Retail media promised closed-loop measurement. Here is what it actually delivered — and what is still missing.
Retail media has been the dominant advertising story for three years. The central promise was simple: buy media closer to the point of purchase, connect it to first-party shopper data, prove ROI that traditional channels could not. The reality is more complicated.
Retail media has been the dominant advertising story for three years. The central promise was simple: buy media closer to the point of purchase, connect it to first-party shopper data, prove ROI that traditional channels could not. The reality is more complicated. Most retail media closed-loop measurement tells brands what happened inside one retailer's ecosystem. It does not tell them what their broader campaign portfolio actually produced. That gap is the next measurement problem retail media has not solved.
Retail media closes the loop within a single retailer's ecosystem. It does not connect to the brand's full media investment — or to what happens at every other retailer in the portfolio.
The fastest-growing advertising channel in history
The numbers behind retail media's rise are genuinely striking. Global retail media advertising spend reached approximately $140 billion in 2024, growing toward $184 billion in 2025 and projected to reach $312 billion by 2030, according to Forrester's forecast. In the United States, retail media represented one in every seven advertising dollars in 2024, with total US spend reaching $54.5 billion.
Amazon commands the largest share: approximately $46.9 billion of global retail media spend flows through Amazon's advertising platforms, representing roughly a third of the entire global market. Walmart Connect generated $4.4 billion in 2025, growing 26.4% year-over-year. Target's Roundel, Kroger Precision Marketing, and dozens of category-specific networks round out a landscape that now counts 277 distinct retail media networks globally as of November 2025, according to industry tracking data.
The speed of this growth is not accidental. It was driven by a specific promise — one that arrived at precisely the moment brands most needed it.
The promise that drove the investment
The timing of retail media's rise is inseparable from the collapse of third-party cookie-based digital attribution. As cookies deprecate, platform tracking restrictions expand, and GDPR and similar regulations erode the user-level data that digital attribution relied on, brands found themselves in a measurement crisis precisely as their digital spend continued to grow.
Retail media offered an answer that seemed structurally compelling: instead of relying on proxies and probabilistic modeling to connect ad exposure to purchase behavior, brands could advertise within the retailer's own ecosystem — where purchase data is deterministic, first-party, and directly linked to the advertising exposure. The closed loop was not a measurement aspiration. It was described as a structural advantage of the channel.
Marketing Dive's analysis, citing Forrester's research, captures the promise clearly: the rise of first-party identity data combined with transactional purchase records offered "the promise of closed-loop measurement." Retailers possess a data asset no other advertising platform can replicate — deterministic purchase data from millions of real transactions.
Nearly two-thirds of marketers increased their retail media investments in 2024 on the basis of this promise. The investment case was straightforward: buy media inside the ecosystem where purchase data already exists, and the attribution problem is solved.
What the closed loop actually measures
The measurement that retail media's closed-loop architecture delivers is real and valuable. It is also narrower than the promise suggests — in ways that have become increasingly visible as brand investment in the channel has grown.
Retail media closed-loop measurement tells brands what happened to their advertising inside a specific retailer's ecosystem. For a brand advertising on Amazon, the loop closes around consumers who saw the sponsored product, clicked through, and purchased on Amazon. For a brand on Walmart Connect, the loop closes around Walmart's customer base, Walmart's app and website, and Walmart's point-of-sale data.
That is a genuine commercial signal. But it is not the same as knowing what the brand's campaign portfolio actually produced commercially.
Consider what the closed loop does not capture for a brand running a typical omnichannel campaign:
What TikTok creator content did to demand — the awareness and consideration generated by creator exposure that primed consumers before they ever searched inside a retailer's platform. Retail media attribution credits the sponsored search impression that converted. It does not credit the creator video that generated the search intent.
What happened at every other retailer in the brand's distribution — Amazon's closed loop closes on Amazon. It says nothing about what the campaign did to sell-through at Walmart, Target, Kroger, CVS, or the independent pharmacy where a meaningful fraction of the brand's volume moves. The loop is complete within a single ecosystem. The rest of the commercial landscape is invisible to it.
What happened in physical stores — eMarketer's November 2024 data makes this structural gap explicit: 80% of consumer spending is in-store, but 90% of retail media advertising is online. The closed loop is primarily a digital-channel measurement architecture applied to a market where the dominant commercial outcome is physical. In-store retail media is growing — IAB released the first industry-wide in-store retail media measurement standards for public comment in September 2024 — but connecting digital ad exposure to in-store purchase behavior remains the unsolved problem at the heart of the channel's measurement infrastructure.
What the brand's full media investment contributed — a consumer who purchased through Amazon after seeing a sponsored product may have been primed by a Meta campaign, a TikTok creator, a TV spot, and three retailer email promotions before the sponsored search moment. The retail media attribution model assigns credit to the sponsored product. The brand's other campaign investments are invisible in that attribution, creating systematic misattribution of commercial contribution across the portfolio.
