Analyze retention

Determine if users are loyal using these session-based and event-based metrics to see when they open the app and which events continue to engage them.

Localytics tracks retention by cohort (the group of unique users who did that event for the first time) in two ways:

Session-based analytics: high-level overview of app user retention based on app opens.
Event-based analytics: more granular look at specific events that continue to engage users over time.

Both retention charts are viewable by daily, weekly, and monthly cohorts.

Note: If a user skips a period but comes back during a later period they will not appear in the skipped period.

Use retention to gauge user loyalty, see which feature's events are best at promoting repeat-engagement, and to compare retention among groups of users who engage with the app at different times. It is also great for comparing the retention of users from different acquisition campaigns.

Navigate to Retention.
Select a date range.
The session-based retention displays. It lists the last 11 periods and how many users joined each cohort during that period. For each period, it lists the percentage of users retained during the following periods.
Change to event-based retention.
Event-based cohorts are defined by a starting event and measure retention for a follow-up event.
View the percentage of users who progressed from one event to a subsequent event.
1. Select your first event.
  The first event you select indicates the cohort.
2. Select your return event.
  This is the retention measure: the percentage of users from each cohort who did the second event in each subsequent day, week, or month. You can choose the same event twice or choose two different events.
3. (Optional) Add conditions to either event.
(Optional) Use filters to compare retention between different groups of users.

This is helpful when releasing a new version of the application or expanding into new countries.

Note: Unlike other areas of the dashboard, filters in the cohort screen are applied to the overall table not individually to each column. For example, if a you apply a country filter and a user recorded that country during the selected time frame, that user will appear in every column they visited, even if they had a different country during some of their visits. This gives you a truer representation of retention.