Customer Cohort Analysis

Analyzes cohort retention data and returns the drop-off point plus concrete actions to fix it.

Best For

•Solo SaaS founders with a spreadsheet of signup cohorts who want to know exactly where users are dropping off.
•Growth-focused operators comparing cohorts before and after a product change who need the impact quantified clearly.
•Small teams without a dedicated analyst who need retention insights translated into specific next actions.

Prompt Template

You are a growth analyst running a cohort retention analysis. I'm going to describe or paste customer/user data. Produce a cohort analysis with actionable retention insights. Output: 1. Cohort Definition: how you're grouping users 2. Retention Table: cohort vs. time period matrix showing retention rates (or describe the pattern if you can't build the full table) 3. Key Findings: drop-off cliff, stabilization point, best cohort, worst cohort 4. Behavioral Correlations: what separates users who retain from those who churn 5. Recommended Actions: 1-2 to reduce early drop-off, 1-2 to improve long-term retention, 1 segment to double down on 6. Metrics to Track Going Forward Data: [DESCRIBE OR PASTE YOUR USER/CUSTOMER DATA]

Pro Tip

Note any product or pricing changes that happened alongside specific cohorts, since the model can only connect a retention dip to a likely cause if you tell it what else changed at that time.

Example Output

A sample of what this prompt produces once you fill in the placeholders.

A cohort analysis identifying a consistent 20-30% drop between Month 1 and Month 2 across all cohorts, crediting the March cohort's new onboarding flow for a 5-10 point retention improvement, and recommending that onboarding change be rolled out to all users.

Recommended Tools

Claude

ChatGPT

Airtable AI

How to use this prompt

Copy the prompt template using the button above.
Paste it into your preferred AI assistant (ChatGPT, Claude, Gemini, etc.).
Replace all bracketed placeholders like [TOPIC] with your specific details.
Send the prompt and refine the output as needed.