The 40% number arrives alone
You learn you're at 28% and the survey is over. Which users, which use case, which missing piece would move the number, the static form never asked.
The PMF survey that explains your 40%.
The Sean Ellis test tells you what fraction of users would be very disappointed without your product. It never tells you why, and the why is the roadmap. Diaform runs the PMF survey as a conversation, so every answer comes back with its reasons attached.
14-day free trial · No demo required
Where static PMF surveys fall short
The open questions get one-liners
"What is the main benefit?" produces "it saves time" a hundred times. The differentiated answer, what it replaced and what that was costing, needs a follow-up nobody asks.
Segments emerge too late
PMF usually exists in a segment before it exists overall. Finding that segment from checkbox data plus thin open text is guesswork; the evidence lived in answers nobody probed.
How Diaform runs the PMF survey
01
The Sean Ellis question, unchanged
Ask the standard question with the standard scale so your 40% benchmark stays comparable across quarters and against industry data.
02
AI probes each disappointment level
"Very disappointed" gets asked what specifically they'd lose. "Not disappointed" gets asked what they use it for and what would change their mind. Every segment produces evidence.
03
The substitute question, done properly
"What would you use instead?" is where positioning insight hides. The AI digs into how well the substitute would work and what would be lost in the switch.
04
Voice answers for founder-grade depth
Users explaining what your product means to their workflow out loud produce the kind of quotes that end up in board decks. Voice or text, their choice.
05
Summaries and sentiment per response
Every completed conversation returns a structured summary, sentiment, and answer confidence, so you read evidence, not raw transcripts.
06
Filter to find the fit segment
Slice responses to find where the "very disappointed" answers concentrate, by the attributes you asked about, and read exactly why that segment can't live without you.
How to run a product-market fit survey
- 01
Survey active users, not signups
The benchmark assumes people who recently experienced the product's core value, commonly those active in the last two weeks who've used it at least twice.
- 02
Send one link
In-app, by email, or in your community. Respondents answer in 3-5 minutes, by voice or text, in 30+ languages.
- 03
The AI runs the conversation
Standard PMF questions, plus live follow-ups tuned to each answer. You define the goals; the AI decides how deep each answer needs.
- 04
Read the number and the reasons
Your disappointed-percentage benchmark, plus summarized reasons per response, segment filters, and CSV export for the deeper cuts.
Frequently asked questions.
The practical details behind setting up, running, and scaling this kind of research with Diaform.
A product-market fit survey centered on one question: "How would you feel if you could no longer use the product?" Sean Ellis found that companies where at least 40% of users answered "very disappointed" tended to grow sustainably, and that number became the benchmark the industry still uses.
More conversations your team can run.
Explore the other ways Diaform can turn a static request for feedback into a useful conversation.
Your customers have more to say.
Get an AI feedback agent that listens, follows up, and turns every conversation into answers your team can act on.
Launch your first AI feedback agent in minutes.