A holdout test can separate a campaign’s observed click credit from its incremental contribution under the conditions tested.
Key moments
- 0:00 A coffee company ran mostly product and promotion ads because they got the clicks and the sales.
- 1:09 It tested product/promotion ads and top-of-funnel problem/solution ads, each against its own holdout group.
- 1:33 The result: the product ads mostly got sales that would have happened anyway; the upper-funnel ads drove more new business.
- 2:17 Demand capture closes existing momentum; demand creation builds it so customers can be captured later.
- 2:35 Clean data and good comparisons make the test reliable.
What this video covers
The point is not that promotional creative has no role. It is to assign each campaign a role using evidence that can test the dashboard story—and to reallocate budget with guardrails.
Read the related guide: Creative Performance Analytics
Upper-funnel media that gets few or no clicks? Here’s how to measure CTV, TV and podcast ads