A Gifting experiment looks great — but is it just network spillover?

InterferenceHard

Problem. A viewer-level A/B of a gifting feature shows treated viewers spend and engage much more. A colleague wants to ship. Why might the estimate be biased, and how would you test it?

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.

← Who should get the discount? Estimating heterogeneous effectsA promo lifts sub conversion 20% — do you ship it? →