Hi! Quick question about targeting and attribute changes. If a user enters an experiment while not logged in (so

b2b

is unknown) and gets bucketed into a variant, then later logs in and we update attributes with

b2b: true

, and the experiment targeting is `b2b = false`: What happens to that user? • Are they removed from the experiment and excluded from analysis? • Or does the original bucket assignment still count in the results? We're trying to avoid B2B users affecting experiment metrics when their status is only known after login.

By default, the SDK isn't going to know the previous state of the user. So, if

b2b

changes, it won't change anything for previous assignments. To prevent it affecting metrics, you can exclude any user where b2b = true from analysis, from the experiment settings page.

Tysm, @strong-mouse-55694!