Retail media is entering a high-stakes moment. The promise of automated buying and AI agents is colliding with how brands actually work. The result, if we are honest, will not be tidy. My view: AI agents will force retail media to pick between true customer value and short-term ad tricks.
This matters because retail media now sits at the heart of growth plans for many marketers. The budgets are large, the margins are rich, and the pressure to “make it easy” is intense. When everything gets automated, the incentives inside these platforms will decide who wins—shoppers or ad systems.
The Coming Fight Over Value vs. Convenience
One executive put it bluntly:
“Brands may seek out buying simplicity, and AI agents could mean a ‘battle for the soul of retail media.’”
I agree. Simplicity sounds great. But simplicity without guardrails can turn into a race to the bottom. If AI agents optimize only for speed and cheap reach, shoppers will get spam, not help. And brands will pay for inventory that looks efficient but doesn’t build demand.
AI agents will be asked to do the work of dozens of traders in seconds. They will pick keywords, placements, and bids. They will shape creative and promotions. Who programs their goals will decide what “good” looks like. Clicks? ROAS? Margin? Loyalty? The answer changes everything.
What’s Really at Stake
The core question is simple: Will retail media deliver shopper value first, or protect an easy revenue stream? If the former, AI will help shoppers find better products, at fair prices, with less noise. If the latter, AI will crank out ads that win auctions and lose people.
Here’s how the tension shows up in practice.
- Optimization targets: Agents trained on short-term sales will over-serve bottom-funnel ads and weaken brands.
- Data privilege: Retailers with closed data may lock out rivals, hurting choice and innovation.
- Creative control: Auto-generated ads can drift from a brand’s voice or misread context.
- Measurement fog: Black-box models make it hard to prove real incrementality.
- Shopper fatigue: More “relevant” ads can still feel like clutter if frequency runs wild.
None of this is destiny. But it will be the default if platforms chase volume and brands chase ease. Convenience is not a strategy.
Counterpoint—and Why It Falls Short
Some will say automation is the only way to manage scale. True. But scale without standards is just noise at speed. Others argue that AI will personalize ads so well that everyone wins. Maybe—if the systems are judged on long-term outcomes, not just cheap conversions.
We should also be cautious about “set-and-forget” promises. If you hand the keys to an agent and never audit its choices, you are not saving time. You are outsourcing judgment.
How Brands Can Keep Control
If brands want simplicity, they should ask for it on their terms. That means building a clear set of rules that AI must follow and auditing results with human sense.
- Define success: Use incrementality, new-to-brand, and profit, not only ROAS.
- Set creative limits: Lock guardrails on claims, brand tone, and product fit.
- Demand transparency: Ask for model features, not just scores and dashboards.
- Control frequency: Cap exposure to protect shopper trust.
- Test often: Run holdouts and clean-room checks to validate lift.
- Protect data: Share only what you must and track where it flows.
Retailers have work to do as well. They should offer clear measurement, fair auctions, and tools that align price with proven value. When retailers win only if shoppers win, the system stays honest.
What I Want to See Next
I want retail media platforms to publish their optimization goals and give brands choices. Do you want growth in new buyers? Repeat purchase? Margin? Let brands pick, then report against that choice. I also want creative systems that explain why they made a change, not just that they did. Explanations build trust.
Finally, I want a shared code of conduct. Set standards for disclosure, frequency, and measurement. Make it easy to compare platforms on real outcomes, not vanity metrics.
The Bottom Line
AI agents will shape retail media’s next decade. The question is not whether they arrive, but what they serve. Let them serve people, not just platforms.
Brands should push for simple tools with smart guardrails. Retailers should tie their profits to shopper value. And we should stop treating convenience as a substitute for judgment. If we get this right, automation will clean up the ad mess, not multiply it. If we get it wrong, the “soul” in that warning will not be a metaphor—it will be what the industry gives up.
Ask your partners for transparency, measure what truly grows your business, and keep a human hand on the wheel. That is how we keep retail media worth the name.
