AI shopping visibility

What separates it from brand visibility?

The unit of measurement. Brand visibility asks whether an assistant names you in a written answer, and its atom is the mention: a sentence recommending a company. Shopping visibility asks whether an assistant shows your product in a carousel, and its atom is the card: an image, a price, a seller, sometimes a checkout button. A brand can be recommended warmly in prose while its products never appear in a single carousel, and a product can appear in carousels for a merchant nobody would name as a brand.

The two also fail differently. A missing mention usually means the pages behind the answer never earned a citation. A missing product card is more mechanical: the card is built from structured product data, so a product absent from every carousel often has a product feed problem or a listing problem rather than a content problem. That is why the honest diagnosis distinguishes the two instead of prescribing content for a feed fault.

Where do assistants show products?

Two surfaces carry product carousels at meaningful volume as of mid-2026. ChatGPT Shopping renders product cards when a question shows shopping intent, with results OpenAI states are “chosen independently and are not ads”. Google’s AI Mode answers shopping questions with “price, reviews and inventory info” drawn from the Shopping Graph, which Google describes as holding more than 50 billion product listings with 2 billion refreshed every hour.

Assistants that write about products are not the same thing as surfaces that render a product carousel a shopper can act on, and the two lists do not match, which is why a shopping measurement spread across a whole assistant panel would be measuring blanks on most of it. Counting the surfaces honestly matters for cost too: a shopping check runs per prompt per surface, so the surface count is a multiplier on the bill, not a detail.

How is AI shopping visibility measured?

By asking the questions shoppers ask, on each surface, and reading the cards out of each answer: which products rendered, whose they were, in what position, at what price. The cards are then matched against the seller's own catalogue, which is what turns a pile of screenshots into a product-level measurement. Two rules keep the number honest: a card that names no seller is reported as not named rather than dropped, and a sponsored card is counted apart, because paid placement is bought rather than earned.

Is your catalogue in the answers?

Related terms

• ChatGPT ShoppingChatGPT Shopping is the product-results experience inside ChatGPT: when a question shows shopping intent, the assistant renders a carousel of product cards with images, prices and reviews.
• Agentic commerceAgentic commerce is shopping where an AI agent completes steps a person used to do, from comparing products to placing the order, with the shopper confirming the purchase.
• Product feedA product feed is the structured file or API stream of a shop's products: titles, identifiers, prices, availability and attributes, kept current as stock changes.
• AI Share of VoiceAI share of voice is the proportion of AI answers naming your brand, measured across a panel of tracked prompts and compared against the rivals named in the same answers.