AI is reinventing retail media, but it continues to grow: Criteo exec

AI tools like ChatGPT are changing how people shop, and ads on a retailers’ sites can matter less when shoppers bounce between platforms. That, in turn, changes the way retail executives view retail media.

Sherry Smith, president for retail media at adtech network Criteo, told Retail Brew we’re in a transition phase for retail media. AI should not be seen as an external threat, she said, but testing and learning with different types of ad formats that can support conversational commerce is important.

One tool for this is Page Intelligence, which uses AI to optimize product pages so retailers and brands can see ad impressions and sales together, not separately, Smith said. Last year, Criteo started testing Page Intelligence and has recently started tests with retailers, according to the company.

Smith spoke with Retail Brew about the state of retail media, growing scrutiny around ad measurement, and the company’s relationship with AI platforms.

Where is retail media in 2026?

It’s continuing to grow, but being reinvented by AI. We’re seeing the industry evolving from more of a [defined] channel into being the retail media industry into more of an AI-powered outcomes-driven ecosystem. That really spans across discovery to conversion and core surfaces like product pages. We see it becoming more dynamic and more optimized in real time, and so it’s really the year that retail media is really growing up. In fact, we believe 2026 is going to be remembered as the year that retail media does grow up and goes from growth at all cost to measurable outcomes.

The industry is moving from more rapid expansion to more reinvention and that is driven by AI and the rising performance expectations that retailers are getting from brands and agencies. There’s greater scrutiny around measurable outcomes and accountability.

Why is there greater scrutiny around measurable outcomes after AI?

You have not only different brands and agencies looking at the metrics differently, but you also have retailers that are providing measurement differently. And I’ll give you a really basic example: A retailer and a brand may choose to look at the “look back window” for when a customer actually buys a product and how that ad actually converts. Someone might look at it for seven days after an ad is shown. Some might have a look back window that spans further than that. So it's not just ROI, but what’s the best way to look at the media as a whole to see what the effectiveness is.

And the reality is that there are still some differences between what brands value versus what retailers value versus agencies, and we are trying to do more to bring a little bit more clarity and consistency into what that looks like within our own unified platform.

What is your equation with platforms like ChatGPT?

We do see it as an enabler. We see it as another traffic driver. The decisioning might be further down the road by the time that it reaches a retailer, and be more upper funnel in nature. They have made a decision, and they’re landing on a product detail page.

We think that from a chat experience, the way that consumers are shopping, they’ve made their decision up to a certain point, and they will go and they will convert on that retail.

But it’s important that retailers are thinking about the way those product detail pages look and how they perform, because they’re not going through their homepage. They’re actually landing on these pages that up to this point really haven’t been optimized or utilized in a way that they should be.

And that needs to change, and we’re doing a lot of work against that. That’s part of how Page Intelligence will help the retailers as well as the brands to prepare for this agentic future. AI might change the way that people continue to discover products, but we believe, inherently, retailers still own that transaction and that trust.