GA4 measures customer retention in three places: the Retention overview report (new versus returning users and retention by cohort), the cohort exploration (who came back, and who bought again, week after week), and audiences that let you act on returning or lapsed customers. It can only see what happens on your website or app, though. Renewals, cancellations and refunds live in your billing system or CRM, so a complete retention view joins the two.
Here is how to set GA4 up for retention, which reports to use, and where its view stops.
Where does GA4 show customer retention?
| Report | Where | What it shows |
|---|---|---|
| Retention overview | Reports > Retention (Life cycle collection) | New vs returning users, user retention and engagement by cohort, user lifetime value |
| Cohort exploration | Explore > Cohort exploration | Retention for cohorts you define, with your own return criteria such as a repeat purchase |
| User lifetime exploration | Explore > User lifetime | Lifetime revenue and engagement by first acquisition source |
| New / established / returning dimension | Most reports, as a comparison | Whether a metric is driven by new or returning users |
A returning user in GA4 is someone who has started at least one previous session. That is a low bar: for an online shop, a returning buyer matters far more than a returning visitor, which is why the cohort exploration is the more useful report.
How do you set up GA4 to measure retention?
- Send a user ID at login. Without it, GA4 counts devices, not customers. A customer who buys on a phone and returns on a laptop looks like two new users.
- Track the actions that mean value.
purchasewith a uniquetransaction_idfor e-commerce, or sign-in and key feature use for software. Mark the important ones as key events. - Set event data retention to 14 months. The default of 2 months limits explorations, including cohorts, to the last two months.
- Filter internal traffic, so your own team doesn’t inflate returning users.
- Get consent right. Visitors who decline analytics cookies can’t be recognized on return, which lowers measured retention in consent-heavy markets.
How do you build a cohort exploration?
- Open Explore and choose Cohort exploration.
- Set cohort inclusion to the first purchase (or first touch, for a content site).
- Set return criteria to another purchase, not just any event, if repeat buying is what you care about.
- Choose weekly or monthly granularity; daily is too noisy for most businesses.
- Break down by first user source or campaign.
That last step is where retention data starts paying for itself. It shows which channels bring customers who come back, which is often a different list from the channels that bring the cheapest first orders.
How do you act on retention data in GA4?
- Audiences of lapsed customers. For example, people who purchased but haven’t visited in 60 days. Share them with Google Ads for remarketing, or exclude recent buyers from acquisition campaigns.
- Predictive audiences, such as likely 7-day churning users, if your property meets Google’s minimum volume. Most smaller sites don’t.
- Triggered experiences. GA4 audience triggers can fire an event when someone joins an audience, which GTM can use for on-site messages; see real-time retargeting with GA4 audience triggers.
- Email and CRM follow-up. GA4 audiences don’t flow into email tools on their own; send the signal through your CRM or BigQuery. Our post on e-commerce marketing automation covers the follow-up flows.
What can’t GA4 tell you about retention?
- Subscription churn and renewals. These happen in billing, often without a website visit.
- Revenue retention. Expansion, downgrades and refunds belong in finance data.
- Who the customer is. GA4 does not store personal data, so it can’t tell you which named accounts are slipping away.
- Deep product behavior. For logged-in products, Mixpanel or PostHog give proper retention tables and per-user histories.
For Sell It Like Serhant, the website and the course platform were separate systems. We merged them into one user profile in Mixpanel, matched by email, so that a buyer’s journey from the site into the course could be followed as one customer (case study). That is the kind of join GA4 alone can’t make. For SaaS specifically, see GA4 for SaaS: key metrics to track.
How do you measure whether retention work pays off?
Compare cohorts rather than totals. Take the customers acquired before a change (a new onboarding email, a loyalty offer, a redesigned account page) and those acquired after, and compare their repeat purchase or return rates at the same age. Where you can, hold back a random group that doesn’t get the change. Revenue from returning customers in GA4’s e-commerce reports is a useful check, but final numbers should come from your shop or billing system.
FAQ
What is a returning user in GA4?
A user who has started at least one previous session on the same device or, with user ID, the same account. Previous sessions don’t need to have been engaged.
Can GA4 calculate churn rate?
Not directly. It can show users who stopped returning and cancellation events you send it, but churn rate should come from billing or the CRM.
How long does GA4 keep retention data?
Standard reports keep aggregated data. Explorations, including cohorts, are limited by event data retention: 2 or 14 months on standard properties. For longer history, use the BigQuery export.
Does GA4 retention include app users?
Yes, if the app sends data to the same property through a GA4 app stream. With a shared user ID, a customer who moves between web and app counts once.
Our GA4 consulting team sets up retention tracking that connects GA4 with your CRM and billing data; get in touch if you want help.
