Keshav · Distribution Intelligence

Incrementality

Did the ad CAUSE the sale, or would it have happened anyway? Compare an exposed cell to a held-out control and measure true lift. Incremental CAC — spend ÷ conversions that only happened because of the ad — is the number that decides whether to scale.

The holdout test

Exposed = cell size (population / audience / summed geo weight). Conversions = what each cell produced.

Test cell (exposed to ads)

Exposed

Conversions

Control cell (held out)

Exposed

Conversions

Ad spend on the test cell ₹

True causal lift

Enter both cells and measure to see whether the win is real.