups give you the "why":

"What is the main benefit you receive from [product]?"

From "very disappointed" users, this tells you what to double down on.

"How can we improve [product]?"

From "somewhat disappointed" users, this reveals the blockers preventing them from becoming advocates.

"What type of people would most benefit from [product]?"

This helps you refine your ideal customer profile.

Step 4: Calculate Your Score

PMF Score = (Very Disappointed ÷ Total Responses) × 100

Exclude "N/A" responses from the total.

Interpreting Your Results

Above 40%: Strong PMF Signal

You have product-market fit. But don't stop there:

Protect your core. Don't ship features that alienate your "very disappointed" users while chasing a broader market
Expand deliberately. Can you serve adjacent segments without diluting the core experience?
Keep measuring at a rhythm that matches how often people actually use the product, because fit can erode after bad releases, competitor moves, or market shifts
Study your "very disappointed" users. What do they have in common? That is where your ideal customer profile comes from

25-40%: Getting Close

You're in the "not quite" zone. Your product has value, but it's not essential yet. This is where most startups live, and where the most productive work happens.

Find your best segment. Which users are giving you the highest scores? Focus there
Address blockers. Ask "somewhat disappointed" users what is missing. Their most repeated 2 or 3 requests are your roadmap
Narrow your focus. You might be trying to serve too many user types at once. Pick the segment that already values you and build for it
Be careful about scaling. Spending heavily on growth before fit is usually how a weak score gets expensive

Below 25%: Fundamental Work Needed

The survey is telling you something is fundamentally off. That is a normal place to start. It does mean changing something significant rather than iterating at the edges.

Talk to users directly. A survey is not enough at this stage. Interviews will tell you more
Question your assumptions. Is the problem real and urgent? Is your solution actually addressing it?
Simplify. Often the problem is a product trying to do too much for too many people
Consider a pivot. Not necessarily a complete restart, but a meaningful shift in focus, audience, or approach

What 40% (or Less) Actually Means for Your Startup

Here's the honest truth: PMF isn't binary, and 40% isn't a magic line.

It's a Signal, Not a Verdict

Some successful companies started below 40%. Some companies above 40% still failed (usually from execution problems, not product problems). The metric is a guide, not a guarantee.

Context Matters

Ellis never characterized the vertical, market, or geography of the startups behind the 40% comparison, so anyone who tells you it is a SaaS number or a US number is adding something he did not say. Use 40% as the target, then pay attention to how your own score moves and how it differs across your segments.

Segment-Level Matters More Than Overall

Your overall score is almost meaningless in isolation. The real insight comes from segmentation: who loves you, who might love you, and who will never love you. A 35% overall score with 65% among power users tells a completely different story than a flat 35% across all segments.

PMF Is Earned, Not Discovered

Superhuman started at 22% in the summer of 2017. Segmenting down to the users who loved the product most brought it to 33%, and Rahul Vohra reports that "within just three quarters of our work to improve the product, the score nearly doubled to 58%." They did not find that fit, they built it through feedback collection, segmentation, and focused product iteration. Worth separating the two moves: the first jump came from measuring a better-defined audience, the second from changing the product. Their segmentation step used the high expectation customer framework, which Julie Supan originated and Vohra applied.

Alternatives to the Sean Ellis Test

The survey is the most practical starting point, and it works better alongside other evidence:

Retention Cohort Analysis

For a deeper look at how to know if you have product-market fit, retention data complements survey data. Look at your retention curves. If users are still engaged after 3, 6, 12 months, you have behavioral evidence of PMF. This complements the survey data with actual usage patterns.

Net Revenue Retention (NRR)

If net revenue retention is above 100%, existing customers are paying you more over time. That is a useful commercial signal alongside the survey, though it reflects pricing and packaging as well as product value.

Organic Growth Rate

What share of your new users arrive through word-of-mouth or organic search? Pull that you did not pay for is one of the more encouraging signals available, though the share that counts as high varies a lot by category.

The "Would You Recommend?" Question (NPS)

NPS asks about recommendation intent, which is a different question. "Would you recommend this?" asks about social willingness. "Would you be disappointed to lose this?" asks about dependence. For deciding whether a market needs you, dependence is the more direct signal, and the two can disagree in useful ways.

The Bottom Line

The Sean Ellis test is the most practical single question for measuring whether a market needs your product. One question, one number, and a segmentation habit that shows you who it matters to.

Run it with users who have had a fair chance to experience the core value. Calculate the score, report the response count next to it, and break it down by segment.

One reading is a state, not a baseline. The real work is understanding why some users would miss you, why others would not, and what you intend to do about it.

FitSignal runs this survey for you, with the score against the 40% line, persona breakdowns, and trend tracking, so the analysis does not live in a spreadsheet.