-
Monitoring User Behavior in GA4: Tracking Segment Overlaps 📊
I’ve been racking my brains trying to figure out how to keep tabs on the behavioral changes of my “new users only” segment over time, as I construct my segment overlap report. It’s baffling me. Any ideas? ❓
It seems these “new user only” folks get booted once they don’t meet the “new kid on the block” label anymore, meaning they’re not sticking with the conditions set in the segment. It’s a bit of a bummer. 🙁
I gave creating an audience a whirl, but it was a no-go. Didn’t help because they weren’t living up to the condition and I’m wanting to track back in time. Creating an audience only starts stockpiling the data.
-
2 Replies
-
In this scenario, one potential solution would be to track new users as part of a cohort. Cohort analysis allows you to track users based on shared characteristics over time. The “new users only” would form one cohort, and after they lose their “new” status, they can be tracked in different cohorts. By setting the time frame when they are considered “new users” (e.g., 30 days), and then following their behavior after this period passes, it becomes much easier to analyze their behavior and evolution over time, even after they cease to be “new users”. It’s powerful because it’ll allow tracking the user’s life cycle and see how their behavior changes as they become more oriented with your product or service. While it may require more time and resources to set up the cohorts and analysis properly, the level of insight offered can help drive decision-making in a more targeted fashion.
-
Considering the restrictions that you’re running into, one solution could be to consider tagging users as “new” when they first join, and keeping that tag over time, even when they no longer satisfy that condition. It might demand some workaround in your system’s setup, but it could be worth a try. For retrospect analysis, try digging into the historical data where you have initial capture point of the user (their sign-up or first interaction date for instance).
If a unique identifier exists, it could be easier to track a cohort of new users and their progression. If such options are not present in your analytics platform, you might want to consider external tools or even custom setups where users are tagged with their start date. Large scale data platforms often have tools for cohort analysis which allows you to exactly do that.
Remember, you’re trying to change your user data categorization into a permanent attribute, rather than a fluid state. This allows you to track the behaviour of this group over time, despite the changes they undergo.
Log in to reply.