The Forrester verdict and what it means
Forrester's State of Retail Media 2025 is the most direct recent assessment of where the gap sits. 86% of commerce media decision-makers in North America and Europe say strengthening measurement and attribution is a high or critical priority. That is not a number from brands who are new to the channel and learning how it works. It is a number from the most sophisticated retail media buyers, after years of investment and experience.
The same research identifies the specific disappointment: "The retail media ecosystem has largely failed to deliver on promises surrounding data usability, leaving many marketers frustrated." Nikhil Lai, senior analyst at Forrester, describes the gap directly: "There's a divergence between retail media, sponsored content and its reality."
From within the channel, the gap is framed as a measurement standards problem. The IAB and Media Rating Council released the Retail Media Measurement Guidelines in 2024 — establishing the first industry-wide framework for how retail media performance should be reported and compared. IAB Europe published updated pan-European retail and commerce media definitions in March 2025. These are genuine steps toward measurement consistency. They also confirm that the standards retail media's closed-loop promise implied were already in place have only just begun to be established.
The three biggest challenges in adopting retail media, according to TransUnion's 2025 Annual Trends Study, are scale, targeting, and measurement. Difficulty proving investment incrementality is cited by 36% of marketers as a factor that could cause them to invest less in retail media — higher than concerns about ROI or channel performance directly.
The incrementality problem embedded in the promise
The most consequential gap between retail media's closed-loop promise and its delivered measurement is the distinction between attribution and incrementality.
Attribution asks: which touchpoints were present when a purchase occurred? Retail media's closed loop answers this question for the touchpoints inside the retailer's ecosystem.
Incrementality asks: which advertising actually caused the purchase, rather than simply being present in the journey of a consumer who would have purchased anyway? This is the question that determines whether the brand's media investment produced commercial outcomes that would not have occurred without it.
Most retail media measurement delivers attribution. Very little delivers incrementality. The distinction matters enormously for budget decisions.
A brand that invests heavily in sponsored product advertising on Amazon and sees a high attributed ROAS is learning how many purchases occurred in the presence of the ad. It is not necessarily learning how many purchases the ad caused. A consumer who searched for a brand they already intended to purchase, saw a sponsored product in the search results, and clicked through has been attributed to the ad — but the ad may not have been the cause of the purchase intention that drove the search in the first place.
This is the attribution inflation problem that retail media shares with every other channel that measures correlation rather than causation. The difference in retail media is that the closed-loop framing created an expectation that the deterministic purchase data solved this problem. It did not. It made attribution more accurate. It did not make it causal.
The move toward incrementality testing — holdout groups, matched market experiments, geographic split tests — is the industry's recognition that attribution alone does not answer the budget optimization question. Incrementality sales measurement is now the third-highest priority for retail media buyers in 2025, cited by 37% in IAB Europe's research. The measurement infrastructure that would make this routine is still being built.
The fragmentation problem compounds the gap
The ecosystem-specific nature of retail media's closed loop is not just a measurement philosophy problem. It is a practical operational problem that grows worse as brand investment in retail media expands across multiple networks.
A brand running campaigns across Amazon, Walmart Connect, Target Roundel, and Kroger Precision Marketing receives four separate attribution reports, each using different attribution windows (Walmart's 14-day versus Kroger's 7-day, as documented in retail media benchmark analysis), different measurement methodologies, different definitions of a conversion event, and different formats for delivering results.
In Europe, 60% of buyers cited a lack of cross-network standards as a significant barrier to investment in 2024, according to Kevel's analysis of the retail media landscape. The IAB's standardization efforts address this over time. In the meantime, the brand's attempt to understand its total retail media investment across multiple networks requires manual reconciliation of reports that were not designed to be compared.
The portfolio-level question — what did the brand's entire retail media investment produce commercially, and how does that interact with its non-retail media investment — remains unanswerable from the data each network delivers. The closed loop closes inside each ecosystem. It does not close across the brand's total commercial picture.
What true closed-loop measurement requires
The measurement architecture that would actually close the loop between brand campaign investment and commercial outcomes requires four things that retail media's current infrastructure does not provide, and cannot provide by design:
A cross-retailer view of sell-through. Understanding what a brand's campaign portfolio produced commercially means knowing what happened at every retailer in the brand's distribution — not just the one whose advertising platform the brand used. Amazon's closed loop tells the brand what happened on Amazon. What happened at Target, CVS, Walgreens, and the convenience channel in the same period is not in the Amazon report.
A connection from digital campaign exposure to physical purchase. With 80% of consumer spending occurring in-store, a measurement system built around digital purchase attribution is measuring a minority of the commercial outcome. Connecting online campaign signals to offline sell-through requires methods that retail media's within-ecosystem architecture cannot deliver — geographic causal inference, sell-through data from physical retailers, and the analytical infrastructure to connect them.
Cross-media causal attribution. Understanding what the brand's sponsored product on Amazon actually contributed requires knowing what primed the consumer before they searched: which creator content, which paid social exposure, which brand campaign touchpoint generated the consideration that drove the search. Retail media measurement attributes the sponsored search. It does not distribute credit across the full campaign portfolio that produced the commercial outcome.
