Hi, I need to split the result of an AB test by new or already existing users. My metric is Retention D7, so I use
conversion delais
and
conversion windows
.
I am able to know if a user is new at the moment he enters the AB test, but I lose this info as days goes on.
How can I create a dimension based on an information I have only when the user enters the experiment ?
What’s the behavior of
dimension columns
when the user have multiple exposures events and the dimension value changes ?
Any help would be greatly appreciated 🙏
we can always know the dimension that is associated with the first experiment exposure for a unit
So if I flag my user as a new user using a dimension at its first exposure, it will be considered new when I split the test results for the whole test duration.
is that right ?
f
fresh-football-47124
10/09/2023, 5:41 PM
Correct
fresh-football-47124
10/09/2023, 5:51 PM
There might be some nuances with the SQL, depending on the date of expsure, and how you define that dimension.