Your customer retention rate is the share of customers you had at the start of a period who are still with you at the end of it.
Take your end-of-period customers, subtract the new ones you gained during the period, divide by the number you started with and multiply by 100. Start with 1,000, end with 1,050 after gaining 200, and your retention rate is 85%.
Enter those three numbers in the calculator above.
The sections below cover the formula, a worked example, how app teams measure day-N retention, and where to compare your result.
How to use the retention rate calculator
The calculator takes three inputs, all for the same period and the same group of customers:
- Customers at the start (S): active customers or paying subscribers on the first day.
- Customers at the end (E): active customers on the last day, including anyone new.
- New customers (N): people who became customers during the period.
Pick one definition of "customer" and keep it.
For a subscription app that usually means an active paid subscription.
Free trial users who haven't converted belong in trial conversion, not retention.
Pull all three numbers from one source, such as your subscription platform or the store consoles.
Mixing sources tends to double-count people who move between platforms.
The customer retention rate formula
The standard customer retention rate formula is:
Retention rate = ((E - N) / S) x 100
Subtracting N matters. New customers didn't exist at the start, so leaving them in would make a fast-growing business look like it keeps everyone.
Worked example: a language app starts October with 1,000 paying subscribers.
During the month 200 people subscribe, and it ends with 1,050. Retained subscribers are 1,050 - 200 = 850, so the retention rate is 850 / 1,000 x 100 = 85%.
The flip side is churn. For the same group and period, churn rate = 100% - retention rate, so 85% retention means 15% churn.
Our churn rate calculator works it from that side and converts monthly churn to annual.
Monthly retention doesn't scale by multiplication
To estimate a longer period from a steady monthly rate, raise it to a power instead of multiplying.
Twelve months at 85% gives 0.85^12, or about 14% of the starting group left after a year.
Real subscription cohorts rarely hold a steady rate, since most losses come at the first renewal, so measure longer periods directly when you can.
Retention also feeds lifetime value. Better retention means more renewals per customer, which the subscription LTV calculator turns into revenue.
Day-N retention for apps
App teams usually track a different number: day-N retention.
It asks what share of the users who installed on a given day came back on day N.
Day-N retention = (users from the install cohort active on day N / users in the install cohort) x 100
Worked example: 5,000 people install on November 1 and 650 of them open the app on November 8. Day 7 retention is 650 / 5,000 x 100 = 13%.
Two details change the result. Adjust defines a "day" as a 24-hour period measured from install, not a calendar date.
And "returned on day 7" is a stricter measure than "returned at any point in the first 7 days," which always reads higher. Use one method and label it.
For subscription apps, track both. Install retention shows whether people use the product, and paid retention shows whether they keep paying.
RevenueCat's State of Subscription Apps 2026, based on data from over 115,000 apps, found first-year retention of 28% for annual plans, 8% for monthly and 1.2% for weekly.
What tests in our database show about retention
Retention numbers move when teams change onboarding, trials and win-back offers.
A few tests from our app retention test library:
- Onboarding. Applica rebuilt ASL Bloom's onboarding around user goals and added a two-screen paywall. ASL Bloom saw +18% 30-day retention, alongside +15% trial conversion and +12% ARPU. Several changes shipped together, and the result is vendor-reported.
- Trial expectations. Blinkist's trial timeline paywall, reported by Marcus Burke, raised trial opt-in significantly and cut trial cancellations by 4%.
- Win-back. Unscripted sent lapsed users a personal win-back trial offer and got over +240% trial starts (a vendor-reported Superwall case study). Reactivated users count as new in the formula above, so track win-back separately.
- Forced conversion. BoldVoice tested a hard paywall with no close button. Trial starts went up, but refunds rose by more than the gain, a reminder that early conversion wins can cost you later.
- Price. TalkingParents raised its price alongside a new feature. About a quarter of customers churned, yet the business came out better financially overall (a before-after comparison, not a controlled test).
The last card is a later-horizon example. Vahe Baghdasaryan reports that a $1.99 paid 30-day trial lifted D90 ARPU from $2.23 to $2.89 (+30%) in health and fitness apps.
Judging that test at day 7 would have missed the result.
How your retention rate compares
Across all verticals, Adjust reports median retention of 26% on day 1, 13% on day 7 and 7% on day 30, with iOS slightly ahead of Android.
Category matters a lot: day 1 ranges from 29% for gaming and social apps down to 13% for food and drink.
For the full tables by category, platform and region, see our app retention rate benchmarks.
For the paying side, use the subscription churn rate benchmarks, and for ways to improve the number, start with our app retention guide and onboarding guide.