Measurement
Incrementality testing
A test that measures what ads actually caused, by comparing a group that saw them with a similar group that did not.
What it is
The group held back from the ads shows what would have happened anyway. The difference between the two groups is the lift the ads created.
Why it matters
For lead generation and e-commerceIt answers the question attribution cannot: would these sales have happened without the ads?
How it is measured
Formula
Lift = conversions in the exposed group - conversions in the holdout group (same size)Platforms offer conversion lift studies; advertisers also run their own holdouts and geo tests. Lift is the difference in conversions between the two groups.
A common mistake
It takes planning and enough volume to read, so teams that try it on small budgets often get answers too noisy to use.
Read the evidenceThe winner is smaller than it looks All terms