Cohort analysis groups users by a shared characteristic—most often the time period they first performed an action—and tracks their behavior over subsequent periods. Unlike aggregate metrics that obscure individual trajectories, cohort tables reveal whether retention is improving or deteriorating across successive user groups, making them the foundation of product-market fit assessment, LTV modeling, and churn management.
What This Cheat Sheet Covers
This topic spans 13 focused tables and 93 indexed concepts. Below is a complete table-by-table outline of this topic, spanning foundational concepts through advanced details.
Cohort TypesCohort Table AnatomyRetention Metric TypesSQL Patterns for Cohort TablesRetention Curve Shapes and PMF SignalsHeatmap Visualization and Pattern RecognitionLeading Indicators in Cohort AnalysisRevenue and SaaS Cohort MetricsCohort Cadence: Weekly vs MonthlyCohort Decomposition by SegmentBI Tool ImplementationsCommon Pitfalls and How to Avoid ThemChurn vs Survival Framing
Cohort Types
| Concept | Example | Description |
|---|---|---|
All users who first signed up in January 2024 | • Groups users by the calendar period of their very first action (signup, install, first purchase). The most common cohort type • anchors all downstream retention calculations to a fixed origin event | |
Users who completed onboarding within 3 days of signup | • Groups users by a specific action they did or did not take, regardless of when they joined • Useful for isolating the retention impact of a particular product experience | |
Users who made their first purchase between Black Friday and Cyber Monday | Acquisition cohorts defined by a discrete calendar window rather than a recurring period (month/week). Captures holiday or campaign effects on long-term behavior. | |
Users acquired via the APAC marketing campaign in Q1 | • Groups users by the region, country, or market in which they were acquired • Reveals whether retention curves differ by geography or channel |