Joe Hale, founder of Verde Digital, takes a look at how AI shopping is likely to affect retailers – really.
Until recently, AI shopping was treated as a future concept for UK retailers, once the technology had matured and consumer behaviour caught up. However, some recent developments in the US shed a little more light on where AI shopping is heading.
JD Sports’ announcement that it will allow US customers to search for and one-click purchase products directly through AI platforms, such as Microsoft Copilot, Google’s Gemini and ChatGPT, marks a clear change of pace. At the same time, Google and Shopify have launched the Universal Commerce Protocol (UCP), designed to standardise how AI accesses product data, pricing and availability.
The hope is that it will offer frictionless discovery, conversational commerce and personalisation at scale. The challenge for UK retailers is how to engage with AI search alongside SEO and ecommerce performance, which still underpins the majority of online revenue.
Having worked with retailers through multiple waves of “the next big thing” such as mobile-first shopping, marketplaces and social commerce, I’ve learned that the winners aren’t necessarily the first movers. They are the brands that understand which parts of their stack must evolve, and which parts must be protected. AI shopping is no different.
JD’s approach is notable because it treats AI as a transactional channel. By connecting LLMs directly to product information, pricing and inventory systems, with checkout handled by Stripe, JD is enabling customers to move from discovery to purchase in a single conversational flow, without visiting a website or app.
That model may be transferable to the UK market, however regulation, margins and consumer expectations differ. But it does indicate the direction of travel, showing how AI is moving from influencing shopping decisions to mediating them.
SEO is not dead
Every time a new way to search emerges, SEO and/or Google are declared dead. It never is; in fact, we are seeing some clients experience record organic revenue figures.
AI shopping will not eliminate search behaviour, but it will add a new way for users to shop. Shoppers using AI will ask questions instead of typing keywords. Instead of scanning results pages, they will receive recommendations. Underneath that conversational layer, however, retrieval still depends on familiar signals such as structured data, authority, relevance and trust.
In other words, the same fundamentals, filtered through an additional layer.
The danger for brands is assuming that AI shopping assistants will simply “know” their catalogue, but this is not the reality. AI relies heavily on how clearly products are defined and connected through feed consistency, entity relationships, schema markup, real-time availability, and historical performance data.
From an SEO perspective, this places renewed importance on basics that are often neglected. Product titles need to be precise and consistent across channels. Attributes such as size, colour and category must be cleanly structured, not buried in free-text descriptions. Internal linking and taxonomy still matter because they help machines understand how products relate to one another.
Brands that weaken these foundations in favour of chasing AI integrations risk becoming invisible inside AI shopping environments. Those that maintain technical hygiene, invest in content that reflects genuine customer intent, and govern their product data carefully will be far more resilient, regardless of which protocol or platform dominates.
AI assistants still rely on structured data, clear product entities, authoritative content and trust signals to surface recommendations. What disappears is the visible ranking. Instead of appearing on page one of a search engine, a brand may or may not be mentioned inside a conversational response.
For UK retailers, this creates a subtle but serious risk. You may not see dramatic drops in traffic or rankings. You may simply stop being part of the conversation.
Protecting SEO fundamentals is therefore a way of ensuring continued visibility in AI-led discovery, even as the interface changes.
The brands that are already winning in AI search are the ones who have put the work in from a brand perspective, which is complimented by SEO fundamentals to support from both a branded and a non-brand lens.
Ecommerce performance still pays the bills
There is a danger that agentic commerce becomes a distraction from more immediate priorities. A rush to be AI-ready to appease senior stakeholders is actually going to result in brands and marketers not focusing on what is actually going to drive immediate revenue.
Many UK ecommerce sites still struggle with issues that directly suppress conversion, such as slow performance, ineffective internal search, weak filtering and inconsistent pricing logic. AI does not solve these problems, but it will expose them.
If an AI assistant surfaces inaccurate stock levels or misleading prices, trust is lost quickly, and in the UK, that trust is hard to regain. JD’s AI initiative sits on top of a substantial investment in digital infrastructure, including a major re-platforming of its US ecommerce operations.
For most UK retailers, readiness for AI shopping depends less on innovation and more on getting the basics right. Product data needs to be accurate and well governed. Site performance must meet rising expectations. Analytics should clearly show how users move from discovery to conversion, so behaviour driven by AI can be understood rather than guessed at. Without these foundations, AI becomes a layer of complexity rather than a source of value.
Expect fragmentation before consolidation
One assumption underpinning much of the current debate is that we are heading towards a single, dominant standard for AI commerce, but I don’t think this will be the case.
Google’s Universal Commerce Protocol is not the only framework retailers have to contend with. Alongside platform-led standards, we are already seeing more informal protocols emerge through commerce engines, payment providers and direct integrations between retailers and AI models. JD Sports’ use of Commercetools and Stripe is one example of a retailer-controlled approach, while platforms such as Microsoft are shaping their own conventions through Copilot and Bing. For UK brands, this points to a period of coexistence rather than consolidation, where visibility depends less on backing a single standard and more on having data, SEO and ecommerce systems that can adapt to several.
We are therefore far more likely to see a fragmented period, with platform-specific implementations, retailer-owned systems and hybrid approaches competing for several years. Each will optimise for different incentives, whether advertising revenue, conversion efficiency, data ownership or loyalty.
For brands, this means flexibility matters more than allegiance. The goal should not be to pick a winning protocol, but to build systems and teams that can adapt without re-platforming every time the landscape shifts.
That requires modular data architecture, strong first-party understanding of customers, and a willingness to resist outsourcing strategic decisions to vendors offering certainty in an uncertain space.
Holding on to what matters
When new technology arrives, the pressure to move quickly is intense. Boards want reassurance, investors want a clear narrative and teams want direction.
My advice to ecommerce brands navigating AI shopping is to move strategically. Protect the fundamentals that drive revenue today, while experimenting in ways that do not compromise them. Do not get caught up on the next wave of SEO, by forgetting the fundamentals which we know continue to drive large percentages of total revenue.
AI will change how people shop. Protocols such as UCP may well become part of the infrastructure, but visibility and profitability will still depend on remembering that technology shapes demand rather than creating it. If you still understand your customers better than anyone else and your SEO and ecommerce foundations reflect that understanding, you are already better prepared for AI shopping than you might think.
Stay informed
Our editor carefully curates two newsletters a week filled with up-to-date news, analysis and research. Click here to subscribe to the FREE newsletter sent straight to your inbox. Why not follow us on LinkedIn to receive the latest updates on our research and analysis?
