Inspect personalization data with BigQuery

Remote Config personalization logs a personalization_assignment Analytics event when a personalization is assigned to a user, allowing you to use BigQuery to inspect and analyze personalization events and associated events.

The following sections describe how to enable BigQuery export for Analytics events, how personalization events are stored, and provide some basic queries to get you started.

Enable BigQuery export for Google Analytics for Firebase

If you're on the Spark plan, you can use the BigQuery sandbox to access BigQuery at no cost, subject to Sandbox limits. See Pricing and the BigQuery sandbox for more information.

First, make sure that you’re exporting your Analytics data to BigQuery:

  1. Open the Integrations tab, which you can access using > Project settings in the Firebase console.
  2. If you're already using BigQuery with other Firebase services, click Manage. Otherwise, click Link.
  3. Review About Linking Firebase to BigQuery, then click Next.
  4. In the Configure integration section, enable the Google Analytics toggle.
  5. Select a region and choose export settings.

  6. Click Link to BigQuery.

Depending on how you chose to export data, it may take up to a day for the tables to become available. For more information about exporting project data to BigQuery, see Export project data to BigQuery.

Next, let’s access and inspect our personalization events in BigQuery.

Access Remote Config personalization data using BigQuery

To query analytics data for an experiment:

  1. Open BigQuery in the Google Cloud console. You can also open it directly from Analytics Events using the view your raw events in BigQuery link at the bottom of the page.
  2. Select your Firebase project and expand it, then expand the analytics_ANALYTICS_PROPERTY_ID entry and click events_.

    Access personalization events in the Cloud console

  3. From the Query drop-down, select In a new tab.

    An auto-generated example query appears.

  4. To view personalization events and associated data, update the query to select personalization_assignment events. The following example query will return a complete personalization assignment event for a specific date shard, limiting results to 10:

    # Select all personalization_assignment events
    SELECT *
    FROM `PROJECT_NAME.analytics_ANALYTICS_PROPERTY_ID.events_DATE_SHARD`
    WHERE event_name = 'personalization_assignment'
    LIMIT 10
    

    Tip: To search all event tables instead of the sharded table, you can replace the events table date with an asterisk (for example, PROJECT_NAME.analytics_ANALYTICS_PROPERTY_ID.events_*). This is not recommended in non-test scenarios or for large data sets.

  5. In the query composer, select Run query. Results appear in the lower pane.

In the next section, we'll discuss what’s included in a personalization assignment event in more detail.

What personalization data is exported to BigQuery?

Personalization data is included in Google Analytics tables in BigQuery and stored in personalization_assignment events.

The basic fields provided in a personalization event are the same as any Analytics event as described in [GA4] BigQuery Export schema. You’ll be mostly concerned with user_pseudo_id i(which can be used to differentiate distinct users), event timestamps, and other user properties.

Personalization-specific details are stored in the event_params field and are described in the following table:

Parameter Data type Description
personalization_id STRING Provides the assigned personalization's universally unique identifier (UUID).
group STRING Indicates whether the user was assigned to the personalization group (P13N) or the baseline (BASELINE) group.
arm_index INTEGER Represents the alternative value assigned to the user, an integer between 0 and 4.
arm_key STRING Contains the parameter name used by the personalization.
arm_value STRING Contains the alternative value string assigned by personalization.
engaged_session_event INTEGER Includes the number of sessions the user is engaged in. See About sessions for more information.
firebase_event_origin STRING Indicates the origin of the event. This will always be fp for personalization_assignment events.
firebase_screen_class STRING Provides the class name of the screen on which the user was active when the personalization assignment occurred. See Automatically collected events for more information.
firebase_screen_id INTEGER Displays the ID of the screen the user was on when the personalization assignment occurred. See Automatically collected events for more information.
first_open_time STRING Provides the timestamp, in UTC milliseconds, of the first time the user opened the app. See Automatically collected events for more information.
ga_session_id INTEGER Provides the Google Analytics session ID. See About sessions for more information. You can use this to correlate the personalization_assignment event with other Analytics events.
ga_session_number INTEGER Provides the Google Analytics session number. See About sessions for more information.

Example queries

You can use a SQL statement like the following to extract the personalization-specific parameters from personalization_assignment events:

    # Expand nested personalization parameters
    SELECT
     timestamp_micros(event_timestamp) AS event_time,
     user_pseudo_id,
     (
       SELECT event_params.value.string_value
       FROM UNNEST(event_params) event_params
       WHERE event_params.key = 'group'
     ) AS personalization_group,
     (
       SELECT event_params.value.string_value
       FROM UNNEST(event_params) event_params
       WHERE event_params.key = 'personalization_id'
     ) AS personalization_id,
     (
       SELECT event_params.value.string_value,
       FROM UNNEST(event_params) event_params
       WHERE event_params.key = 'arm_key'
     ) AS arm_key,
     (
       SELECT event_params.value.string_value
       FROM UNNEST(event_params) event_params
       WHERE event_params.key = 'arm_value'
     ) AS arm_value,
     (
       SELECT event_params.value.int_value
       FROM UNNEST(event_params) event_params
       WHERE event_params.key = 'ga_session_id'
     ) AS ga_session_id,
    FROM `PROJECT_NAME.analytics_ANALYTICS_ACCOUNT_ID.events_DATE_SHARD`
    WHERE event_name = 'personalization_assignment'
    LIMIT 10