In this episode of Retail Reality Check, the host outlines how agentic AI is evolving from a web‑only novelty into the core operating layer for physical stores and back‑office functions. Citing IHL analyst Greg Buzek, the podcast emphasizes that AI must be anchored to a unified data foundation—catalog, POS, and order‑management systems—to avoid creating new silos. The host highlights the massive market opportunity, projecting store‑systems spend at $140.9 B in 2025 and $213 B by 2030, and notes Amazon’s $1 B forward‑deployed engineering effort to accelerate AI rollouts. Concrete front‑end examples include Albertsons’ AI‑driven product discovery, Lowe’s marketplace on associate devices, Sephora’s neuro‑divergent‑friendly quiet hours, and Home Depot’s Google Gemini voice agent. Back‑office transformations feature Kraft Heinz’s AI‑driven deductions management, Bol’s Foods’ AI ordering, and Chronodrive’s shrink‑reduction AI forecasting. A rapid “lightning round” surveys a global wave of AI installs across retailers such as Ahold Delhaize, ASDA, Sainsbury’s, Macy’s, Instacart, and more, covering personalization, delivery, loss‑prevention, and accessibility innovations. The episode closes with a provocative question about whether fragmented systems can support AI agents and an invitation to a July 23 webinar on AI‑accelerated self‑service.
Chapters
Welcome to this week's Retail Reality Check with a focus on AI as the new retail operating layer, plus a review of the latest tech installs brought to you by IHL Group. Before we get straight to the news, I have a quick invitation for you. Join us for our next webinar, Beyond the Counter: How AI is Accelerating Self-Service, Thursday, July 23rd at 11 AM Eastern Time. AI is reshaping the retail experience, and the kiosk market is absolutely at the forefront of that transformation. You will hear directly from industry end users as they share firsthand perspectives on how AI-powered kiosk solutions are delivering measurable value. It's going to be a great one, so definitely don't miss it. Section 1: AI as the new retail operating layer. Moving from the website to the sales floor. Okay, let's dive into this. You know, for the past year or so, the story of agentic AI, meaning AI that doesn't just answer your questions questions but actually takes actions on your behalf has pretty much lived online. But now, it has officially graduated from the website chatbot phase and walked right onto the physical sales floor and into the back office. To really set the stage here, let's look at a key insight from IHL's Greg Buzek. He points out that AI has totally moved from being just this novelty experiment into becoming the mandatory, foundational operating system for the store. We're seeing this play out in a massive way with retailers like Walmart. They're simultaneously rolling out generative AI search for shoppers, deploying AI-powered inventory tracking, and adding augmented reality to associate devices. They're combining these 3 huge initiatives to completely empower their frontline workforce. And following that lead, platforms like Salesforce with their new Agentforce release, along with SAP and Aptos, are bringing these AI agents right to the checkout lane and the fulfillment engine. But There is a huge catch here. An AI agent is literally only as good as the systems it connects to. Which brings us to the real elephant in the room: unification. If retailers want these new AI agents to actually solve fulfillment problems instead of just creating more headaches, they have got to unify their catalog, their point of sale, and their order management data foundation right now. Building a Storefront chatbot? Honestly, that was the easy part. Connecting the catalog, the POS, and the fulfillment engine underneath so they all talk to each other? That is a hard part, and it's exactly the area where most retailers are still running entirely separate, siloed projects. So why the huge push? Well, just look at the size of the prize for getting this right: $140.9 billion. Yep, you heard that right. IHL Group projects that store systems spend is going to reach an astonishing $140.9 billion in 2025, and that's projected to grow to $213 billion by 2030. That is a massive amount of money, making it the single largest line of business in all all of retail tech. This is exactly the prize that vendors are racing toward as they try to build out this new interconnected AI architecture. And to see just how seriously the big tech giants are taking this infrastructure build, we only need to look at Amazon. They are making a massive $1 billion investment to create what they call a forward deployed engineering organization within AWS. This is wild. They are literally embedding their own AI engineers directly into their clients' businesses and security