Home/News/Study: Only 1.28% Product Overlap Between Google AI Mode and the Shopping Carousel

SEO & GEO

Study: Only 1.28% Product Overlap Between Google AI Mode and the Shopping Carousel

2026-09-03 Search Engine Journal SEO & GEO
Study: Only 1.28% Product Overlap Between Google AI Mode and the Shopping Carousel
图:Search Engine Journal

Product tracking firm Productrise published data showing that the products Google's AI Mode surfaces rarely match the Popular products carousel in regular search, even for the same query on the same day. Google told Futurism it has not verified the figures.

Key facts

  • Timing: Data collected August 9–31, 2026, across US and UK queries.
  • Who it hits: Any store relying on organic Google shopping visibility, especially where the same product is sold through several channels.
  • What to do: Nothing to change in ads. Start tracking AI Mode as a separate channel instead of inferring it from carousel performance.

What happened

Between August 9 and August 31, 2026, Productrise monitored over 2 million product listings across more than 100,000 regular search results and AI Mode responses, running the same product-focused queries in both surfaces on the same day in the US and UK and matching items by Google product ID. Only 1.28% of carousel listings carrying a product ID also appeared in AI Mode, averaged daily. When both surfaces returned products, the carousel showed 27.8 products on average versus 3.9 in AI Mode. Among products appearing in both, the first-listed seller differed 49.6% of the time and the price differed on 38.1%; of those price mismatches, AI Mode showed the higher price 68.4% of the time. The queries came largely from products the company already monitors and are not described as a random sample.

What it means for independent store owners

Treat the exact numbers with caution — this is one vendor's dataset on a non-random sample, unverified by Google. The direction still matters. First, AI Mode and the shopping carousel appear to run on largely unrelated selection logic, so carousel performance tells you very little about whether AI Mode will cite you, and vice versa. Analyzing them as one channel will lead you to wrong conclusions. Second, AI Mode surfaces roughly 3.9 slots against the carousel's 27.8. That is close to a sevenfold cut in available positions, and there is no second page to fall back to.

No feed or bidding change is warranted on this alone. What is worth doing is manual observation: pick a handful of your best-selling products, run them in AI Mode every few days, and record whether you appear, which seller is listed first, and whether the price shown matches your storefront. That last one deserves attention — if AI Mode is quoting a reseller's higher price while you sell the same item cheaper, nobody is going to correct that gap for you.

Someone should be watching these changes for you

I run AI automation for independent stores — content, SEO/GEO, product listing, social, and keeping up with policy shifts like this one.

See how it works