Which test for a save-count metric, and why is a plain t-test risky?
Test selectionMediumProblem. You're comparing saves-per-user between two arms of a Homefeed experiment. Saves-per-user is heavy-tailed — most users save a little, a few power-users save enormously. A teammate runs a plain t-test. Is that right, and what could go wrong?
Before you reveal: say your answer out loud, as if you were in the real interview — get your reasoning across clearly first. There is no single correct answer: reading what the interviewer is really after and defending your own thinking is what makes an answer strong.