Turn churn interviews into a prioritized retention plan, not a list of complaints
Ten reasons customers gave you aren't ten problems to fix in parallel.
- Use it for
- PMs and founders who've done churn interviews and have a pile of reasons, not a plan.
Churn interviews reliably produce a list of reasons — too expensive, missing a feature, found a competitor, never got fully onboarded — and the naive next step is trying to address all of them, which spreads effort across everything and fixes nothing decisively. The reasons customers state aren't equally weighted or even always the real cause; some are the actual driver, some are convenient things to say when leaving, and treating them identically wastes the interviews you already did.
This prompt forces a weighting and prioritization pass on top of the raw reasons, instead of treating the interview notes as the finished analysis.
When not to use this
If you've only done one or two churn interviews, there isn't enough signal yet to prioritize confidently — do a few more first. This is for when you have enough interviews that patterns are starting to repeat.
Did this work?
You end up with a ranked short list of what to actually fix, tied to how much churn each one plausibly explains — not a restated list of every reason customers mentioned.
Tested on claude-opus-5. Evidence status is draft; it moves to battle-tested only on recorded runs, never by hand.