How can I check if my product listings are actually showing up in AI shopping results?

Ecommerce clients usually ask this after someone on their team tries ChatGPT and sees a competitor's product in a shopping carousel. The question then is: are we in there or not, and how would we know without checking by hand every morning? This is how we handle it.

Start by naming the surface

"AI shopping results" is not one place. ChatGPT can show shopping carousels. Google has shopping features in AI Overviews and AI Mode. Copilot has its own behavior, and Amazon Alexa returns product results. A product can show on one and not on another, so we always write down which surface we are checking.

Step 1: a short manual test

We open each assistant and ask the questions shoppers ask.

"Best [product type] for [use case] under [price]."
"[Brand] [product] reviews."
"[Product type] gift ideas for [recipient]."
"[Product] vs [competitor product]."

We note whether the product appears, where, and with what title, price and image. Then we repeat on another day and from another location. Results vary, so this is a smoke test, not data. We tell clients that before we start, so nobody builds a strategy from one screenshot.

Step 2: the engines' own tools

For Google, Merchant Center is the foundation for product listings. We check feed status, disapprovals and attribute completeness. Google's documentation is the authority on what AI Mode shopping needs now, so we read it directly and do not rely on third-party summaries. For Microsoft, we use Bing Webmaster Tools and check Microsoft's documentation for what is currently reported on Copilot.

These tools tell us whether the data is healthy. They do not run shopper prompts across assistants and tell us whether the product was recommended.

Step 3: daily tracking

For that we use Promptwatch, which is the platform we run client programs on.

AI Shopping covers ChatGPT shopping carousels and Amazon Alexa product results. A client can upload their own catalog, and we track specific products and their position over time.
Shopping Insights gives product-level analysis. It starts on the Business plan ($579/mo), so we plan for that when a client wants it.
Ads Radar, also from Business, captures sponsored ads in ChatGPT Search, Google AI Overviews and Google AI Mode answers to tracked prompts: advertiser, copy, landing page and the triggering prompt. It does not cover Perplexity, and it does not report on a client's own ad campaigns.
Prompt tracking with persona stacking, so we can run the same shopper question as a budget buyer and a premium buyer.
Search Insights, which as of late October 2026 overlays Google Search Console and Bing Webmaster data, including Copilot, on AI crawls and citations.

Step 4: explain the misses

A yes or no is not enough. When a product is missing, we ask why.

Citations. Promptwatch shows which pages and domains appear in answers for the category, including Reddit, YouTube and review sites. If the winners are discussed in places the client has no presence, that is the gap.
Crawler access. Agent Analytics shows IP-verified AI crawler logs, including bots such as OAI-SearchBot and ChatGPT-User, with error tracking. If product pages return errors, we fix that first. Crawler logs start on Professional ($245/mo).
Product data. Thin titles and missing attributes make products harder to match to a query. We fix them at the source, in the feed and on the product page.
Content. Comparison and use-case pages help the text answer, and the shopping surface sometimes follows the same signals.
Traffic. Visitor analytics shows whether AI referrals reach product pages and convert.

What we tell ecommerce clients

We cannot guarantee a product will show up. We can show where it appears, where it does not, and what the winners have that the client lacks. We also warn that results vary by location and by day, so we judge trends over two weeks, not a single check.

Our cadence

Weekly we look at the products that dropped or appeared, review the citations on the lost prompts and choose one fix. Monthly we report a trend to the client, including traffic from AI assistants. The Action Board generates a prioritized list from the data, and we review it before anything is acted on.

If you only do one thing this week, do the manual test with at least five shopper prompts and write down what you see. Then pick the products that matter most and put them under daily tracking, so the next time a competitor appears in a carousel you already have the answer.