Your retail media budget gets planned like media and produced like paperwork. Eight networks, eight spec sheets, eight sets of rules, and by the time your retail media creative is live, the idea that won the pitch meeting has been resized, re-templated and rewritten into something nobody would've approved as a concept. Nobody can say anymore what the campaign actually says.
Then everyone goes back to the dashboard and argues about bids.
I'm a digital futurist who came up through copywriting and then a long stretch in rooms full of technicians. Here's what I learned in those rooms: none of the numbers matter until the words are right. Retail media is where that lesson costs the most, because it's where the words get chopped up the fastest.
What the industry is finding: only 14% start from one brief
Commerce media is big and getting bigger. One industry revenue report put US commerce media at about $63.4 billion for 2025, up 18% on the year [1].
The newer research is about what's breaking. A 2026 survey of 281 people across brands, agencies and retail networks in six countries [2] asked where commerce media falls apart. The answers line up with everything I've watched happen:
- Asked which parts are least connected, respondents put reporting first at 34%, then creative at 30%. Creative was the least connected area for brands, agencies and networks alike.
- Creative also topped the list of stages most in need of improvement.
- Only 14% said their multichannel campaigns are always planned from a single brief. The same small share said they always get unified reporting.
- When brands were asked what would make them invest more, 46% pointed to proven return and incremental sales.
Read that list like a technician and you'll fund a reporting project. Read it like a strategist and the line that jumps out is the 14%. Eighty-six out of a hundred can't say every campaign they run starts from one brief, which means some of their campaigns go out with no agreed sentence holding them together. The reporting gap is a symptom. The brief gap is the disease.
What the technician sees, and what the strategist sees
The technician looks at retail media and sees a measurement problem. Which network is driving sales? Which placement converts? What's the return on ad spend by retailer? Those are fair questions, and the dashboards are built to answer them.
The strategist looks at the same campaign and asks something the dashboard can't: what did the shopper actually read? Not the plan. The live tile, at the size it ran, with the words that survived the template.
Those two views produce two different meetings, and only one of them finds the actual break. In the technician's meeting, network B is underperforming, so the bids drop and the budget moves. In the strategist's meeting, someone notices network B ran a version of the ad with the proof point cut for length and the headline swapped for a price callout. The network didn't fail. The message did, and nobody could see it from the dashboard because the dashboard only measures what happened after the words were already wrong.
Fix the brief beforeyou fix the bids.
Here's the uncomfortable part for the extreme technicians. You can be the best bidder in the building and still lose, because you're optimizing the delivery of a message that got diluted three handoffs before it reached you. The words move the number. The bids just decide how many people see the words.
Where I learned this: four years of dealer meetings
I spent almost four years with an automaker's dealer advertising association, traveling the Western region and its meetings, helping the ad teams understand what the dealers were really saying and how the money would be spent each quarter.
Think about the structure of that. The brand had a story it wanted told. The dealers had inventory to move, a local market, and opinions about what actually brought people onto the lot. The ad teams sat in the middle, trying to turn both into creative that worked in dozens of local markets. Every handoff, brand to agency, agency to region, region to dealer group, nudged the message toward whoever spoke last.
The quarters that worked weren't the ones with the sharpest media math. They were the ones where somebody wrote down the sentence that mattered and refused to let it get sanded off.
Retail media is the same shape at a bigger size: brand money, spent inside someone else's store, under the store's rules. Without a single brief holding the center, the campaign becomes whatever each network's template allows.
The mechanism: Brief Fracture
Brief fracture is what happens when one campaign idea becomes many briefs. It works through three forces:
- Spec multiplication. Each network has its own formats, sizes, text limits and approval rules. Every spec forces a small rewrite.
- Handoff drift. Each person adapting the work brings their own read of what matters. Small drift per handoff compounds.
- Template gravity. Network creative tools and templates pull every brand toward the same look, which makes the creative easier to produce and harder to tell apart.
Spotting fracture before it ships is one of the reads we practice in the iFix Lab.
The math (illustrative): 96 assets, two thirds of the idea
These are modeled numbers, built to show how fast this compounds. They aren't from any client account.
Spec count. Eight retail networks, each with an average of six placement types, means 48 asset specifications for one campaign idea. Add two size or text variants per placement and you're at 96 assets. Add three and you're near 150.
Fidelity per handoff. Suppose each handoff keeps about 90% of the original message's meaning: the buyer moment, the core claim, the reason to choose you. Then:
- One handoff: 90% of the idea survives.
- Two handoffs: about 81%.
- Three handoffs: about 73%.
- Four: about 66%.
In plenty of retail media workflows, the idea passes from brand to agency to a network's creative team to a template. That's three or four handoffs before a shopper sees anything.
