left arrow icon

Ask every churned user what made them stop

Asking churned users what made them stop explains churn in their own words. In Groove's data, a multiple-choice exit survey got 1.3% completion, one open question by email 10.2%, and a reworded question about 19%; a request for a short call drew around five percent. A vendor puts post-cancellation emails at about 8%. Expect a minority to reply, and look for themes that repeat.

Evidence C, limited evidenceStrong when it fits

Stage
Stage 2. Users, no retention signalDo the first users get value, and do they come back?
Category
Activation and retention
Time
2 to 4 hours, then a week of repliesAwiser's estimate of the active work
Cash
No cash
Live calls
Live calls optional
First result
Within days
Risk
LowAn hour of emails. The risk is a few replies read as the reason everyone left.
Needs
A list of users who stopped, using a window that fits how often people use the productOne open question, sent personallyA place to group the answers by theme

When does it fit?

  • You have users who left and can reach them: Groove sent its exit question to every customer who canceled.[1]
  • You will ask one open question: at Groove, an open question by email got 10.2% responses, nearly eight times the multiple-choice survey's 1.3%.[1]

When should you skip it?

  • You have only a handful of churned users: at the reply rates found, ten people may give you one or two answers.[1][2]
  • People use your product monthly and you call them churned after two weeks: Amplitude says the right period depends on how often people naturally use the product.[4]

How do you run it?

  1. Define churn for your product

    Pick the inactivity window from how often people normally use it, not a fixed number of days.

  2. Write one open question

    "What made you stop using it?" No menu of reasons and no call to book.

  3. Send it personally to everyone

    A plain email from the founder to every user in the window.

  4. Quote replies word for word

    Keep their exact words and group them into themes.

  5. Offer a call only to those who reply

    Asking for a call up front drew far fewer answers; propose one once someone writes back.

  6. Fix the top theme, then ask again

    Act on the reason that comes up most, then ask the next group of churned users.

What usually happens?

  • Wording matters: Groove asked "what made you cancel?" instead of "why did you cancel?", and that framing nearly doubled replies, to about 19%.[1]
  • Vendors report modest rates: ProsperStack, which sells cancellation flows, puts post-cancellation emails at about 8%, and Clozd sees 25 to 35% of churned business accounts accept an interview in its programs.[2][3]

What does the evidence say against it?

  • The detailed founder data is old: Groove's figures predate the five-year window, and no recent documented case with reply rates was found.

How do you read the result?

What to count
Replies that give a reason, and the themes they fall into
When to read it
One week after sending
How to decide
Act on a theme once several people name it in their own words, and ask those who replied a follow-up question: Bob Moesta says talking to people who switched uncovers the underlying causes.[5]

What goes wrong?

  • Asking for a call first: at Groove, a request for a five-minute exit interview drew around five percent, though the conversations it produced were valuable.[1]
  • Expecting most people to answer: every rate found, from founders or vendors, sits well below half.[1][3]

How was it graded?

Evidence
C. Limited evidenceThe one-question email has a detailed founder dataset behind it, but those figures are old, the other rates come from vendors, and no recent documented case was found.
Verdict
Strong when it fitsCheap and direct once you have users who left. Expect a minority to reply, and treat themes that repeat, not single answers, as the signal.

How the grades and verdicts are decided

Sources

  1. 1
    How We Grew Our Customer Exit Survey Responses by 785% (Internet Archive copy) (opens in a new tab)Groove, Alex Turnbull. 2014. Founder report.
    We sent this survey to every customer who canceled.
    A whopping 1.3% survey completion rate.
    Not only was the response rate nearly eight times greater at 10.2%, but we were finally starting to get real, actionable data.
    A tiny difference in framing nearly doubled conversions, and the latter email got a roughly 19% response rate.
    Instead of "why did you cancel?" , we ask "what made you cancel?.
    We began emailing customers who canceled, asking to set up a five-minute exit interview.
    While the response rate wasn't great (around five percent), we did have some really valuable conversations and got great, honest feedback from former customers who didn't always give us straight answers right away.
    Read on October 20, 2026.
  2. 2
    2 Key Cancellation Survey Questions for Customer Retention (opens in a new tab)ProsperStack, Tony Sternberg. September 3, 2024. Vendor data, from a seller of what it measures.
    Hands-on methods like phone calls get great results, but are costly and time-consuming, while post-cancellation emails see low response rates (~8%).
    Read on October 20, 2026.
  3. 3
    Selecting Buyers for Post-Decision Interviews (opens in a new tab)Clozd, Jonathan Stevens. April 20, 2026. Vendor data, from a seller of what it measures.
    Churned Accounts: 25% - 35% participation
    Interview participation rates will vary by industry, job title, and transaction size.
    Read on October 20, 2026.
  4. 4
    Churn rate formula (opens in a new tab)Amplitude, Pragnya Paramita. March 15, 2024. Method, from a seller of what it measures.
    The best period to examine depends on how often you expect people will naturally use your product
    Read on October 20, 2026.
  5. 5
    Bob Moesta on Jobs-to-be-Done (opens in a new tab)Intercom, Des Traynor, with Bob Moesta. May 12, 2016. Method.
    If I talk to people who switched, I am able to uncover those underlying causes, and that is the DNA.
    Read on October 20, 2026.

Sources last checked on October 20, 2026. First published on October 20, 2026.

Awiser is the builder network where founders show their skills, find collaborators, and grow their projects with an AI copilot.