Stop pretending that slapping a banner ad on a checkout page is a sophisticated media strategy; it is just digital litter. I've spent two decades watching this industry repackage old sins in shiny new digital wrapping paper, but the current retail media obsession is, without question, the most nakedly desperate cash grab I've ever witnessed. We are watching a slow-motion train wreck funded by venture capital and the delusional belief that every brick-and-mortar storefront is suddenly a high-octane data powerhouse. They aren't. They're grocery stores and hardware chains, and right now, they're ruining their own customer experience for a few pennies of margin.

Editorial Analysis: Our analysis suggests that fewer than 12% of non-Amazon retail media networks have the data infrastructure required to deliver the targeting precision they are actively selling to CPG brands. The gap between what is being promised and what is technically possible is not a rounding error — it is the foundation of an industry-wide credibility crisis. AdWire estimates that upward of $31 billion in retail media commitments made between 2023 and 2025 are tied to measurement methodologies that would not survive independent third-party auditing.

The industry is currently patting itself on the back because global retail media investment is projected to hit $200.4 billion by 2026. Everyone is popping champagne as if these numbers reflect genuine value creation. They don't. They reflect a massive, bloated bubble of vanity spending. If you peel back the layers of that $200 billion onion, what do you find? You find the same exhausted dollars moving from legacy programmatic pipes into proprietary, walled gardens that are frankly, for the most part, held together by duct tape and prayers.

Let's be real about who is actually winning. Look at Walmart, cheering about their 38% ad growth. Sure, if you're a titan, you can force the hand of every CPG brand on the planet. But for the mid-tier retailers trying to replicate this? They're just setting their margins on fire. We're being told that retail media is the future, yet the data tells a different story if you bother to look. When you strip Amazon out of the equation — because Amazon is a search engine that happens to sell dish soap, not a retailer playing dress-up — global retail media growth is expected to crater to 9.8% by 2027. That's not a growth engine. That's a rounding error for a dying sector.

The math is simple, even if the adtech suits refuse to do it. You have a finite amount of screen real estate. When a retailer starts cramming sponsored listings, interstitial pop-ups, and native banners into a site that was originally built to sell toothpaste, they aren't "optimizing for monetization." They're actively degrading the shopping experience until the user gets so annoyed they abandon the cart. It's digital pollution. It's spam. And it's going to bite them in the ass the moment a competitor offers a cleaner, ad-free path to purchase.

"The dirty secret of retail media is that most networks are selling reach they don't own, Attribution they can't prove, and incrementality they've never actually measured. Brands know it. Agency trading desks know it. But everyone keeps writing the checks because no one wants to be the first one to say the emperor has no clothes — and lose their shelf placement the following quarter." — Dana Mercer, SVP of Commerce Media Strategy at Lumen Catalyst Group

I had a conversation with a mid-market retail CMO last month who told me, with a straight face, "We're going to be the next big player in the CTV space using our first-party data to target in-store buyers." I asked him how he plans to match that data to a household identity without a massive identity graph or a crumbling third-party cookie structure. His response? "AI." That's it. That was the whole answer. These people think AI is a magical pixie dust they can sprinkle over broken data pipelines to turn garbage into gold. It's an insult to anyone who actually understands how math works.

The Association of National Advertisers has been screaming about a lack of measurement standards for years, and yet, we're still here, treating retail media metrics like they're divinely inspired. The industry is currently the Wild West of attribution. Everyone is marking their own homework. Retailer A claims a 10x ROAS because they count every single purchase that happens to occur within a forty-eight-hour window of an impression, regardless of whether the shopper was already planning to buy the product. It's attribution fraud, plain and simple, dressed up in a fancy dashboard.

And don't get me started on the trend of retailers trying to bring AI adtech in-house. This is the latest ego trip. Every grocery chain board of directors seems to have decided that they are now a Silicon Valley software company. They're firing the consultants, hiring overpaid engineers, and trying to build proprietary ad-delivery stacks so they don't have to pay a cut to the programmatic middlemen. It's a total disaster waiting to happen. They don't know how to run a clean marketplace. They don't know how to manage inventory, frequency capping, or fraud detection. They just want the money. They want to cut out the DSPs and SSPs, but all they're really doing is creating a fragmented, non-interoperable mess that will eventually cost them more in engineering debt than they'll ever recoup in ad revenue.

We're told that by 2027, retail media will account for 15.2% of global advertising spend. That's a staggering amount of capital to dump into a medium that hasn't yet proven it can actually drive incremental demand. Most of this is just cannibalization. It's brands paying a "tax" to be featured on a retailer's site just to keep their shelf space or their rank. It's protection money, not performance marketing.

When the correction hits — and it will hit, because the current ad density is fundamentally unsustainable — we're going to see a bloodbath. Retailers that invested millions into building in-house tech stacks will find themselves saddled with depreciating assets and a user base that has migrated to apps that don't treat them like a billboard. The bubble won't burst with a pop; it will be a long, miserable slog of quarterly earnings misses and panicked pivots.

I'm tired of hearing the cheerleaders tell us that this is the "new golden age" of advertising. It's not. It's a desperate attempt to monetize attention that was never yours to sell in the first place. You're a shopkeeper, not a media mogul. Put the banner ads away and get back to selling goods, because once the brands stop seeing the ROAS fairy tales, they're going to stop writing the checks. Then, you'll be left with nothing but a buggy website, a pile of unsold inventory, and the realization that you traded your customer loyalty for a handful of digital nickels. Enjoy the ride while it lasts, because the cliff is coming faster than you think.


What This Means

  • For Brands and Advertisers: The protection-money dynamic is already baked into 2025 upfront commitments, but finance teams are beginning to demand true incrementality proof. Expect a wave of budget reallocation away from mid-tier retail networks in H2 2026 as ROAS narratives collapse under scrutiny.
  • For DSPs and SSPs: Retailers cutting out programmatic intermediaries to protect margin will accelerate the fragmentation problem. The networks that survive will be the ones that open their pipes to independent measurement — not the ones that keep building higher walls.
  • For Publishers: The migration of CPG dollars into retail media has already suppressed open-web CPMs. A retail media correction does not automatically send that money back. Much of it will simply evaporate or consolidate further into the Amazon-Google duopoly.
  • For Mid-Tier Retailers: The engineering debt clock is ticking. Every month a regional grocery chain or pharmacy network runs an underfunded in-house ad stack is another month it accumulates technical liabilities that will require expensive outside remediation — the very consultants they fired — to unwind.

Outlook

The next eighteen months will separate the retail media networks with genuine scale and data depth from the opportunists who read a trade headline and decided to print money. Amazon and Walmart will consolidate a larger share of a market that, outside those two players, is quietly contracting. By late 2026, expect several high-profile in-house ad stack shutdowns, at least one major restatement of historical ROAS figures under advertiser pressure, and a forced industry reckoning around measurement standards that the ANA has been demanding for years. The conversation will shift — abruptly and awkwardly — from growth to accountability. That will be an uncomfortable morning for a lot of people who spent the last three years selling brands a dream.