An effective AI search strategy for ecommerce is the process of making a business, its products, and its expertise understood and trusted as a citable entity. It is not an evolution of traditional SEO. While it uses the same web infrastructure, success depends less on ranking for keywords and more on being included and recommended within an AI-generated answer. This requires a fundamental shift in approach, from optimising for a list of links to optimising for direct synthesis by a machine.
The claim: AI visibility is a distribution problem, not a ranking problem
Our position is that treating AI-generated answers as a simple evolution of a Google search results page is a strategic error. For twenty years, search engine optimisation has been a game of relative ranking. The goal was to appear higher than competitors on a long list of blue links, capturing a percentage of clicks.
AI answer engines change this model entirely. They synthesise information from multiple sources to provide a single, conversational answer. Instead of a list of ten options, a user might get a paragraph recommending two or three specific products or brands. The commercial stakes are therefore higher. It is a binary outcome: either you are mentioned, or you are invisible. This is not a ranking problem; it is a distribution and access problem.
The prevailing view and why it persists
Most businesses we speak with see this as a new task for their existing SEO team or agency. The logic is understandable. The inputs for large language models are, primarily, the publicly accessible web. The people who understand how to structure a website and its content for a search engine to crawl and understand are SEO practitioners. It seems like a natural extension of their remit.
This approach is convenient. It avoids creating new roles or reallocating significant budget. It places a complex, uncertain new channel into a familiar box. But this convenience comes at a cost: it mistakes the foundational requirements for the complete strategy. It assumes the old tactics will work for a new type of machine intelligence that is not designed to simply rank links.
The evidence from client work that contradicts it
Our research into the digital maturity of the UK’s leading home and interiors brands highlights a stark disconnect. In our latest study, The UK Home & Interiors Digital Growth Benchmark 2026, we scored 75 retailers against 54 criteria across six pillars of growth. The results are telling.
47.2 points
The gap between traditional SEO performance and AI readiness
Source: This Is Better, UK Home & Interiors Digital Growth Benchmark 2026 (n=75)
This gap is not isolated to laggards. We found that for every peer group we analysed—Omnichannel Retailers, Multi-category Pureplays, and Specialist/DTC brands—AI Discovery was their single weakest pillar of performance. Even brands scoring over 95% overall, like John Lewis and IKEA, are significantly less prepared for this shift than their traditional search authority would suggest. If strong SEO automatically led to strong AI visibility, this gap would not exist. For more on the scoring, see our benchmark methodology.
The strongest counter-argument, taken seriously
The most serious counter-argument is that you cannot achieve AI visibility without a strong SEO foundation. This is absolutely correct. An AI model cannot cite what it cannot find or understand. A technically broken, slow, or poorly-structured website is invisible to Google’s crawlers and, by extension, the models they inform. Clear, authoritative content and strong domain authority are the price of entry.
But this is where we must name the trade-off. Believing that a strong SEO foundation is sufficient is a passive strategy. It assumes your existing investment will automatically deliver results in this new context. The evidence shows it will not. The trade-off is this: you must either accept being invisible in a growing number of user journeys, or you must actively allocate resources and strategic focus to a new discipline that sits alongside SEO. It requires a different mindset and different work.
What we would do differently, concretely
Instead of simply tasking the SEO team with 'doing AI', we would reframe the objective. The goal is to make your business and its products understood and trusted as a citable entity. This is how we would approach it:
- Shift from keywords to entities: Focus on being unambiguously understood by machines for something. Does the web's knowledge graph know you are the UK's leading specialist in handmade leather sofas? This involves structured data, consistent brand messaging, and building authority around your core concepts, not just keywords.
- Audit content for citability: Review your most important product, category, and advice pages. Are facts, specifications, and unique selling points presented in a way a machine can easily parse and quote? Use clear headings, tables, and lists. Answer customer questions directly on the page.
- Treat your business as a dataset: Ensure absolute consistency in your core data (name, address, specialisms, founder info) across your own site, product feeds, Google Business Profile, Wikipedia, and key industry websites. Machines build trust from corroboration across multiple authoritative sources.
- Measure what matters: The metric is no longer just keyword rank. The new key performance indicator is the frequency and quality of your brand, product, and expert mentions in AI answers to high-intent commercial queries. This requires new tools and a new reporting focus.
This is a strategic project, not a technical task. It combines elements of data management, digital PR, and content strategy. If this describes a challenge you are facing, the first step is to talk with a senior consultant to map the problem.