When AI agents shop for us, how will loyalty programs keep up?
Is the era of "browsing" being replaced by the era of "delegating?"
As AI tools move from effectively being smarter search engines (chatbots), to autonomously executing transactions on behalf of a person (AI agents), the online marketing playbook is being rewritten. Add to this Google’s recent announcement of the Universal Commerce Protocol, or UCP, an open standard co-developed with giants like Shopify, Walmart, and Target which signaled that the industry has officially moved past agentic commerce just being theoretical, and becoming much more tangible.
[READ MORE: Google launches new agentic commerce standard]
For retailers, this shift means loyalty is becoming a way to differentiate products and gain favor with agents. This also means when a consumer asks a shopping agent to "buy a durable rug for a high-traffic room," the agent evaluates various attributes including prices and product specifications, while also assessing the loyalty rewards or discounts a given product might have.
In effect, product data will play an even more important role going forward, especially because Google’s UCP is reportedly expanding the number of product data attributes available for merchants to upload by as many as 60 new fields. All of these new data attributes will help agents determine the suitability of each product for a given customer’s specific needs. Without “enough” data, a product may not even be discovered by an agent, much less matched-up to a shopper’s request.
With this in mind, brands must master three areas of focus to ensure they stay relevant in the next evolution of retail: agentic commerce.
Discoverability: Feed the machine, or be forgotten
When agents do the shopping, a retailer’s product data is the new storefront. If an AI agent cannot read and understand information about the products a retailer offers, it simply cannot buy those products. It’s no secret that many retailers are currently drowning in data debt, treating product feeds as checkboxes to “just get done” rather than as a strategic asset that can be a lever to markedly affect revenue.
The UCP and new Merchant Center tools allow retailers to provide dozens of rich data attributes, including compatibility information, including complementary and compatible accessories, substitutions, Q&As, and product themes. This ensures that when an agent like Gemini evaluates a complex request—such as planning a child's birthday party—it can accurately surface and purchase consumers’ items based on "machine-readable" integrity.
Taken together, this means visibility to the AI agents with high-quality, complete product attributes is now table stakes for retail survival.
Alignment: Surfacing loyalty incentives to the agent
When an agent acts on behalf of a human, loyalty programs must be integrated directly into the flow. One of the other tools retailers and loyalty providers can leverage is a Model Context Protocol (MCP) server. MCP is a tool which provides additional data directly to chatbots and AI agents. With this in mind, a retailer or loyalty provider could include loyalty program rewards or coupon offers that AI agents can see and incorporate into the purchase. This way, a purchase from a store with available discounts or incentives can be applied, adding more value for shoppers.
Notably, Google mentions that they’ll be working with retailers in the coming months to allow the application of loyalty rewards during checkout as part of the UCP features, further proving the need for loyalty to shift from a reward given after a purchase to an incentive recognized by the algorithm during an agent’s “thought” process.
Trust: Proving the system works for the shopper
One of the greatest barriers to mass adoption of agentic commerce is consumers ceding control. To maintain loyalty, retailers must prove that these autonomous systems are secure and still focused on their customers. Luckily, it appears that UCP addresses this by ensuring retailers remain the seller of record, maintaining control over business logic, fulfillment, and returns while defending the retailer from prospective fraud by using secure payment methods like Google Wallet and PayPal.
Google also reports that businesses can utilize Business Agents, which will allow consumers to chat with an agent on behalf of the brand. A blog post notes that in the near future, brands who deploy business agents will be able to train the agent based on proprietary data, access new customer insights, provide complementary offers for related products and enable purchases – including agentic checkout. These moves will also serve to build trust for shopping agents.
Is the "death of clicks" looming? Some 2026 predictions point to a massive shift where purchases happen directly on AI-enabled search surfaces or with agents rather than retailer websites. This means retailers who treat their product data as customer-facing content and integrate their loyalty programs into the agentic layer will be clear winners that stay visible when algorithms finally take the wheel.
Tristan Barnum, chief marketing officer and head of AI innovation, Wildfire Systems.
