I am trying to view a list of users based off of their first time ever having an app session.
What is the difference between the two scenarios, and why do they produce 2 vastly different numbers?
REPORT: App Session $ae_session: 1st Time Filter
COHORT: First App Open $ae_first_open
When running the report of App Session & 1st Time Filter, I see that there are 60 users within a 7 day timeframe of November 22-28.
In contrast, when running the cohort analysis of: First App Open, I see there are 13 users within the same 7 day timeframe of November 22-28.
Best answer by ReubenMoss
Hi April, how are you. I’m a Support Engineer here. While I can’t speak to the exact discrepancy in your case, and I would expect these numbers to be close, let me just clarify a few things and suggest a couple things to check:
First off, App Session triggers automatically when a user has had the app open for at least 10 seconds. It’s conceivable, then, that a user could do First App Open without triggering App Session (if the app is closed quickly or crashes). So I can imagine that maybe you have folks whose First App Open was before that Nov 22 date but whose first App Session was in that range.
Another thing to note, that I’m not sure would be relevant in this case, would be that a user can do First App Open on multiple occasions, particularly if they delete and reinstall the app, or they log in from a new device. I say I’m not sure that matters here since if anything, this would mean you’d perhaps see more First App Opens than first sessions.
The last thing to look at, which can be a ‘gotcha’, is to check your Explore query where you see just 13 users to make sure you’re looking at All Users and not just User Profiles. There could be users who have not yet authenticated, so they’ve not been given user profiles, but are still folks who would have opened the app and counted as sessions back in an Insights query. I wouldn’t be surprised if that alone gets you closer to a match.
Hope that this helps clarify matters for now - let me know if that helps! And if not, feel free to email the support team with some links to those reports so we can take a closer look here.