← The Growth Lab · MEASUREMENT
Attribution Is Not Incrementality
Oct 24, 2026 · Read time: 3 min · DV8 Pacific
Every ad platform reports sales. Few can prove they caused them. The difference between the sales a campaign is credited with and the sales it actually created is one of the most expensive blind spots in marketing, and it grows as budgets move into retail media and AI.
What attribution tells you
Attribution connects an ad exposure or click to a later purchase. Retail media networks are especially good at this: they can show that shoppers who saw an ad went on to buy the product. That is useful, but it answers a narrow question. It shows who bought after seeing an ad, not who bought because of it.
High attributed return on ad spend often comes from shoppers who would have bought anyway: existing customers, people searching for the brand by name, buyers of hero products during a promotion, or demand that already existed. The dashboard looks strong while the campaign adds little.
What incrementality tells you
Incrementality measures what happened because of the advertising: the sales that would not have occurred without it. Industry measurement guidelines define it as isolating the causal effect above a baseline, and recommend controlled experiments where possible. In practice, take a group of shoppers who could see the ads and a matched group who could not, and compare how many bought.
Consider an illustrative example. Suppose 6% of the exposed group bought and 5% of the comparison group bought. The campaign created the difference: one percentage point, a 20% lift. The other 5% would have bought anyway. Attribution might credit the campaign with all 6%. Incrementality credits it with one point.
Judge every campaign on both
High attributed return, high incrementality: the credited sales are real added sales. Scale it.
High attributed return, low incrementality: attribution is likely overstating value. Test before adding budget.
Low attributed return, high incrementality: the campaign may be driving sales elsewhere. Look for halo effects.
Low attributed return, low incrementality: little is credited and little is added. Fix or stop.
Why it matters more for growing brands
For brands with limited budgets, the cost of the blind spot is high. Money spent on campaigns that only capture existing demand is money not spent on reaching new customers. And because many brands judge success on attributed returns, the campaigns that look best are often the ones least focused on growth.
A useful question
One question cuts through most measurement debates: if this campaign stopped tomorrow, how much of its attributed sales would actually disappear? If the honest answer is "not much," the campaign is capturing demand rather than creating it. That is not always wrong, but it should be a deliberate choice.
Sources: IAB and MRC, "Retail Media Measurement Guidelines," January 2024; IAB, "Guidelines for Incremental Measurement in Commerce Media," 2025; DV8 Pacific analysis. The 6% and 5% figures are illustrative.