The GA4 User Retention Report

This guide is built from Google's Analytics Help articles on the Retention overview report, overview reports, the Life cycle collection, user metrics, the dimensions and metrics reference, cohort exploration and data retention. Quotes are Google's wording; the section on how we use the report is our own practice.

What the report shows

The Retention overview is one of GA4's pre-made overview reports. Google says it "summarizes how well your website or app retains users, including the average amount of time users spend engaged on your site or app after they were first acquired and the percentage of users who return each day in their first 42 days."

An overview report is a page of summary cards, and Google notes that "each topic has only one overview report." There is no separate Retention details report: the cards are the report. When you need more control over who counts as retained, you move to a cohort exploration, covered below.

Where to find it

1. Open Google Analytics and select Reports on the left.
2. Click Life cycle > Retention.

If you don't see a Life cycle section, your property may use business-objective reports instead. Google says the Life cycle collection appears when you select "Get baseline reports" while creating a property, and also by default if "you created the property before March 27, 2023" or migrated from Universal Analytics with Setup Assistant. Google also notes: "You won't see the report in the Business objectives collection by default, but an editor or administrator can add the report to your left navigation."

The six cards

By default the report has six summary cards. Google adds that "administrators and editors can change the cards so you may see different ones from these," so your property may differ.

Card	What it shows	Window
New users and Returning users	First-time visitors and people who have visited before, as a total and per day	Your date range
User retention by cohort	The share of each day's new users who came back after 1, 7 and 30 days	Day 1, 7, 30
User engagement by cohort	Average engaged time of each day's new users after 1, 7 and 30 days	Day 1, 7, 30
User retention	The percentage of users who return on each day after their first visit	First 42 days
User engagement	Average engagement time of users who return	First 42 days
Lifetime value	Average revenue from new users	First 120 days

New and returning users

The first card splits your traffic into two groups. Google's user metrics article defines them:

• New users are unique users "who logged the first_open or first_visit event in the specified date range."
• Returning users are unique users "who initiated at least one previous session, regardless of whether or not the previous session was an engaged session" in the date range.

Two things follow from those definitions. A visitor who cleared cookies or switched browsers comes back as a new user, because GA4 sees a new device. And Google warns that "not all new users are active, so new users may exceed active users." Our guide to users and sessions in GA4 explains how these counts relate to total and active users, so we don't repeat it here.

A rising returning-user line is the simplest sign that people come back. Read it next to the new-user line: if new users grow fast, returning users usually grow later, so compare like with like over a few weeks.

Reading the cohort charts

In both "by cohort" cards, "the cohort is the day the user was acquired." Each chart has up to three lines, Day 1, Day 7 and Day 30, and each point on a line belongs to the users who first arrived on that date.

Google's example for the retention chart: "4 users visit your site on October 1. By October 2, 3 users return so the "Day 1" line on October 1 will be 75%. By October 8, 2 users return so the "Day 7" line on October 1 will be 50%. By October 31, 1 user returns so the "Day 30" line on October 1 will be 25%."

So a point shows how a group of new users behaved later, plotted on the day they arrived. That has two practical effects:

• The last days of your date range have no Day 7 or Day 30 value yet, because those users haven't had the time. A line that drops to zero at the right edge is usually this, not a real collapse.
• A dip on one date points at that day's new users. A campaign that brought in less interested visitors, or a launch with a bug, shows up there.

The engagement chart works the same way but with time instead of a percentage. It uses average engagement time, which Google defines as "the average time that your website was in focus in a user's browser or an app was in the foreground of a user's device." Our engagement guide covers how that time is measured.

The two 42-day cards are different again. User retention "starts with 100% user retention when all users visit for the first time" and "decreases from the first day as users drop off." User engagement shows the average engagement time of users who return, and Google notes: "Only returning users are included in the average engagement time." A flat engagement line with a falling retention line means fewer people come back, but those who do stay as long as before.

