Subscription Retention Chart
This page describes Charts v3.
This chart is also available in our legacy Charts v2. See Charts v2 and v3 differences.
Subscription Retention shows you how paying subscriptions renew and retain over time by cohorts, which are segmented by subscription start date by default, but can be segmented by other fields like Country or Product as well.
Available settings
- Filters: Yes
- Segments: Yes
- Absolute/Relative Selector: Yes
- Retention Period Selector: Yes
Segmentation & cohorting
When segmenting by subscription start date, cohorts are segmented by the start of a paid subscription. Subsequent periods along the horizontal table indicate how many subscriptions continued renewing through those periods.
When segmenting by other dimensions, such as Country or Store, cohorts are segmented by the values within that dimension (e.g. App Store or Play Store for the Store segment), and include all paid subscriptions started within the specified date range. Subsequent periods along the horizontal table indicate the portion of subscriptions that successfully renewed out of all those that had the opportunity to renew in a given period.
How to use Subscription Retention in your business
Measuring Subscription Retention is crucial for understanding:
- How your product is delivering value over time
- How your Realized LTV is likely to change over time, and
- What subscriber segments exhibit the strongest and weakest retention
By studying this data, you may learn which customer segments are worth focusing your product and marketing efforts on more fully, or where you have the opportunity to improve your pricing to increase retention over time.
Calculation
For each period, we measure:
Subscriptions: The count of new paid subscriptions started within that period.
- This differs from other charts that are cohorted by a customer’s first seen date. This chart is cohorted by first paid purchase date.
- Trial conversions are cohorted based on their trial conversion date, not the trial start date. Trials that never converted to paid are not considered in this chart at all (since they were never a paid subscriptions).
- Product changes from product X to product Y are considered churn for product X and a new paid subscription to product Y, cohorted by the effective date of the product change.
Retention through each available period
- Retention: The number of subscriptions in the cohort who made an additional payment in a given period that have had the opportunity to do so.
- Available period: Use the date range selector to determine the range of subscribers to include in the chart. For each cohort included in that date range, all available retention periods will be provided.
When segmenting by subscription start date, each subscriber in that cohort will have had an equal opportunity to renew within a given period, but when segmenting by other dimensions, each subscriber may have different subscription start dates, and therefore different renewal opportunities.
As a result, the retention rate within each period cannot be directly compared to the total size of the cohort, because we will only measure the retention rate of the portion of that cohort that had the opportunity to renew in that period (e.g. for Month 3 retention, we will only include subscribers whose subscription start date is at least 3 months ago).
Formula
When segmenting by Subscription Start Date
[Retained Subscriptions at x period] / [Subscriptions] = Retention
When segmenting by other dimensions
[Retained Subscriptions at x period] / [Subscriptions that have had the opportunity to renew] = Retention
Retention is always calculated relative to the size of the initial cohort that has had the opportunity to renew. Retention is not calculated relative to the prior period.
FAQs
| Question | Answer |
|---|---|
| How are resubscribes handled in Subscription Retention? | If a user churns, but later resubscribes, this will be considered as a new subscription, cohorted by the resubscription date. |
| How are grace periods handled in Subscription Retention? | Grace periods extend the time that a subscription may be active within to give stores the opportunity to still successfully renew a subscription that has not been cancelled. The Subscription Retention chart excludes these periods from our retention calculation, since it is intended to reflect the number of subscriptions that successfully renew in subsequent periods. This will cause reported retention to be shorter for a subscription that enters a grace period and fails to renew from it if comparing that same subscription in the Active Subscriptions chart, since we consider the subscription active until its grace period ends and its entitlement is revoked. |
| How are refunds handled in Subscription Retention? | Refunded subscriptions are treated like subscriptions that churned during the period that the refund occurred in on the App Store & Play Store, since their access to the product is immediately revoked. On Stripe, however, you can decide whether or not to revoke access, which will determine whether the underlying subscription is treated as churned or not. |
| How are product changes handled in Subscription Retention? | This chart measures retention on a per-product basis. If a subscriber changes products, they would be counted as “churned” on the old product and start a new “retention journey” on the new one. Their new cohort would be based on the date the transaction for the new product was completed. |
| Can I measure Subscription Retention across all product durations? | Subscription Retention is filtered by product duration based on the selected Retention Period, and does not support measuring retention across all product durations. This is because the x axis of the chart is based on the product duration, which would not work when mixing product durations. To compare the retention of products with different durations, use the Cohort Explorer to cohort New paying customers and measure Retained subscriptions. |
| How are incomplete periods displayed in Subscription Retention? | The final, incomplete period of a given cohort is indicated with a hashed background. It shows how many subscriptions are set to renew by the end of the period, taking into account the auto-renewal preference of the subscriptions. |
| How are introductory offers treated in this chart? | Subscription cohorting depends on the type of offer. Free trials are not included while the customer is still in the free trial period because the user has not paid yet. If the trial converts, the subscription is then cohorted into this paid conversion date. Subscriptions with paid introductory offers are cohorted into the initial purchase date of the offer because the customer has made a payment, even if it is discounted. |