Retention Cohorts in KODAI show how groups of customers retain over time. This guide explains exactly what each part of the view means—axes, curves, and numbers—so you can interpret your data and act on it.
What you see on the page
The Retention Cohorts view shows a chart and usually a table. Each row is a cohort—a group of customers who started in the same period (e.g. the same month). Columns represent later time periods (month 1, month 2, month 3, etc.). The value in each cell is the retention rate or count for that cohort in that period: what share of the cohort was still active, or how many customers remained.
Cohort axis (rows)
The cohort axis is the vertical dimension: it lists the start period for each cohort. For example, "Jan 2026", "Feb 2026", "Mar 2026". Each row answers: of everyone who started in this month, how did they retain in the following months? Older cohorts have more data points (more columns filled); newer cohorts may only have one or two periods so far.
Time axis (columns)
The columns are the periods after the cohort start. "Month 0" or "Start" is the cohort's first period; "Month 1" is the next, and so on. So for the Jan 2026 cohort, Month 1 might be Feb 2026. This lets you compare retention at the same "age" across cohorts: e.g. Month 2 retention for Jan vs Feb starters.
What the numbers in each cell mean
Each cell is the retention metric for that cohort in that period. It can be shown as a percentage (e.g. 85% still active) or a count (e.g. 170 of 200). Percentage is useful to compare cohorts of different sizes. A drop from 90% to 70% between Month 1 and Month 2 for a cohort means 20% of that cohort churned in the second period. Darker or colored cells often indicate higher retention; lighter or empty can mean lower retention or no data yet.
Reading the retention curve
If the view shows a curve or trend per cohort, the curve plots retention over time for that cohort. A steep drop in the first months suggests early churn or onboarding issues; a flat curve after that is healthier. If newer cohorts sit below older ones at the same period age, retention has worsened over time—worth investigating pricing, product, or support.
How to use this for decisions
Use cohorts to see which signup periods are at risk and when. Then cross-check with Customer Intelligence: filter by segment (Failed, Expiring Soon) and see if the same cohorts show up. Combine cohort insight with targeted outreach or product changes to improve retention before the next dip.
In short
Rows = cohorts (start month), columns = periods after start, cells = retention rate or count. Read the curves and compare cohorts to spot drops and trends. Use this with Customer Intelligence to act on at-risk groups.