Step 1: Define Events

An event is any action a user performs on your app that is tracked. Retention is when a user performs an event and then returns to do another event. Select the First Event and Return Event from the list for your analysis.

Each event can be further filtered using + Attributes. Add event filters for specific user actions, such as Platform and App Version.

Use + Attributes again and combine multiple attributes using AND/OR for each event. For example, use the OR filter to combine data across two platforms. If you want to get data for a specific platform and look at the visits to a specific website, use the AND filter.

You can add multiple first events and return events using the OR function if you want to perform the retention analysis for multiple events.

Step 2: Filter Users

By default, you can perform behavior analysis on all users. However, you can also query for a group of users filtered using User property, User activity, User affinity, or Custom segments. Filtering users is similar to creating a segment. For more information, refer to Create Segment.

Step 3: Retention Options

Retention Type

After defining your events, you can choose how to calculate the data. The following options are available:

• Unbounded: Unbounded retention provides the percentage of users who were retained until the selected day or later. This metric includes users who performed the return event on or after that day. Here, the user is marked retained on all days until the user performs the return event for the last time. For a higher level of granularity, choose hour, week, or month instead of day.
  For example, the user installed the app (first event) in Week 0. The user opened the app (return event) on Day 0, Day 1, and Day 3 but did not open the app on Day 2. The user is counted as retained for Day 0, Day 1, Day 2, and Day 3. The user will be counted as retained for Day 2 as the user was active on the platform for the later period on Day 3.
• N-Day: N-Day retention provides the percentage of users retained on the selected day. It also includes the users who performed the return event on the selected day. For a higher level of granularity, choose hour, week, or month instead of day.
  For example, the user installed the app (first event) on Day 0. The user opened the app (return event) at least once on Day 0, Day 2, and Day 4 but did not open the app anytime on Day 1 and Day 3. The user will be counted as retained only for Day 0, Day 2, and Day 4. The user will not be counted as retained for Day 1 and Day 3 because the user was not active on these days.
  
  

Split By

With Split by, you can compare the data using a specific attribute. Add the desired attributes from the list with which you wish to compare. You can add a maximum of ten attributes.

Duration

By default, retention analysis is performed for seven days. You can select your desired duration in the list.

Step 4: Chart

The visualization type for retention analysis is a line chart. The chart is created as a weighted average of the percentage of retained users for the selected day or selected granularity. This calculation applies to all retention types.

Granularity

You can break down the chart into Daily, Weekly, or Monthly granularities for easy visualization. The retention users are displayed based on the selected granularity.

A few limitations of the retention granularity are:

• Hourly retention is available for a day (24 hours).
• Daily retention is available for 90-day durations.
• Weekly retention is available for 180-day durations.

Compare to Past

The Compare Mode enables you to compare your analyses with data from a specific time period. Rather than performing separate analyses for each insight, this mode allows you to utilize a single analysis to gain insights from multiple perspectives. Click Compare to Past at the top of your chart and select the duration you wish to compare to. You can also select a custom date range starting with or ending with to do a custom comparison equal to the baseline duration. For more information, refer here.

Step 5: Tables

Along with the chart, the data is available in a tabular format.

The Users column shows the number of users who performed the first event on a specific day. The Day 0, Day 1,...Day n columns provide the percentage of users according to Unbounded Retention, N-Day Retention, and First Occurrence.

The day granularity can be replaced with any granularity, such as hours, weeks, or months. The percentage numbers are color-graded for a better holistic view.

You can transpose the table. This allows you to view and download the table in your preferred format by shifting its vertical and horizontal orientation.

Download

After viewing the information, you can download the chart or table. The charts are downloaded in PNG format, and the table is exported in CSV format.

Save to My Studio or Custom Dashboard

Your analysis is now ready. To access it later, you can save the analysis.

To save:

1. Click the icon in the upper-right corner and click Save to My Studio.
2. Additionally, you can save the analysis to a custom dashboard. Choose the required option from the list.
   
3. Select the dashboard where you want to save the analysis. You can choose an existing dashboard or create a new one.
   
4. Enter a name for your analysis to help you identify.
5. Enter a brief description of the analysis.
6. Click Save.

You can access the analysis by navigating to Analyze > My Studio. To access the custom dashboard, navigate to Dashboards.

Next Steps

• MoEngage Dashboards
• Session and Source