In-Store Footprints: The Omnichannel Expansion
Online retail media had its gold rush. But that digital frenzy is cooling down. Broad industry data points to a sharp growth deceleration, with spending growth set to drop from 25.1% in 2024 to a much flatter 15.6% in 2025. Advertisers are hitting bid exhaustion on typical web banners. CPMs are topping out. This makes offline channels the next logical target. Physical stores aren't just brick-and-mortar shells anymore. They're becoming active, high-intent media grids. Forbes and RetailNext project the sector will hit $231 billion by 2030, driven largely by turning the store floor into an ad platform.
Advertisers want digital-grade metrics in physical spaces. No more guessing. They're tired of paying for static endcap cardboard and hoping for the best. Brands want to know who walked by, how long they stood there, and if they actually bought the item. Translating web impressions into store realities means treating aisles as digital screens. Retailers are building new inventory out of thin air. They're mapping physical paths to create structured, bids-friendly ad space. The shift isn't just about sticking displays on walls; it's about building a responsive, measurable infrastructure where physical traffic is valued exactly like web traffic.
Think about the sheer scale. Most purchases still happen offline. That is where the volume is. If a retail media network only sits on a website, it's missing the boat. Brands want an omnichannel look. They want the display banner on the phone to connect with the digital screen on the endcap. That requires a unified network. It's a massive coordinate system. The store becomes a physical extension of the web browser, but with real-world carts and physical feet.
Transaction Data and Privacy-Safe Attribution
The receipt is a retailer's primary asset. It's the ultimate paper trail. It doesn't use tracking cookies or guess based on algorithms. It is direct proof of purchase. Today, modern networks connect their programmatic buying tools straight to these checkout registers. Take Kroger's work with The Trade Desk as an example. They compare exposed shoppers directly against matched control groups who didn't see the ads. It's clean, database-matched attribution, not shaky panel projections. When you own the transaction data and own the physical space, brands expect you to prove the lift. However, scaling this volume of checkout logs can create massive data ingestion bottlenecks. Building matching frameworks without crashing requires a robust, performant storage platform, which is why scaling AI data infrastructure is critical to successfully operating enterprise-level measurement databases.
But tying store movement to a specific receipt is a legal minefield. Privacy compliance isn't optional. GDPR and CCPA rules are tight, and a slip-up burns cash fast. A major GDPR breach can cost up to 4% of a company's global revenue. That's a massive hit. To avoid this, networks use privacy clean rooms. They clean and group data so individual identities are never exposed. Systems track anonymized traffic flows and group buying habits rather than tracing specific customers. Shoppers get to keep their privacy. Brands get the conversion validation they need. Retailers protect their margins. It's a delicate balance, but it's the only way to scale in-store ads without getting hit by regulatory fines. Indeed, database security is paramount when storing and using first-party consumer records in clean rooms. Compromises in nearby client database platforms can be devastating (as described, for instance, in our report on the Nottingham student records breach that exposed personal and financial databases).
Advertisers are becoming hyper-sensitive to compliance. They'll walk away from networks that can't guarantee data safety. Anonymized data clean rooms allow brands to upload their own first-party files, mix them with the retailer's checkout files, and check performance securely. It proves that the in-store display drove the buy without revealing who bought it. You get pure performance metrics without the compliance liability.
Smart Telemetry: Mapping The Shopper Journey
Old-school store tracking was elementary. You stuck a clicking counter on the front door and hoped for the best. Sensor technology changed that. Now, smart telemetry tools use IoT sensors, Wi-Fi logs, and smart shelves to map how people walk through a store in real-time. If a shopper lingers in the chip aisle, the system logs the entry and the dwell time. This map changes how layout designers plan store formats. This telemetry is a subset of physical-world interaction tracking, which presents data-gathering hurdles similar to the issues faced in OpenAI's robotics relaunch, where acquiring physical-world interaction data is crucial to modeling real-world spatial environments.
