Key Takeaways
- Retail media networks sell brand-building potential, but most lack the independent incrementality proof to justify moving dollars from proven brand channels.
- Sponsored listings drive conversion, not awareness, and conflating the two lets networks charge brand-level CPMs for performance-level outcomes.
- Shopper data is only valuable if it connects to media outcomes you can verify outside the network’s own reporting.
- Budget fights between brand and trade teams signal that retail media’s role in the funnel is still undefined.
- CPG brands that run incrementality tests before scaling retail media consistently find more efficient reach through channels they already own.
- The right question is whether or not retail media can build your brand at a cost that beats your next-best alternative.
—
Retail media networks aren’t a brand-building channel by default. They become one only when you can prove the exposure is reaching buyers who would not have converted anyway, at a cost that justifies pulling budget from channels with stronger reach and creative control. Right now, most networks can’t show you that proof. They’re showing you their own attribution dashboards instead.
The Pitch Is Getting Louder While The Proof Gets Thinner
Every major retail media network is now selling upper-funnel placements. Display, video, connected TV, off-site programmatic inventory, all packaged with the promise that shopper data makes it smarter than traditional brand media.
The shopper data angle is real. Knowing that someone buys in your category every six weeks has genuine targeting value. The problem is that targeting precision isn’t the same thing as brand impact, and networks routinely present the former as evidence of the latter.
When a network tells you that your campaign drove a 15% lift in brand awareness among category buyers, ask one question: compared to what control group, measured by whom? If the answer involves the network’s own measurement tools and no holdout methodology you can audit, that number isn’t data.
What Incrementality Measurement Actually Requires
Incrementality isn’t complicated in concept. You expose one group to your media, you withhold it from a matched control group, and you measure the difference in the outcome you care about. The retail media industry’s version of this is frequently something much weaker.
A real incrementality test for retail media brand-building should include:
- A clean holdout group that is statistically matched to your exposed audience and genuinely withheld from the campaign.
- A brand metric you define in advance, not a post-campaign metric the network selects because it moved.
- An external measurement partner with no financial interest in the outcome.
- A cost-per-outcome comparison against your current brand channels running in the same period.
- A minimum test duration long enough to capture brand effect, typically six to eight weeks for most CPG categories.
If a network resists any of these conditions, that resistance is information. We have written about this measurement challenge in the context of how retail media strategy for CPG brands needs to extend beyond sponsored listings. The same rigor applies when the pitch moves from performance to brand.
The Sponsored Listings Trap
This is where most CPG performance marketers are getting burned. Sponsored listings are a conversion tool. They capture demand at the moment of purchase intent. They work. We use them. Our clients see return on ad spend that justifies the investment.
But a sponsored listing isn’t building your brand. It’s harvesting brand equity that was already built somewhere else. When a network repackages those same shopper signals into an upper-funnel product and charges you brand-level CPMs, you’re paying twice for the same buyer: once to build awareness on a channel with real reach, and again to capture them inside a walled garden that would have converted them anyway.
This isn’t a theoretical concern. We see it in client account structures regularly, especially in categories where Amazon, Walmart Connect, and Kroger Precision Marketing are all pitching simultaneously. The overlap in audience targeting across networks means you can easily be running three “brand-building” campaigns that reach the same 40% of your audience three times while leaving the rest of the market dark.
Our post on retail media budget funding and network selection gets into how this fragmentation drives waste at the budget level. Brand building amplifies that problem because the budgets are larger and the feedback loops are slower.
The Budget Fight Is A Signal, Not A Scheduling Problem
When brand teams and trade teams fight over who owns retail media, it’s because no one has defined what retail media is supposed to do in the funnel for that specific brand. Trade teams see it as a distribution tool. Brand teams see it as stolen budget. Both are partially right, which means neither is getting the strategy they need.
The resolution is a clear test-and-learn framework that assigns retail media to one objective at a time, measures whether it achieves that objective against an alternative, and builds the case for investment incrementally. That framework should come from your agency, not from a network’s sales team with a quarterly quota.
How To Pressure-Test The Brand-Building Claim
Before you move any brand budget toward retail media networks, run this checklist:
- Can the network provide a third-party-validated holdout study from a brand in your category?
- Does the off-site inventory they are selling appear on a supply-path transparency report you can review?
- Are the brand metrics they are promising ones your CMO would recognize, or are they platform-specific proxies?
- What is the effective CPM for verified, non-duplicated reach, not the headline CPM?
- What happens to your conversion rate on that network if you pause the brand campaign for four weeks?
That last question is the most important. If the answer is “nothing changes,” the brand campaign was not doing brand work. It was adding spend on top of purchase intent that was already there.
—
FAQ
What is the difference between retail media brand building and retail media performance?
Performance retail media targets shoppers at or near the point of purchase, typically through sponsored listings and search placements. It’s measured by ROAS or conversion rate. Brand-building retail media targets shoppers earlier in their decision process, typically through display, video, or off-site placements, and should be measured by awareness lift, consideration, or new buyer acquisition.
How should CPG brands measure incrementality in retail media?
Start by defining the outcome before the campaign launches, not after. Choose a brand metric your team already tracks. Build a holdout group with a partner who has no stake in the result. Run the test for at least six to eight weeks. Then compare the cost per unit of brand lift to what you’re achieving on your existing brand channels. If retail media wins on that comparison, scale it. If it does not, you have your answer without burning a full brand budget to find out.
Why are retail media CPMs rising even as attribution gets harder?
Networks set CPMs based on demand, and demand from CPG advertisers is growing because retail media is positioned as a cookieless, first-party alternative to programmatic display. Rising demand hasn’t been matched by improvements in measurement transparency or cross-network deduplication. The result is that brands are paying more for reach they cannot fully verify and cannot easily compare across networks.
Should brand budgets ever fund retail media?
Yes, but only after you have proven incremental brand impact through a controlled test. Moving brand budget into retail media without that proof isn’t a strategy. It is a bet placed based on a sales pitch. Run the test first, scale on the evidence, and always keep a cost-per-outcome comparison to your next-best brand channel in the room.
—
Ready to pressure-test your retail media strategy with real incrementality thinking? See how Junction 37 builds performance media frameworks for CPG and DTC brands.
—
Chris Pyne, Founder and CEO of Junction 37 and its sister venture Series A. He built Cortex, J37’s proprietary AI-driven planning ecosystem, and pioneered the integration of predictive marketing science into client strategy. Previously, Chris held C-suite roles at OMD USA and MediaCom, where he led planning for $7B in billings and 700+ employees.