Since I noticed a misalignment between the count of Users and the count of User IDs on Google Analytics (for mobile apps), I need to understand technically how Google counts Users in mobile apps. While I have extremely clear that on Web Users are counted based on cookies, on mobile app I still have doubts (I never developed a mobile app). I made some research and I'd like to have a confirmation from people who knows it.
Users in Firebase / Google Analytics represents the count of the field user_pseudo_id (StackOverflow answer). A comment to this question seems to confirm this assumption, as well as this sentence in the official documentation to set the User ID: "Setting a user ID is never required for Analytics to work correctly. If you're only interested in finding events belonging to the same user for the same app on a single device, you can use the user_pseudo_id. This value is generated automatically by Analytics and is stored within BigQuery for each event."user_pseudo_id should correspond to the App Instance ID (BigQuery Export Schema). If it's correct, why on the documentation the App Instance ID is referenced as an example? The field allows customizations?App Instance ID can change in some cases (device factory reset, app uninstall, data cleared, etc. --> Instance ID API), and that would explain the misalignment.Since, as mentioned before, I have zero experience in mobile app development, can someone confirm this flow? Is everything correct or am I missing something?
Thanks!
Well, this is good and complex question, I will try to simplify it. First of all, I will start a bit wider and then I will get back to your questions.
I start with this assumption: the metric Users in Firebase / Google Analytics represents the count of the field user_pseudo_id (StackOverflow answer). A comment to this question seems to confirm this assumption, as well as this sentence in the official documentation to set the User ID: "Setting a user ID is never required for Analytics to work correctly. If you're only interested in finding events belonging to the same user for the same app on a single device, you can use the user_pseudo_id. This value is generated automatically by Analytics and is stored within BigQuery for each event."
-> Basically metric 'Users' is the result of reporting identity. It depends what you collect, how you configure your GA4 UI, received consents (if some).
user_pseudo_id should correspond to the App Instance ID (BigQuery Export Schema). If it's correct, why on the documentation the App Instance ID is referenced as an example? The field allows customizations?
-> No, this field can't be customized. The field expects only specific format (string, length). The example value is DC05363C3AD34B47A33536755FCC9C3E
The App Instance ID can change in some cases (device factory reset, app uninstall, data cleared, etc. --> Instance ID API), and that would explain the misalignment.
-> Yes, this is true.
Just to give an idea about the identifiers that Google is collecting:
null into user_pseudo_id column.*If you use "Device-based" reporting Identity you are reporting only upon collected App Instance ID
If you love us? You can donate to us via Paypal or buy me a coffee so we can maintain and grow! Thank you!
Donate Us With