Cohort analysis
A cohort is a group of users who share a starting event — the week they first visited, the month they signed up. Cohort analysis follows each cohort over time so you can compare like with like. It separates 'are users behaving differently' from 'is the mix of users changing', which a single blended average can hide.
What this means
You assign each user to a cohort based on when they did something for the first time — typically first visit or signup. Then you measure a metric (retention, conversions, revenue) for each cohort across the periods that follow. The result is a grid: cohorts down the side, elapsed time across the top.
Why it beats a blended average
An overall average mixes brand-new users with long-tenured ones, so a change in the proportion of new users can move the average even if nobody's behaviour changed. Cohorts hold the starting point fixed, so you compare each group on equal footing and can see whether a product or marketing change actually improved the experience for later cohorts.
Watch cohort size: small cohorts give noisy curves, and incomplete recent cohorts (not enough elapsed time) should not be compared to mature ones.
- Cohort = users sharing a start event and period
- Separates behaviour change from mix change
- Don't compare immature cohorts to mature ones
How it appears in analytics and logs
A cohort grid shows how each starting group behaves over subsequent periods. A blended metric that looks flat can hide cohorts that are getting better or worse — the cohort view reveals it.
Diagnostic use case
Use cohorts to see whether retention or conversion is improving for comparable groups, instead of being fooled by a shifting blend of new and old users.
What WebmasterID can help detect
WebmasterID's first-party event timeline lets you group users by an acquisition period and read their later behaviour as a cohort.
Common mistakes
- Comparing a young cohort to a fully matured one.
- Reading noisy curves from very small cohorts as signal.
- Using a blended average where the user mix is shifting.
Privacy and accuracy notes
Cohorts are defined by a shared event and time bucket, reported in aggregate. WebmasterID builds them from first-party events without cross-site identity.
Related pages
- Retention rate
Retention rate measures how many users from a starting cohort come back in a later period. It depends entirely on definitions: what counts as 'returning', over what window, and which cohort. A 7-day and a 30-day retention rate answer different questions, and neither is comparable to a churn figure computed a different way.
- Churn rate
Churn rate measures how many customers (or how much recurring revenue) you lose in a period. Like retention, it is defined by choices: the window, what counts as 'churned', and whether you count customers or revenue. Customer churn and revenue churn can diverge sharply, so the basis must be stated.
- Segmentation for conversion analysis
Segmentation divides visitors into groups — by source, device, geography, or behaviour — so you can compare conversion within comparable cohorts. A single blended conversion rate can hide that one segment converts well and another barely at all. The discipline is choosing segments that answer a question without slicing so finely that each group becomes noise.
- Event Explorer
Group first-party events into cohorts over time.
Sources and verification notes
Last reviewed 2026-06-24. Facts are checked against primary/official sources where available; uncertain specifics are marked “Data not yet verified” rather than guessed.