What that costs. If the full-strength idea converts at a certain rate and the diluted version converts proportionally less, losing a third of the message is a big chunk of efficiency gone before bidding even starts. On an illustrative $3M monthly commerce media budget, even a 15% loss in conversion efficiency from dilution is about $450,000 a month of work the media has to make up.
No bid strategy recovers $450,000 a month of lost meaning. A better sentence can. Use these figures to reason with; your own budget will tell its own story.
At 90% fidelity per handoff, a third of the message is gone before bidding starts.
The retail media creative brief that survives eight networks
A brief that survives eight networks is one page, and its most important line is the one that says what adapters may not change. Mine has six lines: the buyer moment, the one thing, the proof, the difference, the next step, and what can change. Every line is something an adapter can check their work against, and the last one does most of the governance work, because it turns every adaptation from a judgment call into a check. If you're the agency owner holding eight network relationships, or the CMO answering a CFO who wants proof of incremental sales, that page is the one document both of you can defend.
The full brief, with what goes on each line and the rules adapters work to, is a template members get inside the iFix Lab.
Second-order effects that get blamed on reporting
- Networks start to look the same. When creative gets diluted to template level, performance differences between networks come down to audience and placement alone. You lose the ability to tell whether a network or a message is working.
- Reporting gets blamed for a creative problem. Teams ask for unified reporting, which matters. But even perfect reporting can't fix a campaign where every network ran a different version of the idea.
- Incrementality gets harder to prove. Retail media sits right next to the point of purchase. Diluted creative leans on catching shoppers already in the aisle, which makes the network look good in its own reporting and adds little that wouldn't have happened anyway.
Where the common advice breaks between $1M and $10M a month
Each network will tell you to follow its best practices. At $1M a month on one or two networks, that's sensible. Their guidance reflects what works on their shelf.
Between $3M and $10M across many networks, following every network's best practices produces a campaign with no center. Each network's guidance is right for that network. Together, they pull the idea apart.
A second rule cracks too: “retail media is lower funnel, so lead with the product and price.” That's technician logic. It keeps you competing on the shelf where the shopper is already comparing, and it makes your spend less incremental, which is the one thing brands say would earn more budget. Price is a number. The reason to choose you is a sentence.
What changes once you're already past $10M a month
- The master brief becomes a governance document. One page, owned by the brand, that every network adaptation must trace back to. The network teams execute it as written.
- You need a creative-fidelity review. Someone checks a sample of live assets per network against the master brief on a schedule. Drift gets caught within weeks, while it's still cheap to fix.
- Quarterly planning has to hold the store's calendar and the brand's story at once. That's the lesson from the dealer meetings: the money moves best when the people spending it hear what the people closest to the buyer are actually saying, and the brief reflects both.
- Retail data should feed the brief as well as the targeting. What shoppers search and compare is buyer language. The technician uses it to target. The strategist uses it to write.
The decision framework I'd use
Before the next commerce media cycle, count the handoffs between your brief and a live tile. Every one of them costs you part of the idea, the way every photocopy of a photocopy loses a little more of the page. Most teams have never written that number down.
The full rollout, step by step, from the brief to how each network earns its budget, is a Lab tool.
Questions senior buyers ask
How would I know this isn't the same old brief template with a new name?
Run the side-by-side test at the end of this note on your own live assets first. If a stranger can match every network's tile to one sentence, you don't need it. If they can't, you've seen the leak with your own eyes.
What makes retail media creative different from social creative?
It runs inside someone else's store, beside your competitors, under the store's specs and approval rules, often at the moment of comparison. That's why the proof line and the difference line carry more weight there than almost anywhere else.
Should we let networks produce our creative?
For format work, often yes. For the message, keep control. Give them the master brief and approve against it.
How do we handle networks with very different audiences?
Change the buyer moment if you must, through the master brief, on purpose, so the change is a decision someone made and can defend.
How do we prove retail media is incremental?
Ask each network for holdout testing, by audience or by store region, and treat refusals as a signal. Compare total sales in test and control alongside the network's attributed number.
The test I'd run this quarter
Pull one live asset from each network for your biggest campaign. Lay them side by side with the original brief. Don't open the dashboard yet.
If a stranger couldn't tell they came from the same idea, you've found where the money's leaking, and no bid change was ever going to plug it.
Fix the sentence, then let the bids do their job. Reading the sentence before the bid is what the iFix Lab trains. Every week the Signal Briefing gives you what moved in search, discovery, signals and conversion, each with a verdict and a move, so your master brief keeps up before the next commerce media cycle is planned. The Lab is $97 a month. Cancel any time; month one comes with a 14-day refund. Join the iFix Lab. Next note: what happens when AI makes the assets nearly free and the vision becomes the only scarce thing left.