On October 14, 2020, Google announced the launch of Google Analytics 4, a new version of its well-known analytics platform that replaced Universal Analytics.
One of the main changes introduced by GA4 was the stronger focus on the user. While Universal Analytics was built more heavily around sessions, GA4 gives much more importance to user-based analysis.
Because GA4 can combine data from websites and apps within the same property, it needs ways to recognize users across platforms and devices. To do that, it can rely on different identity methods.
Contents
GA4 can identify users in different ways
GA4 can use different identity spaces to recognize users. The most relevant ones in this context are:
User-ID: the identifier provided by your business
If your business allows users to sign in, which is very common in e-commerce and other account-based websites, you can send your own identifier to Google Analytics. When that happens, GA4 can associate activity from the same signed-in user across sessions and across different devices or platforms. Google describes User-ID as a business-generated identifier that helps connect behavior across sessions, devices, and platforms.
This is not collected automatically. It has to be implemented as part of your tracking setup.
The main limitation is that it only works when the user is authenticated and you are actually sending that identifier. If the user is not logged in, GA4 will not be able to rely on User-ID for recognition.
Google signals: Google’s signed-in data
Another way GA4 can identify users is through Google signals. This is based on Google users who are signed in and have the relevant personalization settings enabled. When Google signals is available, GA4 can use that information to improve cross-device reporting and user recognition. Google includes Google signals as part of its reporting identity options.
This can be especially useful because it helps connect activity from the same user across different devices.
However, Google signals is not always available for every user, and it is also affected by privacy settings, consent, and regional restrictions. It is not a guaranteed identifier for every visit.
Device ID: the automatically assigned identifier
When GA4 cannot rely on User-ID, it can use a device-based identifier instead. In GA4 reporting, this is typically the client ID for websites and the app instance ID for apps. Google explicitly describes device ID in GA4 this way.
This method makes it possible to recognize repeated activity from the same browser or app instance, but it does not reliably identify the same person across multiple devices.
In other words, device-based identification is useful, but it is more limited than User-ID or Google signals when the goal is to understand cross-device behavior.
How does GA4 decide which user identifier to use?
GA4 does not treat all identity methods equally. In reporting, it prioritizes stronger identity methods when they are available. User-ID is the clearest option when you implement it correctly, because it comes directly from your own system and is tied to authenticated users. Google recommends using real, unique User-IDs and explains that they improve user accuracy and cross-platform measurement.
If User-ID is not available, GA4 can rely on other identity spaces such as Google signals and device ID, depending on your reporting identity settings.
That means GA4 is always trying to use the best available method to recognize the user, but the final result depends on your implementation, your property settings, and the identifiers actually available for that visit.
GA4’s user-centric model is one of the biggest differences compared with Universal Analytics. Instead of looking only at isolated sessions, it gives you a better chance to understand how people interact with your business over time and across devices.
That said, the quality of user recognition depends heavily on how your setup is implemented. If you can send a proper User-ID, you will usually get the most accurate view of authenticated users. If not, GA4 may still rely on Google signals or device-based identifiers, but those options are more limited.
Understanding these identity methods is important if you want to interpret GA4 data correctly and make better decisions based on users rather than just sessions.
