NewAugust 2026 · Incrementality
Controlled experiments that measure the revenue a channel actually caused — above and beyond what would have happened anyway.
MTA tracks which touchpoints appeared in a customer’s journey. MMM finds patterns between spend and revenue over time. Neither can answer whether a customer would have bought anyway, even without the ad. Incrementality testing answers exactly that.
What’s new
01Test vs. control experimentsa holdout group is withheld from your ads; the difference is your incremental impact
02Geo testingcomparable groups built by geography when a randomized holdout isn’t possible
03Calibration for MTA and MMMresults feed back to correct over- and under-attribution
04Actionable outputsresults expressed as iROAS and calibrated ROAS
Why it matters
Ad platforms report ROAS on their own attribution logic, which is almost always inflated — they only see their own touchpoints and have an incentive to take credit. And a significant portion of ad spend goes to customers who would have converted organically; high-ROAS retargeting is the classic example.Signal lossGeo-based incrementality doesn’t rely on user-level data or pixels, so it holds up as iOS privacy changes and cookie deprecation give MTA more blind spots.