Same Rank, Fewer Clicks: The Store Pages AI Overviews Are Eating

The ranking report says fine. The revenue report disagrees.

For twenty years, ecommerce SEO ran on a simple chain: rank higher, get more clicks, sell more. That chain has snapped in the middle. A category page can hold position one for months while the clicks underneath it quietly drain away.

The best large-scale evidence comes from Ahrefs, which sells SEO software and so has an interest in how search visibility is measured. According to MediaNama's coverage, Ahrefs compared desktop click-through rates on 300,000 keywords between December 2023 and December 2025. On keywords with AI Overviews, the top result lost 58% of its clicks, up from a 34.5% loss in its April 2025 study. Position two lost 50.8% and position three lost 46.4%.

Source: Ahrefs data on desktop click-through rates, as reported by MediaNama.

The broader trend points the same way. SparkToro, using Similarweb clickstream data, found 68.01% of U.S. Google searches ended without a click in early 2026, according to Search Engine Land. SparkToro itself cautions that its long-term comparisons use different data panels and are not directly equivalent.

Not every lost click is an AI Overview

Before you blame AI for everything, remember that zero-click search predates it. MediaNama's coverage of the Ahrefs study notes that featured snippets, local packs and top stories also pull clicks away from organic listings. AI Overviews are the newest and largest thief, not the only one.

The cleanest evidence on AI summaries specifically comes from Pew Research Center, which tracked the real browsing of 900 U.S. adults in March 2025 across 68,879 Google searches. Pew is a nonprofit with no product to sell, which is why its numbers are worth more than most.

Google disputes the framing. In August 2025, Head of Search Liz Reid wrote that total organic click volume has been "relatively stable year-over-year" and that Google is sending "slightly more quality clicks to websites than a year ago," meaning clicks where people do not bounce straight back.

We continue to send billions of clicks to the web every day and are committed to prioritizing the web in our AI experiences in Search.

Both things can be true. Total clicks across the web can hold steady while the clicks to your specific pages collapse. What matters for a store is which of its pages sit in the path of the answer.

Penalty or answer? Read Search Console like a detective

A penalty and an AI answer look similar on a traffic chart and completely different in Search Console. The difference is in which metric moves first.

Search Console keeps 16 months of Performance data, so use all of it to compare the same season year over year. Then split your queries by intent. Put brand and product-name searches in one group, and question-shaped searches ("best," "how to," "vs," "is it worth") in another. If the question group's click-through rate collapsed while the brand group held, you are looking at AI answers, not a penalty.

Google's own reporting helps, with a gap. Its generative AI performance reports, rolled out globally in August 2026, show impressions in AI Overviews and AI Mode by page, country and device, but no clicks. Google's AI features documentation also says clicks from AI features are counted inside the regular Web search type, so they are blended with everything else. For turning these numbers into something leadership can read, see our guide to SEO reporting metrics for executives.

Which store pages take the hit

The damage follows the shape of the query. Pew found that AI summaries appeared on 60% of searches phrased as questions and 53% of searches of ten words or more, but only 8% of one or two word searches. Short, specific product searches were largely spared. Long, curious ones were not.

For a typical store, that makes the buying guide the most exposed page on the site. "Best trail running shoes for wide feet" or "how to choose a standing desk" is exactly the query Google can summarize from ten sources. The product page for a specific shoe, searched by name, is far safer, for now.

The "for now" matters. Semrush tracked more than 10 million keywords through 2025, and Search Engine Land reported that the mix of queries showing AI Overviews shifted sharply. Semrush owns Search Engine Land, so treat this as vendor data reported by a sister publication.

AI Overviews are also far more common than they were. Comscore data reported by Search Engine Land found them on 39.4% of U.S. desktop searches in June 2026. The same report found YouTube, Amazon, Reddit and Wikipedia were the most cited sources for retail searches, which tells you who you are competing with for a place inside the answer.

Your Merchant Center feed is now a storefront

On a buying query, your web page is no longer the only way in. Google increasingly assembles shopping answers from structured product data, and the feed you send to Merchant Center is becoming as important as the page you rank.

In January 2026, Google announced the Universal Commerce Protocol with partners including Shopify, Etsy, Wayfair, Target and Walmart, along with "dozens of new data attributes" in Merchant Center built for conversational shopping, such as compatible accessories and substitutes. In May 2026, it added conversational attributes and AI performance insights in Merchant Center, plus a Universal Cart for buying across retailers with Google Pay.

The scale behind it is enormous. Google says AI Mode's shopping results draw on a Shopping Graph of more than 50 billion product listings, with more than 2 billion refreshed every hour. When a shopper asks for "barrel jeans that aren't too baggy," the answer is built from attributes, not from whoever has the best backlinks.

The purchase itself is moving too. In September 2026, Google began switching on native checkout in AI Mode and Gemini by default for eligible Shopify stores, as we covered in our September 21 to 25 roundup. An order placed there never touches your site, so it will not show up as an organic session at all.

The real value of being cited

If clicks are shrinking, is it worth fighting for a spot inside the AI answer? Yes, but be honest about what you are buying.

The real value of a citation is presence at the moment of decision. A shopper comparing three standing desks who sees your brand named in the answer has heard of you before they ever reach a product page. That shows up later as branded search and direct visits, not as a click from the Overview itself. Our guide to optimizing content for AI Overviews covers how to earn that placement.

Why pulling the plug backfires

It is tempting to pull the plug. Google now makes that possible: the Search generative AI control in Search Console lets a site stop appearing in AI Overviews, AI Mode and Discover's AI summaries.

Sites that opt out will not receive traffic or impressions from our generative AI features.

That warning comes from Google, as quoted by Search Engine Land. The query still gets an AI answer, assembled from your competitors. Google says the setting is not a ranking signal for regular results, but the trade is still lopsided for most stores.

The older tools do not solve it either. Google's AI features documentation says nosnippet, data-nosnippet, max-snippet and noindex control what appears in AI features, but those same controls also shrink or remove your regular search snippets. Google-Extended governs AI training in some of Google's other systems, not whether you appear in AI Overviews.

Give shoppers something a summary cannot

An AI answer can summarize what is known. It cannot hand someone the thing they came for. The pages that still win clicks are the ones where the next step only exists on your site.

Rebuild your guides around those. A buying guide that ends with a sortable comparison of in-stock products, real test photos and a fit tool gives the shopper a reason to click that a summary cannot replace. A guide that only restates spec sheets is exactly what the AI answer already did better.

Where to start

Rankings used to be the scoreboard. For online stores, they are now just the ticket to be considered. The stores that win the next few years will be the ones that show up inside the answer, in the feed and on the page where the purchase actually happens.