<!-- What Is Cohort Analysis? Dating Business Glossary | https://whitelabeldating.com/glossary/cohort-analysis/ | Run by High Intent Media, which also runs Dating Partners and DatingAPI. -->

# What is cohort analysis?

> Cohort analysis is a way of measuring a business by following groups of members who started at the same time, such as everyone who first paid in March, and tracking what each group does in later months. It shows whether retention is really improving.

## How does cohort analysis work?

You group members by when they started, then track each group over the following weeks or months. Google Analytics describes a cohort as a group sharing a common characteristic, set by "inclusion criteria" such as the date a member first visited, and "return criteria" that they must meet to count in later periods. Cohorts can be daily, weekly or monthly.

Stripe's billing analytics does the same for payments: it puts subscribers into cohorts by the month they started paying and shows, month by month, how many are still active and how much revenue the cohort still brings in.

A worked example with round numbers (an example, not a benchmark):

| First paid in | Payers at start | Month 1 | Month 2 | Month 3 |
|---|---|---|---|---|
| January | 200 | 150 (75%) | 120 (60%) | 100 (50%) |
| February | 220 | 176 (80%) | 145 (66%) | |
| March | 250 | 210 (84%) | | |

Each row is a cohort. Reading down a column compares cohorts at the same age. Here each newer cohort keeps more payers in month 1, which suggests something improved.

## Why does cohort analysis matter for a dating site or app?

It stops growth from hiding problems. A dating brand that adds many new payers each month can show rising totals while each group leaves faster than the last. A single [churn](/glossary/churn/) figure blends old loyal payers with new ones; cohorts separate them.

Cohorts also show the effect of a change. If you change onboarding, prices or member checks in April, compare the April cohort with earlier ones at the same age. That is the most honest way to see whether [retention](/glossary/retention/) improved.

## How is cohort analysis different from a monthly report?

A monthly report shows a snapshot of everyone at once; a cohort table shows each group's path over time.

| | Monthly report | Cohort analysis |
|---|---|---|
| Unit | All members this month | Members who started together |
| Good for | Totals, cash, targets | Retention, the effect of changes, LTV |
| Can hide | Worsening retention behind growth | Little, once enough months pass |

## What should you ask a provider about cohort analysis?

Ask whether the platform shows cohorts by sign-up month and by first payment month for your own brand, and whether you can export the raw data. Ask whether cohorts can be split by traffic source, so you can see which channels bring members who stay. Our guide to [dating app analytics](/guides/dating-app-analytics/) covers the rest of the reporting set.

## Common questions

### Should cohorts be based on sign-up date or first payment date?

Use both for different questions. Sign-up cohorts show how well you turn new members into active and paying ones. First-payment cohorts show how long payers keep paying.

### How many months of data do I need?

You can start after two or three months, but the picture is clearer after six to twelve, because many dating plans run for several months.

### Can I do cohort analysis in Google Analytics?

Yes, for activity. Google Analytics has a cohort exploration report. For payment cohorts you usually need your billing data, because analytics tools do not always see renewals.