Lifetime value

The last card is "Average 120d value," which Google describes as "the average revenue generated by new users over their first 120 days." Google's example is a user who buys a shirt and then pants the following week: lifetime value includes both purchases.

The dimensions and metrics reference says the LTV metric "is the sum of the purchase, in_app_purchase, app_store_subscription_renew, and app_store_subscription_convert events minus the refund events." That means the card is empty for a site without purchase tracking. A lead-generation site will see little here; an online shop needs working purchase and refund events before the number means anything.

Retention report vs cohort exploration

The Retention overview answers one fixed question: of the people who arrived on a day, how many came back. A cohort exploration lets you choose the question. Google says you can build cohorts "based on your user's acquisition date, an event they triggered, their transactions, or their conversions," and set a separate return criterion.

	Retention overview	Cohort exploration
Type	Standard report, summary cards	Exploration in Explore
Who is in a cohort	Users acquired on a day	Your inclusion criterion: first touch, any event, transaction, conversion or a chosen event
What counts as a return	Fixed by the report	Your return criterion
Granularity	Daily cohorts; Day 1, 7, 30 and 42-day curves	Daily, weekly or monthly
Breakdowns	None on the cards	A breakdown dimension, up to 60 cohorts
Data retention setting	Not affected	Limits the date range

Use the overview to notice a change, and the exploration to explain it, for example to see whether users from paid search return less than users from organic search. Google also notes that in explorations "cohorts are based on the user's device data only," so the two can disagree with a User-ID setup. Our cohort guide covers inclusion and return criteria and why its numbers differ.

The data retention setting is something else

The word retention means two different things in GA4. The Retention report measures whether users come back. The data retention setting in Admin decides how long user-level and event-level data is kept before it is "automatically deleted from Analytics servers."

Changing the setting doesn't change this report. Google says the data retention setting "does not affect standard aggregated reports" and "only affects explorations and funnel reports." The Retention overview is a standard report, so it can show long date ranges even on the default setting, while a cohort exploration can only look back as far as the setting allows. Our data retention guide explains what it affects and how to change it.

Changing the cards

Like every overview report, the Retention overview has a date range picker, comparisons, sharing and export. Google's overview reports article says Editors and Administrators can choose which cards appear, and "the summary cards chosen will appear for everyone with access to the report by default." Comparisons are the quickest way to split the cards, for example mobile against desktop; our comparisons guide shows how to set them up.

How we use the report

This is our own practice, not Google guidance:

• We look at the returning-user line and the Day 7 retention line monthly, and mark launches and campaigns with annotations so a dip has a likely cause next to it.
• We ignore the last week of any Day 7 line and the last month of any Day 30 line, because those users haven't had time to return.
• When a line moves, we rebuild the question in a cohort exploration with a breakdown such as first user channel before drawing conclusions.
• We only read the lifetime value card once purchase tracking has been checked, and we compare it with what the shop's own system reports.

Frequently asked questions

Where is the retention report in GA4?

Go to Reports, then Life cycle > Retention. If your property uses business-objective collections, the report isn't there by default, but an Editor or Administrator can add it to the left navigation.

What does Day 7 retention mean in GA4?

It is the share of users first acquired on a date who returned 7 days later. In Google's example, 4 users arrive on October 1 and 2 return by October 8, so the Day 7 value for October 1 is 50%.

Why does the retention line drop to zero at the end of the date range?

Users who arrived in the last days of the range haven't reached Day 7 or Day 30 yet, so there is nothing to show. It is not a sign that retention collapsed.

Is the Retention report affected by the data retention setting?

No. Google says the data retention setting does not affect standard aggregated reports and only affects explorations and funnel reports. It limits a cohort exploration, not the Retention overview.

Why is the lifetime value card empty?

Lifetime value is built from purchase, in-app purchase and subscription events minus refunds. If your site doesn't send those events, there is no revenue for the card to show.

For more on measurement, see our guides to GA4 basics, explorations, the Tech reports and the Events report, or see our services.