Isolate retargeting's true lift with a ghost-ad holdout

Experiment designHard

Problem. A retargeting campaign shows huge attributed conversions, but you suspect over-attribution. Design an experiment that isolates its true incremental lift.

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.

← Measure a channel's incrementality without user trackingForecast installs and revenue with seasonality →