Incrementality, not attribution. The question that matters for budget optimization is not how many purchases were present in the ad's attribution window. It is how many purchases the ad caused. Answering this requires causal inference methods — difference-in-differences, Bayesian structural time series, holdout-based experiments — that operate above the platform level, on commercial outcome data that the platforms do not own.
Where behavioral intelligence sits in the measurement stack
Retail media is not going away. The channel's scale, growth rate, and genuine measurement advantage within its own ecosystem make it a permanent and important part of the omnichannel marketing mix. The closed loop it delivers within a single retailer's ecosystem is better measurement than most other digital channels provide within their own context.
The gap is not that retail media's measurement is bad. The gap is that it closes the loop on a subset of the commercial picture, and the subset it closes is not the one that answers the questions brand CMOs and CFOs actually need answered.
Those questions are: what did this campaign produce across the full distribution footprint? What was the incremental commercial contribution of each channel — not just the one where the conversion was recorded? How should the brand allocate budget across retail media, social campaigns, creator content, and brand advertising to maximize offline sell-through across every retailer and distribution channel in the portfolio?
These are behavioral intelligence questions. They require a system that sits above the individual retail media networks, connects to sell-through data across the full commercial footprint, applies causal inference methods to isolate true incremental contribution, and returns the kind of structured direction that allows the brand to optimize its total campaign investment rather than the performance of individual network buys in isolation.
Retail media closed the loop inside the ecosystem. Behavioral intelligence closes the loop between the brand's full campaign investment and its commercial outcomes across every channel where those outcomes actually occur. These are not competing approaches. They are different layers of the same measurement stack — one of which has been built, and one of which is now urgently needed.
Sources and references
- Forrester Research. State of Retail Media, 2025. Global retail media market: $184 billion in 2025, projected $312 billion by 2030. 86% of commerce media decision-makers say strengthening measurement and attribution is a high or critical priority. Via TechBullion and Osmos.ai.
- Kevel. Looking Back on Retail Media Trends and Challenges from 2024. Global retail media ad spending: $140 billion in 2024. US: $54.5 billion, one in seven advertising dollars. Amazon: $46.9 billion (one-third of global spend). 60% of European buyers cited lack of cross-network standards as barrier to investment. March 2026.
- Fugo.ai. Retail Media Growth, Statistics, and Trends for 2026. "80% of consumer spending is in-store, 90% of retail media advertising is online." eMarketer Forecast, November 2024. 277 retail media networks globally as of November 2025.
- Marketing Dive / Forrester. The Future of Retail Media Networks. "The retail media ecosystem has largely failed to deliver on promises surrounding data usability." Quote from Nikhil Lai, senior analyst at Forrester. August 2025.
- Marketing Dive. What Marketers Must Do to Get Retail Media Measurement Right. Paul Frampton, President CvE: "The rise of the first-party ID and data offers the promise of closed loop measurement. However, there is both hype and nascence in the reality."
- Osmos.ai / Forrester. Retail Media in 2026. "The gap between what retail media promises (closed-loop measurement) and what most networks actually deliver (last-click attribution within limited windows) is the single largest source of advertiser frustration."
- Infillion. The Omnichannel Measurement Gap Retail Media Still Hasn't Closed. Nielsen finding: "many advertisers assume retail media naturally solves attribution because of its proximity to transaction data, even though connecting exposure to outcomes across environments remains complex." February 2026.
- IAB and Media Rating Council. Retail Media Measurement Guidelines released 2024. IAB and IAB Europe: first industry-wide in-store retail media measurement standards released for public comment, September 2024. IAB Europe: pan-European definitions published March 2025.
- IAB Europe. 2025 Attitudes to Retail Media Report. Incrementality sales measurement cited as priority by 37% of buyers. November 2025.
- TransUnion. 2025 Annual Trends Study. Three biggest retail media challenges: scale, targeting, measurement. Difficulty proving incrementality cited by 36% as potential reason to invest less.
- Walmart Connect / various. Attribution window discrepancy: Walmart 14-day vs Kroger 7-day. Via RMIQ Retail Media Guide 2025.
- RMIQ. 2025 Retail Media Market Guide. Amazon: 77% of US retail media market; $56 billion global advertising revenue for 2024. Walmart Connect: $4.4 billion in revenue, 26.4% growth.
Veinera is building the behavioral intelligence layer that connects your full campaign investment to offline commercial outcomes — across every retailer and distribution channel in your portfolio, not just the one whose advertising platform you used. Book a 30-minute walkthrough with a Veinera specialist, no commitment.
Related reading
- The offline data desert — why the most valuable behavioral signal is the hardest to reach · Apr 19, 2026
- The AI feature problem — why bolting intelligence onto reporting does not produce intelligence · May 10, 2026
- Explanation over reporting — why analytics is due for a rewrite · Mar 28, 2026