teams. The whole goal is to press AI rollout times from taking several months down to just mere days. Now, we are already seeing some absolutely incredible front-end applications of this integrated approach. Take Albertsons, for example. They are integrating Criteo Sponsored Product Discovery directly into their Google Cloud Conversational AI search. This means they're able to hit shoppers exactly when they ask open-ended meal planning questions. So if you're a shopper and you ask the app for a taco recipe, relevant sponsor products just seamlessly pop up, right there in the conversation. It makes retail media feel totally organic and effortless. But hey, let's shift to the back office for a second, because AI agents are totally transforming operations there too. Look at how Kraft Heinz is attacking deductions management. They're using Genpact and Microsoft Azure AI agents to autonomously manage their accounts receivable. What these agents do is automatically pull deduction data directly from customer portals and carrier systems, and then they digitally check and match up those debit memos against proof of delivery. Then, the agents auto-categorize the claims. By doing this, these specialized agents are drastically reducing financial leakage. They're catching all these invalid deductions that manual human oversight might just accidentally miss. Plus, they are resolving these disputes up to 20% faster. It's just a perfect example of how AI can step in and solve a really complex, fragmented workflow. Section 2: Lightning Round. Global tech installs. Let's move to and see how this builds. It's time to shift gears for a breathless, rapid-fire look at a ton of global tech installs happening right now. Buckle up. First up, Lowe's just put its third-party MyRedVest marketplace directly onto associate devices across 1,700 stores. This allows staff to sell extended online inventory right there in the aisle, which definitely echoes a very similar QR code strategy that Walmart successfully tested previously. Over in the beauty space, Sephora is officially rolling out quiet hours across its global footprint. They're adjusting store lighting and turning down the music to create these wonderful low-sensory environments for neurodivergent shoppers. Say goodbye to those incredibly frustrating phone trees. Home Depot is swapping them out for a new Google Gemini AI voice agent that can understand exactly what a customer wants in under 10 seconds. In grocery news, Ahold Delhaize just scaled its personalization capabilities across nearly 2,000 stores. Meanwhile, the Midwestern chain Bol's Foods tapped Afresh AI to completely replace all their manual store ordering and handwritten production guides. And speaking of fresh food, the French drive-through grocery pioneer Chronodrive actually managed to cut their fresh produce shrink in under a single month! They did it by deploying Relux's AI forecasting system. Across the pond in the UK, ASDA is making some huge moves. They expanded their Deliveroo tie-up to over 850 stores for rapid delivery, and they just became the very first UK retailer to deploy the Amazon Retail Ads service. The major delivery platforms are also expanding way beyond restaurant food. DoorDash is now delivering directly from Autoparts.com, while Uber Eats just added major names like Kiehl's, FedEx Office, and Academy Sports to their app. On the loss prevention front, Sainsbury's is rolling out FaceWatch AI facial recognition to 150 stores. Their early tests showed it actually stopped 90% of repeat offenders, though they are managing it carefully after an early instance of human error misidentification. Digital innovation is just flying right now! Macy's is finding some great early success with their Ask Macy's conversational AI assistant. At the same time, Instacart is currently piloting direct add-to-cart vertical video ads with brands like Hellmann's and Siete Foods. And wrapping up our lightning round with a really fantastic milestone for accessibility: McIver's Market is officially the first grocer to accept EBT/SNAP payments directly on the digital screens of Instacart's AI-powered Caper Carts. Whew! Okay, that was a lot of innovation happening all at once. But as we bring this explainer to a close, I want to leave you with one provocative question: if an AI agent walks onto your sales floor today, if it tries to check your inventory, modify an order, or answer a shopper's highly specific question, Are your fragmented, siloed systems even capable of answering it? Because the retailers who build that unified data foundation today are the absolute winners of tomorrow. Thank you so much for joining me for this explainer. Be sure to check out the full narrative, including over 50 news stories and deep-dive case studies, at www.ihlservices.com. See you next time!
