Big question
Did retail media cause the sale, or claim one that was going to happen anyway?
The position
Discount the network's number, and set the discount from your own matched comparison rather than from a benchmark. Nothing published compares retailer-attributed sales with brand-recorded sales across a real sample.
Why it is live
The baseline is the network's own model and you cannot inspect it, and the network reporting the sale also sold you the media. The IAB and IAB Europe issued incremental measurement guidance for commerce media in November 2025, and the ANA published retail media measurement standardisation work on 18 August 2026, both because retailer definitions and windows produce answers nobody can compare.
The case for the attributed figure
InferredThe retailer sees the basket, which is more than any other seller of media can see, and the association is computed on real orders.
The case against
ContestedTwo primary attribution documents disagree with each other, one setting default windows and the American one saying in its own words that it does not prescribe them.
Where the weight sits
Against the reported figure, strongly. Standards now exist and comparability does not, and a standard nobody has been audited against changes nothing on a plan.
What would change this
A measurement body accrediting attributed sales as a named metric for a named network, or one large brand publishing its own retailer against books comparison.
Where this is taught
Connected Retail Media
Sources
- IAB and IAB Europe incremental measurement in commerce media, 3 Nov 2025secondary
- ANA retail media measurement standardisation, 18 Aug 2026secondary