Home / Phase-6 / Cohort Analysis
Phase 6
Cohort Analysis
Phase progress0 / 0 completed
Cohort analysis is a behavioral analytics method that groups users by shared characteristics (e.g., sign-up date, feature usage, campaign source) and tracks their behavior over time. Unlike aggregate metrics, which can mask meaningful patterns, cohorts let product managers see when and why users engage, churn, or convert. This module introduces the foundations of cohort analysis, explores different cohort types, and demonstrates how they support retention, growth, and product-led strategies. By the end, you’ll learn how to design and interpret cohorts and use insights to drive both product improvements and revenue growth.
1) What is Cohort Analysis?
A method that groups users into cohorts who share a common starting point or behavior. It helps teams track retention, churn, and feature adoption over time—bridging the gap between analyzing all users at once and analyzing individual users.
2) Types of Cohorts
- Time-Based: Users grouped by sign-up date or activity window.
- Acquisition: Users grouped by acquisition source or channel.
- Behavioral: Users grouped by actions taken (or not taken).
- Segment-Based: Users grouped by attributes like location, plan type, or device.
- Size-Based: Comparing user behavior at scale (e.g., beta users vs. general launch).
3) How to Conduct a Cohort Analysis
- Set a goal: Define the question (e.g., “Improve Week-4 retention for new users”).
- Define cohorts: Choose the grouping key (time, source, behavior, segment).
- Select metrics: Retention, activation, conversion, ARPU, etc.
- Build a cohort table: Rows = cohorts; columns = time since start (D1/W1/M1…).
- Visualize: Use heatmaps or line charts to spot drop-offs or improvements.
- Interpret & act: Identify drivers, run experiments, and monitor subsequent cohorts.
Resources
Done with this module?
A/B Testing & Experimentation DesignNext Module
Next module:
Interpreting Qualitative vs Quantitative Data