Retention rate

Retention rate is the percentage of users from a group that started at the same time who are still coming back to a product after a set number of days. For apps, it is usually read at Day 1, Day 7 and Day 30 after install.

What is retention rate?

Retention always comes with a cohort, meaning a group of users who share a start date: the day they installed the app, first visited the website or placed their first order. A claim like “the app retains 40% of users” means nothing until you say which group and after how many days.

What counts as “coming back” also has to be defined up front. In GA4’s cohort exploration, you can set the return criterion to any event, a transaction, a conversion or a specific event (Google Analytics Help, accessed October 2026).

Retention rate formula

Day N retention = Users in the cohort who return on day N / Total users in the cohort × 100%

Google gives an example: 4 people visit a website on October 1. On October 2, 3 of them come back, so Day 1 retention is 75%. On October 8, 2 are left, so Day 7 retention is 50% (Google Analytics Help, Retention overview, accessed October 2026).

That method is called Standard. GA4 also offers Rolling (counting only people who returned in every consecutive period) and Cumulative (counting people who returned in any period). Run the same data through all three and you get three different numbers.

Where to measure retention: GA4 and App Store Connect (2026 update)

• GA4’s Retention overview report shows the percentage of users returning each day over the first 42 days. Cohorts in GA4 are based on device data and do not use User-ID.
• On March 25, 2026, Apple shipped its biggest update to Analytics in App Store Connect, adding cohorts by download date, download source and offer start date (Apple Developer News, March 2026).
• Apple defines retention as the percentage of devices that installed the app on the selected day and opened it again after N days, counting only users who agreed to share usage data. A cell can stay blank if it has not met the privacy threshold (Apple, App retention, accessed October 2026).
• Apple offers peer group benchmarks for Day 1, Day 7 and Day 28 retention, showing the 25th, 50th and 75th percentiles for apps in the same category and business model (Apple, Peer group benchmarks, accessed October 2026).

Each tool samples differently, so retention in App Store Connect, GA4 and your MMP will not line up. Pick one source for tracking trends.

Example: choosing ad sources by who stays (hypothetical)

A food delivery app in Ho Chi Minh City runs two sources in the same week:

• Source A: 12,000 installs, CPI of 15,000 VND, total spend of 180 million VND, Day 7 retention of 18%, leaving 2,160 users.
• Source B: 8,000 installs, CPI of 22,000 VND, total spend of 176 million VND, Day 7 retention of 30%, leaving 2,400 users.

On CPI, A looks cheaper. Work out the cost per user still active on day 7, though, and A comes to about 83,300 VND while B comes to about 73,300 VND. Optimizing for installs would send the money to the wrong place.

Retention rate vs. churn rate vs. DAU/MAU

The DAU/MAU ratio is explained in the DAU entry.

Frequently asked questions

What is a good Day 1 retention rate?

There is no single number that fits every industry. Games and ticket booking apps get used at very different rhythms. If your app is on the App Store, Apple’s benchmarks show which percentile you sit in compared with apps in your peer group.

If retention drops, should I fix the ads or the product?

Split cohorts by install source first. If every source drops together, the cause is usually the product or a new release. If only one source drops, look again at that source’s audience and creative.

If you want app campaign goals set on users who stay instead of installs, see how Adtek’s App User Acquisition team works or get in touch through the contact page.