This message was deleted.

I think you will need to exclude them from the experiment. Because you did not randomize which beta tester got the new feature, adding them all to the

variation

and not splitting them across the

variation

and the

baseline

will result in biased results. If I'm understanding you correctly, you should keep those beta testers being forced to get the variation and experiment on the rest of your users.

Ok, thanks for your fast response!