A quick guide to read cohort charts for actionable metrics
Cohort analysis is a great way to find out more about user engagement by dividing users into cohorts. A cohort is a group of users who share a common characteristic such as the day, week or month of joining.
Cohort charts or tables can be pretty complicated and confusing at first, making it quite difficult to read/understand for most people. In the following example, we’ll look at different ways to analyze the patterns and make better sense of what your users are doing.
The first column shows the cohorts (month wise) in which users were acquired and the second column shows the total users for that cohort. Month 0,1,2….shows customer retention N months after acquisition. Assume each cell represents the percentage of returning users in a specific month.
Note: Cohort 1 = users joining in Feb; Cohort 2 = users joining in March……Cohort 7 = users joining in August
There are three main ways to read the chart:
- Horizontally(row): analyze a single cohort’s retention rates over time. Example: the decreasing retention rates for cohort 1 across 7 months
- Vertically(column): analyze month-wise retention rates among cohorts. Example: compare June, July and Aug metrics for cohorts 4, 5 and 6
- Diagonally: find retention rates for a specific month over all cohorts Example: compare Aug retention rates across all cohorts diagonally
