AIuthority

New Analysis Reveals ChatGPT Relies Heavily on Google Shopping for Product Carousels

By Charles Ryder

The latest research on AI-powered shopping points to a simple reality: ecommerce discovery may look different on the surface, but Google still appears to be doing much of the heavy lifting behind it.

A large-scale analysis published in early March 2026 found that ChatGPT sources 83% of its product carousel results from Google Shopping through what researchers call shopping query fan-outs, or QFOs. That figure stands out not just for its size, but because it confirms a theory many search and ecommerce professionals have had since OpenAI launched ChatGPT’s shopping research feature in late November 2025.

In plain terms, when users ask ChatGPT for product recommendations, the carousel it generates does not seem to come from a fully independent shopping index. The evidence suggests it leans heavily on Google Shopping’s organic product results.

Chart illustrating ChatGPT's 83% reliance on Google Shopping for product carousels, showing data from the Peec AI analysis. The Research Behind the Headline

The most detailed study came from Tom Wells of Peec AI, who analyzed more than 43,000 products and found a stable pattern that had held for about four months. His conclusion was hard to miss: most ChatGPT carousel products closely matched listings already ranking in Google Shopping’s top 40 organic results.

This was not a one-off finding.

Semrush had previously run its own test across 100 prompts, repeated five times each. It found that the top product shown in ChatGPT’s carousel appeared in Google Shopping’s top three organic positions 75% of the time. Researchers also spotted separate shopping-related queries in ChatGPT’s network behavior, which suggests the model uses a different retrieval process for product carousels than it does for general web context.

That distinction matters. ChatGPT shopping does not appear to be simple AI summarization layered over standard search. It looks more like a dedicated product retrieval pipeline, one that still depends heavily on Google’s shopping ecosystem.

How ChatGPT Shopping Seems to Work

OpenAI introduced its shopping research experience to help users ask natural-language questions such as “What’s the quietest cordless vacuum for a small apartment?” and get curated recommendations, comparisons, and deal information.

Behind the scenes, though, the system appears more structured than most users would assume.

Researchers found that ChatGPT uses two types of fan-out queries:

  • Contextual QFOs for general information and supporting text
  • Shopping QFOs for retrieving product-specific results used in carousels

These shopping QFOs are shorter, highly specific, and usually unique to the prompt. That suggests ChatGPT is generating targeted product-search instructions in the background, then pulling from a shopping-oriented result set that closely mirrors Google Shopping visibility.

The study also found that:

  • 60% of matching products came from Google Shopping’s top 10
  • 84% came from the top 20
  • Bing played only a minimal role
  • Just 0.16% of products were found exclusively in Bing

That last point is especially notable given OpenAI’s relationship with Microsoft. Anyone expecting Bing to serve as the primary shopping backbone for ChatGPT is not seeing that reflected in the data.

Why This Matters for Brands and Retailers

This shifts the ecommerce optimization conversation in a meaningful way.

If ChatGPT is becoming a product discovery channel—and the current evidence says it is—then Google Shopping rankings are no longer only about Google. They may also determine visibility inside AI-driven product recommendations.

For brands, that raises the stakes. If a product does not rank well in Google Shopping organic results, it may be effectively absent from ChatGPT product carousels as well. As Tom Wells put it, products outside Google’s top 40 are essentially excluded.

For ecommerce teams, the implications are immediate:

  • Optimize Google Merchant Center feeds
  • Use clear, intent-matching product titles
  • Keep pricing, GTINs, availability, and imagery current
  • Strengthen review signals and product sentiment
  • Treat structured product data as part of AI visibility strategy

This is where traditional SEO, shopping optimization, and what many now call Generative Engine Optimization (GEO) start to overlap.

Diagram explaining the The Bigger Competitive Story

There is a broader platform story here too.

For years, many assumed AI assistants would weaken Google’s role as the default starting point for commerce discovery. This research suggests a more complicated reality: AI may be changing the interface while still relying on Google’s infrastructure and product data underneath.

So while the user may begin with ChatGPT, the underlying product visibility may still be decided by Google Shopping.

That carries major strategic consequences. It reinforces Google’s moat in ecommerce discovery, raises fresh questions about how AI systems source product information, and pushes brands to stop thinking about channels in isolation. Search, shopping, and AI recommendations are converging quickly.

With ChatGPT reportedly handling tens of millions of shopping-related queries each day and commercial intent continuing to rise across AI conversations, this dependency could shape the next phase of online retail more than many marketers expect.

FAQ

What did the March 2026 analysis find?

The analysis found that 83% of ChatGPT product carousel results aligned with Google Shopping through shopping query fan-outs, indicating a strong dependence on Google’s organic shopping listings.

Does ChatGPT use Bing for shopping results?

Only to a very limited extent. The research found that Bing played a minimal role, and just 0.16% of products appeared exclusively in Bing.

Why should ecommerce brands care?

Because visibility in Google Shopping may now influence whether products appear in ChatGPT carousels. Strong feed quality, structured data, reviews, and optimized product listings could affect both search and AI discovery.

What are shopping QFOs?

Shopping QFOs, or shopping query fan-outs, are targeted retrieval queries ChatGPT appears to generate behind the scenes to pull product-specific results for its shopping carousels.

Conclusion

This analysis is a clear wake-up call. ChatGPT may feel like a new shopping frontier, but the available evidence suggests its product carousel ecosystem still depends heavily on traditional search visibility—especially Google Shopping. For brands, publishers, and ecommerce teams, AI optimization is no longer a separate track. It is tied directly to feed quality, organic shopping performance, and structured product data. If you want to keep up with changes like this and better understand AI visibility, AIuthority is a solid place to start.