It turns out that location is everything, but not in the way most people think. High-speed zones like entrances are ad dead zones. People move too fast to notice screens. The real value is in deceleration zones. These are spots where shoppers wait or compare items. Think of deli counters, pharmacy queues, and comparison shelves. That's where you put screens.
This telemetry also enables dynamic content. Thanks to real-time occupancy and environmental sensors, displays can swap creatives instantly. If occupancy spikes near the pharmacy on a cold, rainy day, the screens pivot from energy drinks to cold remedies. It's digital-grade optimization, but built for the messy physical store floor. Brands pay premium rates because their ads match the immediate context of the room.
We are talking about platforms like RetailNext's Aurora sensors. These systems monitor traffic without capturing faces. They detect heat signatures and paths. They prove capture rates. If fifty people pass an interactive display and six spend more than thirty seconds looking at it, the system logs it. That's real, verified dwell time. This lets retailers charge brands based on actual eyeballs, not general assumptions about store traffic.
Self-Service Workflows and Asset Automation
Launching a physical ad campaign used to be a massive chore. In the old model, brands spent weeks negotiating with a retailer's internal creative teams. It was slow and expensive. That's shifting. Retail media networks are deploying self-service workflows and creative automation tools like Adobe GenStudio. Now, brand teams upload their product shots, choose templates, and generate compliant, store-ready digital formats in minutes. It removes the creative bottleneck entirely.
But making the ad is only half the battle. You have to prove it worked. Legacy networks used to send performance slide decks weeks after the campaign ended. If an ad fell flat on its face, you didn't know until the money was spent. That doesn't work anymore. Real-time reporting access is a critical demand. Studies show that 40% of organizations would speed up and increase their retail media spend if they had better, faster access to insights. When brands see daily performance updates on physical screens, they buy with confidence. Automated workflows take the administrative work away from the retailer and give advertisers the quick verification they need to keep writing checks.
It reduces simple friction. If a marketing manager has to spend half their day emailing back and forth to approve a simple graphic, they'll move their budget somewhere easier. Automation builds scale. It lets small, mid-tier, and massive programmatic buyers run physical promotions with the same ease as a Google Search ad. Self-service gives power back to the brand, and it lets the retailer focus on operations.
Beyond ROAS: The Shift to True Incrementality
For years, Return on Ad Spend was the default scoreboard. But ROAS is a deeply flawed metric. Standard ROAS calculations get easily distorted by long attribution windows, halo effects, and rules that credit an entire grocery basket to a single clicked ad. If a consumer sees an in-store digital display for shampoo, goes to checkout, and buys eighty dollars of groceries, standard attribution might credit the shampoo ad for the whole eighty-dollar purchase. That is not marketing. That is bad math.
Brands are catching on to these tricks. More than one-third of advertisers now actively slow or reduce their spending when a retailer cannot prove actual sales lift (according to eMarketer and Osmos data). They want to know if the ad drove a purchase that wouldn't have occurred otherwise. That's incrementality.
To safeguard their budgets, networks are adopting strict incrementality testing. They run onsite A/B splits—showing ads to a test group while keeping a control group organic—or use geo-matched regional store testing to isolate campaign-driven sales. This isolates the exact lift. It shows causality. As highlighted in the Wall Street Journal's special feature on retail media results, the industry has to move beyond simple vanity metrics to prove what actually works in physical spaces. Incrementality is the new baseline. RMNs that fail to build this testing infrastructure will watch their budgets shift to networks that do.
Look at Skai's 2025 State of Retail Media report. It shows that advertisers are no longer asking if their ads were seen. They're demanding to know what sales happened specifically because of that placement. If you can't isolate the lift, the budget plateaus. By setting up rigorous control stores that do not run the promotions and comparing their sales to active test stores, retailers can show the precise dollar lift. It's the only metric that matters when budgets get pinched.