Best practices
It is recommended that you do not register a user ID as a custom dimension.
Creating unnecessary high-cardinality custom dimensions, like User ID, negatively impact your reports and explorations, and cause data to be condensed under the (other) row. Instead, use the User-ID feature.
Analyze user-based data
The following describes features you can use to interact with and analyze user-based data:
Use case |
Feature |
---|---|
I want to explore segments of users and drill into individual user activities. |
User Exploration |
I want to monitor user properties as they are collected, troubleshoot issues as issues arise, and understand a user's behavior as the user explores my website or app. |
DebugView |
I want de-duplicated reporting based on user ID. I want to compare users who are signed in with those who aren't signed in. |
In Universal Analytics, User ID is only used in specific User ID reports. Thus, User ID is registered as a custom dimension to be used in non-User ID Views, and compare traffic. In GA4, with User-ID in the reporting identity, User ID de-duplication is present across all reports. |
I want to create remarketing audiences based on user IDs. |
Create remarketing audiences based on User-ID data |
I want to see user-level data by user ID. |
GA4 BigQuery Export has both Client ID (user_pseudo_id) and User ID (user_id) in the schema. |
I want to join my Analytics data with first party data (e.g. CRM data) so I can measure offline key events. | BigQuery Export |
I want to run additional analysis on my GA4 data (e.g. run ML models) and import the insights back into Google Analytics. | BigQuery Export and Offline Event Data Import |