Published, July 15, 2026
Following Advertima’s contribution to the IAB Europe Guide to AI in Retail & Commerce Media, we sat down with Andreas Sofos, Senior Director Data Science at Advertima, to explore the technology behind Audience Intelligence in Retail Media, why verified measurement matters and what the future of In-Store Retail Media could look like.
If you had to explain Advertima’s technology to your grandmother, where would you start?
Andreas: “Grandma, imagine you’re in the health and beauty aisle, shampoos, skincare, that section. There’s a screen there playing an ad for men’s razors. If you see this ad, would you buy this product?“
Grandmother: “No, why would I?“
Andreas: “Fair enough. What would you actually want to see there instead?“
Grandmother: “Maybe a good hand cream.“
Andreas: “Now imagine that same screen, right as you’re standing there, shows you a discount on hand cream instead. Would you look? Would you buy it?“
Grandmother: “Yes, probably.“
Andreas: “Well, that’s what we do. Today, that screen showed you the razors anyway, because it had no idea a lovely lady your age was standing right in front of it. What we build is the part that notices, without ever knowing who you are, just that a lady like you is there and makes sure something worth your attention shows up instead of something random.”
“AI is how we do it.
It’s not the destination.”
Andreas Sofos, Senior Director Data Science at Advertima
The IAB Europe guide talks about AI. But what problem are we actually solving?
“Online Retail Media works because you can measure everything: who saw an ad, whether they engaged, whether they bought. In a physical store, none of that existed. You had a screen, and you hoped people looked at it.
That’s the gap we solve. We bring the same measurement that made Online Retail Media work into the physical store, making sure the right ad reaches the right audience at the right time, so a store can become a real Retail Media channel where everything is measured and tied to outcomes, instead of just a place with screens in it and where online and in-store measurements can be linked together. AI is how we do it, it’s not the destination. The destination is that stores can finally prove what digital advertising has always been able to prove.”
The guide features Advertima’s Audience-based programmatic case study. Let’s go beyond what’s written. Walk me through the project from your perspective.
“The brief came through Sensodyne and Publicis: activate shoppers aged 20 to 45 by gender inside Carrefour UAE stores. At the time, ads were played on a fixed rotation no matter what happened in front of the screen or whether people were there or not.
We plugged Audience Intelligence into the existing media stack. The Neuron served as the Demand-Side Platform (DSP), while PlaceExchange served as the Supply-Side Platform (SSP). This allowed a specific toothpaste creative, for example whitening or stain removal, to be triggered only when the target shopper was actually present in front of the screen through programmatic Audience-based Activation rather than a fixed loop.
The results were impressive across the funnel, both on ad engagement and sales uplift, compared to the fixed ad rotation.”
What surprised even you during the project?
“What surprised me was how fast and effectively the entire ad tech stack came together. Getting that many moving parts synchronized this cleanly, this quickly, is genuinely hard to pull off.
Moreover, the collaboration between Advertima and the analytics team of Precision Media, MAF Carrefour’s media unit, linking upper funnel ad engagement metrics all the way through to lower funnel sales uplift, led to a clear, impressive overperformance for Audience-based programmatic delivery versus the fixed loop.
No matter which KPI you looked at, screen time needed to reach 1,000 target impressions, screen time or impressions needed to reach 1% of sales uplift, or pure incremental sales lift, every layer of the funnel told the same story: audience-based delivery drove efficiency and performance across the board.”
One sentence in the guide really stands out: “From selling screen time to selling verified impressions.” Why is this such a fundamental shift?
“Screen time determines how long an ad will play per hour. But it does not ensure its impact. But impact is the reason why brands use advertising in the first place. Impression, provided by Advertima, is a verification that a shopper was there while a specific ad was shown. It’s a proof of what brands are going for, and how Online Retail Media, which became so successful, works. In-store needs to provide the same measurement standards.”
If you could remove one KPI from Retail Media forever…
“I’d remove OTS, Opportunity to See, as it’s commonly calculated today. It’s a modeled figure, built from footfall and a series of layered assumptions, and it served the industry well before more advanced technology existed to measure audience presence directly.
But from where I sit, every metric and every insight should be backed by data and evidence, not to be built on top of assumptions. In Advertima, we’re already experiencing this transition, and it’s genuinely exciting to see verified measurement taking hold in the physical world.”
What’s the hardest thing to explain to customers?
“In general we believe our value proposition is pretty clear. What made Online Retail Media successful in the first place, linked measurements for performance evaluations, is brought into physical stores. Where it takes a bit more explaining, mostly in the first conversation, is that in-store is a one-to-some or one-to-many environment, not one-to-one like online. A screen doesn’t face a single shopper every ten seconds, it often faces several people with different intents at the same moment. That’s precisely why this required more sophisticated technology to solve than online targeting ever did, and it’s exactly the layer we built. What’s genuinely rewarding is seeing customers connect that explanation directly to the performance numbers in front of them, the concept and the results reinforcing each other.”
Imagine it’s 2030. What will people laugh about when they look back at Retail Media in 2026?
“Seeing how fast retailers are adopting this technology and moving toward a verified impression-based model for in-store, I’m confident that by 2030 most of what’s considered standard practice today, screen uptime, loop plays, modeled reach, will look outdated. The market is still emerging. The early movers will acquire market share and budget share. Network reputation matters. So by 2030, the early adopters will have laid the foundation for sustainable long-term success, with the internal measurement frameworks, teams and processes built around it. “
Finish this sentence: Audience-based Activation matters because…
“…it’s the difference between hoping someone saw your advertisement and knowing they did. In-store advertising has existed for years, what’s been missing is the accountability that made Digital Media trustworthy in the first place, and Audience-based Activation finally brings that into the store, proving significant uplifts in both upper and lower funnel metrics”
Is there anything I didn’t ask that you think people should understand about this technology?
“Just that this technology is built on privacy-by-design. No facial recognition, no biometric data, nothing personally identifiable is ever accessed, processed or stored. All visual input is processed locally on our edge-AI and deleted within milliseconds; we never send images to the cloud, and we never keep what we don’t need. No photos. No videos. No personal information. Just metadata. Our technology is intentionally built to compensate for the fact that we do not use personal information but are still able to provide marketers with the insights they need.
We collect only anonymous audience segments. Our system groups people into segment-buckets based on broad traits like age range, gender, or group size, not individuals. What’s sent to the cloud is aggregated, stripped of identity, and impossible to trace back to anyone. No profiles. No tracking. No risk. Just insights. “
Want to explore how AI is reshaping retail and commerce media?
Download IAB Europe’s Guide to AI in Retail & Commerce Media to discover the latest industry perspectives, practical use cases and the Advertima case study featured in the publication.
