Date of slack thread: 5/15/24
Anonymous: Hello team, I had a question around overriding two statsig experiments in the same layer using a stableId
. When I override one experiment, the other experiment override is deleted since the stableId
matches. I know that experiments in the same layer are mutually exclusive, but is there a way to place myself in two experiments in the same layer at the same time for testing purposes.
Statbot: Overrides in a layer are designed such that a user can only be overridden into one experiment within that layer. This is to maintain the mutual exclusivity of experiments within the same layer. If you override an experiment for a user with a stableId
, and then attempt to override another experiment in the same layer for the same stableId
, the first override will be replaced by the second one, as per the documentation’s guidelines on how overrides work for experiments in a layer source. For testing purposes, if you need to see the behavior of multiple experiments in the same layer, you would need to test them separately or consider reorganizing your experiments so that they are not in the same layer if mutual exclusivity is not required for your test cases.
Statbot: Click below to mark as resolved.
Vineeth Madhusudanan (Statsig): Not within the same layer, sorry Eric. Mutual exclusion is a hard promise here :(
Anonymous: Ahh ok, thanks for the quick response!
Statbot: This thread has been marked as resolved. Press below to re-open.