▶️ Video
✓ Link checked
Free
Intermediate
Why we picked it Haus runs incrementality tests for a living, and here their founder talks through how you actually separate ads that cause sales from ads that just show up next to sales that were coming anyway. It grounds the holdout idea (withhold ads from a slice of your market, then compare) in how brands really run these experiments. Watch it once to build the mental model before you design your own test.
Measuring Top of Funnel and Incrementality Testing with Zach Epstein from Haus
On Haus on YouTube by Haus (Zach Epstein) Podcast style conversation, roughly 30 to 45 minutes
- A holdout is simple in principle: keep ads away from one comparable group or geography, run them for another, and the gap is your true incremental lift.
- Platform-reported numbers flatter top of funnel spend the most, since that is where overlap with organic demand is largest.
- You do not need perfect infrastructure to start. A clean geo split and a few weeks of patience gets you a real answer.