The Invisible Shelf

There was a time when visibility was mostly a physical problem. If your product sat on the bottom shelf in Tesco, fewer people saw it. If your Amazon PDP was weak, you lost ranking and traffic. If your title, bullets, imagery and reviews did not do the job, you became harder to find and easier to ignore. That same problem now exists in AI. The difference is that the shelf is invisible. Consumers are increasingly using platforms like ChatGPT, Gemini and other AI interfaces to research products, compare options and build shortlists before they ever click through to a brand or retailer. Adobe reported that traffic to U.S. retail websites from generative AI sources rose 1,200% between July 2024 and February 2025. Adobe also found that 39% of U.S. consumers had already used generative AI for online shopping and 53% planned to do so in 2025. That matters a lot. If your brand is not discoverable inside those platforms, a growing number of shoppers will never meaningfully see you. Not in the shortlist. Not in the recommendation set. Not in the answer. You are absent before the click even happens. Discovery is moving upstream For years, brands have focused heavily on the point of conversion. AI changes where the funnel begins. Instead of typing a short keyword into a search bar, consumers can now ask full questions. What is the best electrolyte drink for runners? Which sunscreen works for sensitive skin and sport? What are the best protein snacks for kids? The winner is no longer just the brand with the biggest budget. It is the brand that is easiest for AI systems to understand, trust and retrieve. That means your brand needs to be legible to machines before it can be chosen by humans. Adobe’s data suggests this traffic is not low quality either. Visitors arriving from generative AI sources in retail browsed 12% more pages per visit and had a 23% lower bounce rate than other traffic. The new shelf is made of answers. In physical retail, shelf position influences whether a shopper notices you. On Amazon, search rank influences whether a shopper clicks you. In AI platforms, answer inclusion influences whether a shopper knows you exist at all. If an AI platform recommends three brands and you are not one of them, you do not just lose a click. You lose the consideration moment itself. That is why we call it the invisible shelf. It exists before the PDP. It shapes the shortlist. And most brands are not managing it properly yet. This is the same problem in a new interface. None of this should feel alien to eCommerce teams. An unoptimised Amazon PDP is invisible shelf space. Poor retail execution in Tesco is invisible shelf space. Weak product data is invisible shelf space. Buried AI discoverability is invisible shelf space too. The principle is unchanged: if shoppers cannot find you, the quality of your product barely matters. Adobe’s UK reporting points in the same direction. It found that 35% of UK consumers surveyed had already used GenAI tools to help with shopping and 47% planned to do so that year. Why some brands show up and others do not AI systems rely on signals. - Clear product attributes. - Consistent naming. - Useful PDP copy. - Structured information. - Relevant reviews. - Trusted external sources. - Strong category associations. The brands that show up are usually the ones that have done the hard work of being understandable. If AI platforms become a standard part of how people discover products, then brands that are not represented properly will lose inclusion at the moment shortlists are formed. And that can compound. Just like strong Amazon rankings compound. Just like physical shelf dominance compounds. The invisible shelf compounds too. A growing share of discovery is now happening inside AI platforms. If your brand is not visible there, then for a growing number of shoppers, you effectively do not exist. That is the real message in Adobe’s research. The future of discovery will not just be won on the shelf you can see. It will be won on the one you cannot.