Retail media faces spending squeeze over measurement mess

By August 2026, the ANA’s measurement initiative had a clear organizing target: force alignment around the basic definitions that retail media networks had been using differently. Bizcommunity, citing Dentsu’s analysis, described a market where performance could not be reliably compared because networks defined and reported results through incompatible methodologies, especially in attribution, where “some retailers use a seven-day window, while others extend this to 14 or 30 days,” making any common framework less a refinement than a prerequisite for coordination.

That is why the ANA’s move mattered in August: it advanced the debate from broad complaints about inconsistency toward a shared framework for what should be standardized first. Bizcommunity argued that standardization had to cover baseline metric definitions and attribution rules, while Retail Customer Experience underscored the urgency by warning that “attribution windows inherited from digital assume a tighter loop than physical retail has” and that “set a seven-day window… people buy monthly and you will systematically under-credit,” exactly the kind of distortion a common framework was meant to address.

The ANA-led push is a response to a market whose scale has made measurement fragmentation impossible to treat as a niche annoyance. More than 200 retail and commerce media networks now operate globally, increasing platform fragmentation and complicating campaign measurement for marketers, while Business Insider described the landscape as “noisy and fragmented for CMOs,” with advertisers effectively running multiple incompatible playbooks that make planning, comparison, and scaling far harder than the growth story suggests.

What turns that fragmentation into a budget problem is that the inconsistencies are not cosmetic: Traders Union reported an ISBA and MediaSense audit of five major UK retail media networks across more than 200 criteria found significant inconsistencies in metrics, including five different versions of what constitutes an attributed sale. That helps explain why IAB Europe research found 53% of ad buyers said a lack of standardization was a barrier to retail media investment, and why one study found reported iROAS could vary by 6.5x on average, with 83% of campaigns flipping from positive to negative based solely on methodological choices.

The ANA-backed effort is not promising precision so much as a common language for judging retail media results. In the August 2026 case study, the shopper purchase rate framework is presented as a universal way to combine baseline scorecard inputs into one view, so brands can compare total investment, units sold, and return across placements without relearning each network’s reporting logic, while ROAS remains a supporting measure rather than the only lens.

That confidence-building logic is becoming concrete in measurement rules. On September 17, 2026, IAB Europe opened revised in-store retail media measurement standards for public comment, proposing that retailers stop using ad frequency as the basis for audience counting and instead start with footfall, use one shared calculation, disclose the assumptions built into it, and report sales over four-week periods before and after a campaign. The fix also depends on narrowing definitions so networks report against the same core metrics and disclose how they derived them; Version 2 says Ad Play is not the right starting point for media measurement, defines an in-store impression as visual contact with the ad, and notes that the December 2024 IAB and IAB Europe standard used an OTS estimate with eight inputs, while the September 17, 2026 draft reduced the formula to six inputs and ranks four methods by precision, requiring retailers to disclose which one they used.