When the AI agent goes shopping, data quality makes all the difference

What must retailers and brands do to ensure they can continue to sell their products in an era when it’s not consumers, but AI agents, who click the “Buy” button? According to GS1, everything hinges on data quality.

A new narrative

Many retailers, suppliers, and online platforms are at a loss today: how do you stay relevant in a world that will soon be driven by agentic AI? How can you ensure that shoppers will continue to find your products when chatbots take over from traditional search engines? Or, taking it a step further: what will be the impact when AI agents start making purchasing decisions on behalf of consumers?

Business strategists Rien van den Bosch and IT Director Senne Van den Bergh of GS1 Belgium & Luxembourg observe that there’s a lot of FOMO— fear of missing out—among companies. Should they invest in GEO (generative engine optimization) in addition to SEO? GS1’s message is clear: get your data in order. With unique identification standards, you’ll be prepared for tomorrow’s world of agentic commerce. Google uses the GTIN (Global Trade Item Number, or globally unique identifier) as the key to recognizing a product and distinguishing between variants.

Toward 100% data quality

“In September, we attended a workshop in Zurich with Google, OpenAI, and Meta, among others. What stood out: even the major platforms are still figuring it all out themselves. Our message, therefore, is: stay calm—you’re not too late. First and foremost, take a few steps you should have taken long ago to ensure you’re discoverable—steps that many companies haven’t fully implemented. If you simply fine-tune your classic SEO, that alone will solve a lot of problems.”

GS1 standards help with this. “Without a unique identification code, for example, Google has a hard time distinguishing between product variants. That used to be less of a problem: as a human, you can interpret what’s being referred to and make a choice, but an AI agent doesn’t work that way. 90% data quality is no longer enough—we need to aim for 100%.” A key factor is the implementation of the so-called UCP protocol, an open-source standard specifically developed to enable autonomous shopping via AI. UCP serves as a common language between AI agents, apps, search engines, online stores, and payment providers. And it cannot function without accurate, structured data.

It’s becoming even more important

GS1 has long emphasized that data quality is immensely important. In the past, this was primarily with regard to logistics processes: the earlier in the supply chain everything is correct, the fewer problems arise later on. “But if your data quality isn’t right today, the result will be that you’ll fall by the wayside. SEO ensures you’re found; data quality ensures the agent selects the right product. So it has become even more important. Moreover, the unique identification code isn’t just important for being found online—it’s also essential for complying with future European legislation.”

There’s no need to panic, but sitting back and doing nothing isn’t an option either. At the Category Management Congress hosted by LD&Co and RetailDetail—one of the three pre-events at RetailDetail Night—on November 19 in Brussels, Rien van den Bosch from GS1 will explain how brands and retailers can best prepare for agentic commerce. YouGov, Bonduelle, Mars, and LD&Co will also be speaking that afternoon. Reserve your ticket using the button below.