Cohort Analysis Design
Defines the cohort by an event rather than a date range, or you end up measuring seasonality.
Free — no card needed
Create a free account
to open Cohort Analysis Design — and the other 364 prompts across 21 categories.
Already have an account?
Running it, start to finish
- Define the cohort by a shared event, not a date range.
- Structure every comparison by period-since-event.
- Check what else changed between cohorts before attributing anything.
What you get back
The output this produces, every time.
- Defines cohorts by a shared event rather than a calendar window.
- Enforces equal-age comparison, which is the most commonly broken rule in cohort analysis.
- Names what varied between cohorts, since that frequently explains the difference.
Getting better results
Where this usually goes wrong, and how to avoid it.
- Never read down a calendar column. It compares an old cohort with a young one and concludes the old one is better, which is an artefact rather than a finding.
- Be careful with cumulative views. They always rise, which makes them reassuring and makes a deteriorating cohort invisible.
- Do not extrapolate an unflattened curve. If no cohort is old enough to show a floor, any lifetime value derived from it is a guess with decimal places.
More Data Analysis prompts
All Data Analysis →- Data Analysis
Segmentation Analysis
Produces segments that are actionable and stable, since statistically clean but unusable is the standard failure.
ClaudeChatGPT002Open → - Data Analysis
Metric Definition Review
Treats two teams reporting one metric differently as a governance problem rather than a query bug.
ClaudeChatGPT000Open → - Data Analysis
A/B Test Design and Reading
Designs a test that can actually detect the effect you care about, and states the decision rule before the data arrives.
ClaudeChatGPT000Open →
Written for The AI University. Every prompt in this library is original work — authored, tested and revised here, not collected from elsewhere. 365 of them, free with an account.