Mobile App Behavior

Track the detailed ways users interact with your app.

The Behavior reports help you discover how users interact your app. You can use this set of reports to track a wide range of information about your users, including:

  • the total number of screens seen per session
  • the order in which screens are viewed
  • the number of technical errors, including app crashes
  • how often users return to the app
  • how long sessions last
  • how long individual app elements take to load
  • specific content actions unique to your app, like a social recommendation or an ad click

Many of the Behavior reports require additional set up in the app tracking code that should be completed by the developer. Links to detailed setup instructions are included each report description below. Data does not appear in some of these reports until you complete the required additional set up.

Although these reports require a little extra work to set up, we strongly recommend you use them. The Behavior reports are powerful tools that can help you reach a deep understanding of how users use your specific app. The customizable tracking settings of these reports let you to track the individual elements of your app, allowing you to focus the reports on your specific business objectives. The way you set up these reports in your tracking code determines how data appears in your reports. The more organized you are about setting up them up, the easier it will be to read and interpret your reports.

Below are descriptions of each Behavior report in your Google Analytics account. Links to additional setup instructions are included where necessary:


Get a high-level summary of how users interact with app content. The Overview includes usage metrics like Screen Views, Crashes, and the Total number of Events that occur on a screen. Click view full report to jump to that specific report for a deeper analysis.


Evaluate the performance of each screen in your app. This report breaks out screen usage metrics, like the number of Screen Views, Unique Screen Views, and the % Exit from a Screen.

Behavior flow

Visualize the path users traveled from one Screen or Event to the next. This report can help you discover what app content keeps users engaged with your app. See user movement between Screens, Events, or a blended view of both Screens and Events. You must set up Events in your app tracking code before they appear in the Behavior Flow. See the section on Events, below, for details.

This Behavior Flow Report is a part of the family of Flow Visualization Reports. Read the About Flow Reports and Using Flow Reports articles in this Help Center for information on using Flow Reports.

Crashes and Exceptions

This report shows the name and a brief description of the top Exceptions, or technical errors, that happen in your app. Crashes are automatically parsed out from other exceptions. You can define more exception types in your app tracking code for more detail on other exceptions. Network failures and empty search results are two examples of common exceptions you might want to define.

Defining exceptions in the app tracking code requires some technical knowledge of your app and should be completed by the developer. Visit our Developer Guides for detailed instructions on setting up the exceptions report for Android and the exceptions report for iOS.

App Speed

Use App Speed reports to see how long different requests take to load in your app. You might, for example, want to track how quickly a game level loads or a search returns results.

The App Speed reports require additional set up in your app tracking code that should be completed by the developer. Visit our Developer Guides for detailed instructions on setting up User Timing for Android and User Timing for iOS. No data appears in these reports until you define in your app tracking code which specific actions and load times to track.

After you complete the additional set up, toggle between these three tabs to view the data:

Explorer: Timing Category displays the name you’ve associated with a specific load time you’re tracking. The Average User Timing metric shows you the average length of time it takes to load that action.

Distribution: User Timing Bucket displays the sum total the length of time (in seconds) all tracked actions take to load. Also provides a percentage sum total of sessions the sample size represents.

Map Overlay: The global distribution of Average User Timing. Darker shades indicate a longer load time.

All of these reports use the User Timing Sample to display the number of users that are used to calculate the average length of time for a specific action. By default, the App Speed sets this sample rate to 1% of total users, but you can adjust this setting in your app tracking code for each action you want to track.


This important set of reports points out who your most loyal users might be based on how often they interact with your app. Try to retain the users that engage with your app more often and for longer. In conventional marketing, this is the audience segment that usually completes more conversions and transactions.

Because of the complexity and importance of this data, we’ve divided Behavior into the following individual reports:

New vs. Returning: The percentage of users that have previously opened your app, and the percentage of users that are first time users

Loyalty: The number of sessions that are the nth occurrence in your app.

In the report table, Session Instance 1 is a session that occurs with no prior session recorded. It is the 1st session of the app associated with a specific user. The number of Sessions in this table row are the number of first-time sessions. The Sessions for Session Instance 2 are the number of sessions that occur with 1 prior session recorded or, the 2nd session for that many users; etc.

Here are some examples of data in this report, and what it means:

  • If you see 500 Sessions in the row for Session Instances 1, this means that 500 sessions were first time users.
  • 400 Sessions in Session Instances 2 means that 400 sessions were started by users that have used your app one time.
  • 100 Sessions in Session Instances 26-50 means 100 sessions were started by users that have used your app 25-49 previous times.

Recency: The amount of time that passes between the close of one session and the opening of another.

Session Duration: How long a session lasts, in seconds.

Sessions remain open as long as activity is tracked, and are closed after 30 seconds of inactivity has passed. Developers can also manually customize this limit. Consult our Developer Guides instructions on setting up sessions timing for Android and session timing for iOS.

Each of the Behavior reports provide scorecards for the high level session metrics, including the number of Sessions, Average Session Duration, average number of screens per session (Screens/Session), and the Goal Conversion Rate.


Events are unique content actions in your app that don’t require a new screen to load, like like a download, social recommendation, or an ad click. The Events reports are highly customizable and cannot provide any data until you complete additional set up.

Setting up Event Tracking requires technical knowledge and access to your app, and should be completed by the developer. Visit our Developer Guide for detailed setup instructions on setting up Event Tracking for Android and Event Tracking for iOS.

When you set up Events, you choose define and track up to five components, or categories, for each Event. The values you assign to these categories is the data that appears in the Events reports. The more organized you are about setting up Event Tracking, the easier it will be to read and interpret your reports. For more information and examples of these categories, read About Events.

After you set up Event Tracking, use the Events Overview report to see the summary of how well all events perform. Top Events displays data on the highest performing Events. Use the Screens report to see which screens host the most popular events.