Most ecommerce brands have no idea what AI says about them. They've checked their Google rankings. They monitor their Trustpilot score. They know what shows up when you search their brand name.
But they haven't asked ChatGPT. They haven't searched Perplexity. They don't know if Google AI Overviews is recommending them or their competitor when a potential customer searches their category.
In 2026, this is a significant blind spot. AI assistants are part of how your customers research and decide, and what those assistants say about your brand shapes purchase decisions you never see in your analytics.
What AI actually says (and doesn't say)
The range of AI responses to brand queries is wider than most merchants expect.
At one end: AI knows your brand well, mentions you consistently in category queries, and makes positive, specific recommendations. "For sustainable activewear, [Brand] is consistently well-reviewed for quality and fit. Their returns policy is flexible, and customers regularly highlight the durability compared to mainstream alternatives."
At the other end: AI has never heard of you, or conflates you with a different brand, or repeats outdated information from before a product reformulation or relaunch.
In the middle, which is where most brands sit, AI mentions you inconsistently, hedges its recommendations, or includes you in a list without differentiation. "Some options in this category include [Brand A], [Brand B], and [Brand C]." Mentioned, but not recommended.
The gap between "mentioned" and "recommended" is significant. A mention puts your brand in front of a potential customer. A recommendation closes the gap.
The five AI responses you might get
Recommended confidently. AI names you specifically, describes what you're good for, and endorses you. This happens when you have strong editorial coverage, consistent positive reviews, and clear positioning in AI's understanding of your category.
Mentioned in a list. You appear alongside two or four other brands without differentiation. AI knows you exist but doesn't have enough to distinguish you. Common for brands with some third-party coverage but no clear editorial endorsement.
Not mentioned. AI doesn't include you even for queries directly relevant to your category. Either you're absent from the sources AI draws on, or you're present but not prominent enough to make the shortlist.
Incorrectly described. AI says something about you that isn't accurate: wrong product range, outdated pricing, wrong country of origin. This happens when the most-indexed information about your brand is old or when third-party sources contain errors.
Mixed sentiment. AI synthesizes both positive and negative signals from across your review and coverage footprint. If your Trustpilot is strong but there's a Reddit thread with a well-upvoted negative experience, AI may reflect both.
Why this matters beyond brand awareness
Your customers are making decisions based on what AI says. A potential customer asks Perplexity for the best option in your category. If AI gives a confident recommendation for your competitor and a hedge about you ("some customers report [issue]") they're likely going to your competitor.
AI can surface old problems. A product recall from four years ago, a customer service issue that generated press coverage, negative Reddit threads. AI doesn't distinguish between recent and historical information the way a human would. Older negative information can resurface in AI responses even if you've long since addressed the underlying issue.
AI shapes the comparison frame. When someone asks "Is [your brand] better than [competitor]?" AI's answer sets the frame for how customers think about the comparison. If AI consistently positions your competitor as the premium choice and you as the budget option, even if that's not accurate, that perception becomes self-reinforcing.
AI's opinion of you is largely outside your direct control. You can't edit what ChatGPT says about you. You can influence it over time, by building the coverage, reviews, and content that shape what AI learns. But you can't change it quickly, and you can't do it without first knowing what it currently says.
How to find out what AI says about your brand
The manual version: run 15–20 queries across ChatGPT, Perplexity, Google AI Overviews, Claude, and Gemini. Cover discovery queries ("best [product type]"), comparison queries ("[your brand] vs [competitor]"), and trust queries ("is [your brand] worth it?"). Record the results. This takes about two hours and gives you a snapshot.
The limitations: it's a small sample, you'd need to repeat it monthly to track changes, and you won't get a quantified score, just qualitative impressions.
SeenByAI automates this at scale: hundreds of queries per run, with quantified scores for AI Search and AI Knowledge across the three stages of the customer journey (Discover, Consider, Trust). The result is a clear picture of where AI currently stands on your brand, how you compare to competitors, and what actions to prioritize.
What to do when you find something concerning
Outdated information: The fix is fresh, accurate content on your own site, combined with updated entries on third-party sources (review sites, directory listings, Wikipedia if applicable). AI Search tools pick up fresh content quickly. AI Knowledge models are slower. They need the correction to make it into the next training data snapshot.
Missing from category queries: This is a coverage and review problem. Which review sites cover your category? Are you included? Which buying guides appear when you search "best [your product type]"? These are your targets.
Negative review sentiment: You can't delete external reviews, but you can dilute them by generating more positive recent ones, and you can respond to them thoughtfully. AI reads your responses. Address the underlying issue if it's real. AI will pick up on whether the problem that generated the negative reviews still exists.
Incorrect information: Find the source of the incorrect information and, if possible, update it. Then create accurate content on your own site and in authoritative sources that AI is likely to index.
The first step is always knowing where you stand and having a plan to fix what you find. Most brands haven't looked.
SeenByAI finds where competitors are beating you in AI, and get a prioritized plan to close the gap. Get started free.
Written by Stu Miller, Founder of SeenByAI and CEO & Co-founder of Smart Insights. Stu has spent 16 years helping ecommerce businesses grow their digital marketing, and built SeenByAI after experiencing the AI visibility problem first-hand running his own Shopify store.
Related reading: