Incrementality vs Last-Click: What Cost Teams Should Demand

Last-click attribution mis-prices channels. What finance and growth should agree on instead.

Markdown source

Last-click attribution is a convenient lie. Incrementality tests are an expensive truth. Cost teams should fund a light incrementality habit before continuing to trust platform-reported ROAS at face value.

The convenient lie

Last-click attribution credits whichever channel happened to close the sale, systematically overvaluing high-intent channels like brand search and undervaluing upper-funnel channels that created the demand in the first place.

What incrementality actually measures

The true lift a channel creates versus what would have happened without it, typically measured through holdout tests or geo experiments rather than an attribution model reading the same click data differently.

A realistic starting point

Fund a small number of incrementality tests on the channels carrying the largest last-click share of budget, rather than attempting to overhaul attribution across every channel at once.

FAQ

Are incrementality tests expensive?

They cost less than continuing to misallocate a large ad budget based on flawed attribution—treat them as insurance rather than overhead.

Can small stores run incrementality tests?

Yes, with simplified designs like geo holdouts, though reaching statistical confidence takes longer at lower spend levels.

---

CostRadar's simulations help teams stress-test channel budget shifts before committing spend, a lighter-weight complement to full incrementality testing.