Cohort Explorer
This page describes Charts v3.
This chart is also available in our legacy Charts v2. See Charts v2 and v3 differences.
The Cohort Explorer allows you to measure the performance of various cohort definitions over time and across multiple performance measurements.
Cohorts
The Cohort Explorer supports three different cohort definitions:
- New Customers: The count of new customers first seen by RevenueCat in a given period, cohorted by their First Seen Date.
- Initial Conversions: The count of customers first converting to any product (including paid or unpaid, and subscription or one-time purchase) in a given period, cohorted by their Initial Conversion Date.
- New Paying Customers: The count of customers making their first payment in a given period, cohorted by their First Purchase Date (not including free trials). This includes customers who were later refunded.
Please keep in mind that the count of New Paying Customers and Subscriptions started in a given period will likely differ, since one customer may have multiple subscriptions over their lifetime, or may convert to paid through a one-time purchase instead of a subscription.
Comparing cohort definitions
Each cohort definition is a unique way of grouping customers together to understand how each unique cohort performs over time.
Because these measures each have unique cohort definition, each period references different groups of customers. For example, the Apr '24 cohort of New Paying Customers may include some customers who happened to also be first seen in Apr '24, and are therefore in the Apr '24 cohort of New Customers, but it may also include some customers who were first seen in prior months.
In addition, the Apr '24 cohort of New Customers may include customers whose first payment won't occur until after Apr '24.
Therefore, these different cohort definitions should not be thought of as a conversion funnel for the period they reference. Rather, they are independent ways of grouping customers together to understand how each unique cohort performs over time.
Measures
The Cohort Explorer supports seven different measures that can be used to understand the performance of your chosen cohort over time:
- Revenue: The amount of revenue generated in a given period by that cohort.
- Revenue (net of taxes): Revenue generated in a given period by that cohort, minus our estimate of revenue deducted from the stores for taxes (e.g. VAT, DST, etc).
- Proceeds: Revenue generated in a given period by that cohort, minus our estimate of revenue deducted from the stores for taxes and commission.
- Realized LTV: The cumulative amount of Revenue generated by that cohort since their inception until a given period.
- Realized LTV / Customer: The cumulative amount of Revenue generated by that cohort since their inception until a given period, divided by the count of customers in that cohort.
- Retained Subscriptions: The count of subscriptions from that cohort which remain active as of a given period. (This count can go up over time if some members of the cohort start their subscription late, or if more customers reactivate in a given period than the number who churned)
- Subscriptions Set to Renew: The count of subscriptions from that cohort which remain active and set to renew (not cancelled) as of a given period. This measure can give you an indication of when in their subscription cycle customers tend to cancel their subscription.
Proceeds reflect RevenueCat's estimate of what you will earn from the stores for the revenue you generated, but keep in mind that the App Store's payment schedule is based on Apple's Fiscal Calendar, which does not align with calendar months. Learn more here.
In addition, to learn more about how RevenueCat estimates taxes and commissions deducted from the stores, click here.
The count of "customers" in that cohort refers to whichever cohort definition is being used. So when looking at a cohort of New Paying Customers, Realized LTV / Customer will divide Realized LTV by the count of New Paying Customers.
Available settings
- Filters: Yes
- Segments: No
How to use the Cohort Explorer in your business
The Cohort Explorer can be used to answer many different questions, like:
- At an average CAC of $x, what is my typical time to payback? (Realized LTV / Customer by New Customer Cohorts)
- What is my revenue retention of the customers who first paid in the last year? (Revenue by Paying Customer Cohorts)
- How can I expect Realized LTV to grow over time for my most recent paid customer cohorts based on the performance of my prior cohorts? (Realized LTV / Customer by Paying Customer Cohorts)
- What portion of my Initial Conversions become paying customers that remain paid after a given period of time? (Retained Subscriptions by Initial Conversion Cohorts)
Calculation
Periods (ie. the x axis of the table) are calculated based on each individual customer, not based on the cohorts:. For example, consider the case where you are cohorting new customers by monthly cohorts and are showing data by monthly periods. A customer who was first seen on 31 January would be in the January cohort. For this customer, their month 0 (ie. the first column of the cohort explorer table) would be 31 January to 27 February (the customer's first month), not 1 January to 31 January (the cohort month).
Example calculation for some of the measures:
Revenue of New Customer Cohorts
For each period, we:
- Count the New Customers that were first seen in that period
- Provide the sum of revenue generated by that cohort in each subsequent period
Realized LTV of Paying Customer Cohorts
For each period, we:
- Count the New Paying Customers that made their first payment in that period
- Provide the cumulative sum of revenue generated by that cohort as of each subsequent period
Realized LTV / Customer of Paying Customer Cohorts
For each period, we:
- Count the Paying Customers that made their first payment in that period
- Provide the cumulative sum of revenue generated by that cohort as of each subsequent period, divided by the count of Paying Customers in that cohort
Retained Subscriptions of Initial Conversion Cohorts
For each period, we:
- Count the Initial Conversions that made their first purchase or started their first subscription of any kind in that period
- Provide the count of subscriptions that were active from that cohort as of each subsequent period
FAQs
| Question | Answer |
|---|---|
| How are refunds handled in the Cohort Explorer? | When we see a refund processed by a store, the revenue is deducted and the subscription no longer considered retained in the period where the refund occurred (not the original purchase date). |
| How can the Cohort Explorer be compared with Subscription Retention? | When using the Active Subscriptions measure in the Cohort Explorer, it is fundamentally measuring the same thing as absolute Subscription Retention, but keep in mind that the Subscription Retention chart is always segmented by Product Duration so that the retention periods always reflect expected payment periods. In addition, the denominator of Cohort Explorer is customers, while the Subscription Retention chart measures subscriptions. Customers who are resubscribing, changing products, or who have multiple subscriptions in parallel will start a new retention curve in the Subscription Retention chart, while they will contribute to that customer's original cohort in the Cohort Explorer, not to the start date of the new subscription. |
| How does Realized LTV in the Cohort Explorer compare with the Realized LTV charts? | When you compare the New Customers cohort using equivalent filters, currency, and lifetime windows, cumulative Realized LTV through day 7 should match the 7-day Realized LTV chart. Both attribute refunds to the refund date within the customer's lifetime. For example, revenue from a purchase on day 4 that is refunded on day 10 remains included at day 7, then is removed at day 10 and from the 14-day value. |