How much did turning ads on (or off) actually create?
Unlike attribution, this excludes conversions that would have happened anyway without ads and shows pure incrementality. Pick whichever of the 3 methods fits your situation.
🧪 Control group (suppression)
Randomly block ads for a subset → compare exposed vs unexposed concurrently. Cleanest.
🟢 New launch (on)
Compare before/after turning on something new → the lift it created.
🔴 Shutdown (off)
Compare before/after turning off something running → the drop from stopping it.
Compares a holdout group (ads randomly blocked) vs an exposed group over the same period. Random assignment gives the highest causal confidence.
Compare an exposed group vs. a randomized holdout group (ads suppressed) over the same period to see the true incremental lift from ads. The most reliable of the 3 methods.Needs: Exposed/holdout · Conversions · Group size
Drag & drop CSV file
or click to select a file
See results instantly with sample data, no CSV needed