For a retail brand, AI visibility is whether a shopper's AI assistant names your store at all before that shopper has even decided where to buy, in a physical store, on a marketplace, or on your own site. That is a wider problem than ranking an ecommerce product page, because the research IBM and the National Retail Federation published in January 2026 shows the AI conversation now happens earlier than the channel decision, not after it. Get left out of that conversation and a retailer never gets the chance to compete on price, service or loyalty, because the shopper never hears the name.
Source: IBM Institute for Business Value and NRF study of 18,000+ consumers in 23 countries, published January 7, 2026, linked in the article.
How many shoppers are actually using AI to research purchases?
A large and specific share, according to the only study here with a disclosed, large sample. The IBM Institute for Business Value's study with NRF, published January 7, 2026, surveyed more than 18,000 consumers across 23 countries in the third quarter of 2025, alongside 200 senior retail, consumer goods and ecommerce executives. It found that 45% of consumers now use AI for help somewhere in their buying journey: 41% to research products, 33% to interpret reviews, and 31% to hunt for deals. The same study found 72% of consumers still shop in physical stores, a figure the report does not tie to AI use, but one that matters for a retailer: the AI research phase and the in-store purchase are not mutually exclusive. A shopper can ask an assistant what to buy and then walk into a mall to buy it.
Is this the same problem as AI visibility for ecommerce?
Related, but not identical. AI visibility for ecommerce brands is mostly about what happens when an assistant is asked to name or compare products for an online purchase, where structured product data, a clean catalog, and AI-referred traffic to the storefront itself are the direct levers. Retail visibility is broader: a shopper researching where to buy a stand mixer, a pair of running shoes, or a sofa is often not asking "which website," they are asking "which brand or which store," and the AI assistant's answer shapes a decision that might end at a physical register, a marketplace listing, or a brand's own site, in roughly that order of likelihood for many categories. A retailer that only optimizes its own ecommerce pages is solving half the problem; the other half is making sure the brand itself, not just its online store, is what gets named.
What are retail executives actually doing about this?
Less than the shopper-side numbers suggest they should be. In the same IBM-NRF study, 54% of surveyed brand and retail executives reported persistent challenges with data consistency across channels and systems, and 51% cited limited in-house AI expertise. Caroline Reppert, NRF's Senior Director of AI and Technology Policy, put the underlying shift plainly: AI is changing how consumers shop, and every aspect throughout the shopping journey. Matthieu Houle, CIO of ALDO Group, described the practical effect from the retailer's side: AI is turning shopping into a trusted conversation, much more than a search. Both read the same way for a PR or marketing team: the brand no longer controls the first impression a shopper forms, an assistant does, and a retailer with inconsistent product data or no public-facing content for that assistant to draw on is handing the first impression to whichever competitor has cleaner data.
What actually makes an assistant recommend one retailer over another?
Mostly the same signal that drives AI citation in every other category: independent, verifiable third-party content, not a brand's own marketing copy. An analysis of more than 800,000 AI responses found a stark gap between brands with no verified review presence and brands with even a minimal one, and that pattern holds for retail specifically. A product review on an independent site, a "best of" roundup from a publication an AI system trusts, or a reporter's write-up of a store opening all give an assistant something concrete to repeat. Earned coverage already accounts for the large majority of what AI systems cite across categories, and retail is not an exception to that pattern just because a lot of retail marketing budget has historically gone to paid placement and in-store signage instead.
Does local search and store-listing data still matter?
Yes, and arguably more than before, because an assistant answering a "where can I buy X near me" query is pulling from the same structured local data Google Search has used for years: a Google Business Profile, consistent store hours and addresses, and location-specific inventory signals where a retailer provides them. A chain with inconsistent listings across locations, duplicate or outdated store pages, or no structured data describing what a specific branch actually carries gives an assistant nothing reliable to pull from for a local query, and an assistant with nothing reliable tends to default to whichever competitor's data is clean.
What should a retail brand actually do first?
Three things, roughly in this order. First, fix the data consistency problem the IBM-NRF executive survey flagged: a product catalog, pricing, and store information that disagree across channels and systems is the single most common reason an AI assistant gives a wrong or outdated answer about a retailer. Second, build or claim the review and roundup presence that assistants actually cite, the same unglamorous, ongoing work any brand doing AI-visibility work has to do, rather than treating it as a one-time project. Third, treat earned coverage, a trade publication piece, a reporter's product review, a local business story, as infrastructure for AI visibility rather than a separate marketing line item, since that is the content an assistant is actually drawing its answers from. 5WPR's category-specific index for home improvement retailers shows exactly this pattern at work: the two biggest chains barely get cited from their own domains, while review sites and Wikipedia carry most of the weight.
What happens when an AI assistant gets a retailer's information wrong?
More often than most retail teams realize, and there is no fast way to correct a specific chatbot's specific answer. The fix runs the same way it does in every other AI-visibility category: there is no customer-support ticket a retailer can file with an AI company to edit a single response, so the only durable lever is strengthening the accurate public record an assistant is likely to draw its next answer from, current store data, an accurate product catalog, and real third-party coverage that states the facts correctly. A retailer that waits for an error to go viral before addressing its underlying data and coverage gaps is already behind; the data and coverage work has to exist before the wrong answer does.
The direct answer, for anyone who read this far
Forty-five percent of shoppers are already using AI somewhere in their buying journey, and seventy-two percent are still walking into physical stores, which means the AI research conversation now runs ahead of, not instead of, the channel decision retailers have spent decades optimizing for. The brands that show up in that conversation are the ones with consistent product and store data across channels, a real review and earned-coverage presence an assistant can cite, and a public record accurate enough that a wrong AI answer is the exception rather than the norm. If you want a clear read on how AI assistants currently describe your stores or your products, get in touch with our team.



