AI Overviews Hit 14% of Shopping Queries: What It Means

For about two years, ecommerce had a quiet little secret. While publishers and review sites watched AI Overviews chew through their traffic, online shops mostly got left alone. Google rarely slapped an AI summary on top of a "buy" search, so the clicks kept coming.

That truce is over. AI Overviews now appear on 14% of shopping queries, and the speed of the change is the part that should make you sit up. We track this stuff daily across client accounts, and we still did a double take when the numbers landed.

So here is the honest version of what is happening, what we actually see in our own campaigns, and what we would do about it if it were your store. No panic, no doom. Just the maths and a plan.

What does "AI Overviews on 14% of shopping queries" actually mean?

AI Overviews are the AI-generated summary boxes Google parks at the top of the results, above the blue links you have been chasing for years. On a shopping query, that box can recommend products, compare options, and pull in prices before a shopper ever scrolls to your listing.

The 14% figure comes from a study by Visibility Labs, which checked more than 20.9 million shopping keywords (the ones that trigger a Shopping box) and found that AI Overviews appeared on roughly 2.9 million of them. As Search Engine Land reported on the research, that is a 5.6x jump in four months, climbing from about 2.1% in November 2025 to 14% by March 2026.

You can read the full Visibility Labs analysis for the methodology, but the headline is simple. The category that thought it was immune just got drafted into the AI Overviews club.

"Focusing on AI SEO is no longer a luxury, it's becoming a necessity," said Jeff Oxford, founder of Visibility Labs. We would only add: it stopped being optional for publishers a year ago, and now retailers have caught up to the same reality.

Why now? Google quietly turned shopping into an AI product

This did not come out of nowhere. The 14% spike lines up almost perfectly with Google bolting AI directly onto shopping in November 2025.

In its official announcement on 13 November 2025, Google rolled out agentic checkout, conversational AI Mode shopping, and shopping inside the Gemini app. It also confirmed its Shopping Graph now holds more than 50 billion product listings, with 2 billion of them refreshed every hour.

Read that back. Fifty billion listings, two billion updated hourly. That is not a feature Google built to sit on the bench. When a company wires that much infrastructure into search, the share of queries getting an AI treatment only goes one direction.

Our take: the timing is the tell. The infrastructure shipped, and the AI Overview share on shopping queries climbed within weeks. If you are weighing up where to put effort, we cover the moving parts in our guide to AI search visibility and generative engine optimisation, because the rules for getting cited are not the rules you grew up with.

The click problem, with real numbers attached

Here is where retailers get understandably twitchy. If an AI summary answers the question, why would anyone click through to your product page?

The most credible data we have on this comes from the Pew Research Center's study of real browsing behaviour from 900 US adults. When an AI summary appeared, users clicked a traditional search result just 8% of the time, compared with 15% when there was no summary. That is nearly half the clicks gone.

Worse for the "well, they will click the links inside the summary" optimists: Pew found people clicked a link inside the AI summary in only 1% of visits. And users were more likely to end their session entirely after seeing a summary (26% of pages) than without one (16%).

We will be straight with you. Those Pew numbers are across all query types, not shopping specifically, and they predate the November shopping push. But the direction of travel is consistent with what we see in client Search Console data once an AI Overview lands on a money keyword. Impressions hold up. Clicks sag. We dug into that pattern in our breakdown of the click-through rate drop AI Overviews are causing, and shopping is now firmly inside that story.

Which shopping searches are getting hit, and which are not

Not every product search is equally exposed, and this is the single most useful thing to understand. Google is not blanketing every query. It is reading intent.

Here is the split we see play out, and it is the difference between losing sleep and sleeping fine:

• Research and comparison queries get hammered. "Best robot vacuum for pet hair", "Sony vs Bose headphones", "which air fryer is quietest". These are questions with many valid answers, which is exactly what an AI summary loves to chew on.
• Branded and specific transactional queries are mostly left alone. "Buy Nike Air Max 270 size 10", "Dyson V15 price". The shopper already knows what they want. Google has less reason to interrupt with a summary, so for now it usually does not.

This matches the broader pattern in Search Engine Journal's coverage of BrightEdge research, which found AI Overview coverage grew 58% across nine industries between February 2025 and February 2026, with Google leaning on query intent rather than category to decide when a summary shows.

One caveat we will flag honestly. You will see other sites quote precise figures like "83% of best-product queries trigger AI Overviews". We could not trace that exact number back to the Visibility Labs primary source, so we are not going to repeat it as gospel. What we can say with confidence: high-consideration, research-style shopping queries are dramatically more exposed than ready-to-buy branded ones. That distinction is real, and it should shape your whole approach.

The 86% that everyone forgets to mention

Quick reframe, because the panic crowd needs it. If AI Overviews appear on 14% of shopping queries, then 86% of them still serve plain old results with no AI box in the way.

That is the overwhelming majority. And it is the part of the market that gets neglected while everyone screams about the 14%.

Our strategy in client accounts is deliberately two-sided. We chase visibility inside the AI Overviews where they appear, and we deliberately go after the long-tail, specific, transactional queries where they do not. "Wireless earbuds" might get a summary. "Wireless earbuds for small ears with USB-C under 80 pounds" almost certainly will not, and the buyer typing that is a lot closer to checkout anyway.

Specific beats broad right now, in a way it has not for years. If you have been ignoring the boring deep tail of your catalogue, this is your sign. Our work on ecommerce SEO that actually converts leans hard into that long-tail layer for exactly this reason.

How to get your products cited inside shopping AI Overviews

Defence is half the game. The other half is showing up inside the summary so you are the brand Google recommends instead of your competitor. Here is the playbook we run.

1. Get your structured data genuinely clean. Product, Offer, AggregateRating and Review schema, with the real fields filled in: brand, SKU, price, availability, currency, review count. Google's AI pulls from structured signals, and half-completed markup is a wasted opportunity. Schema spent a decade as the overlooked plumbing of SEO. It just became load-bearing.
2. Build comparison content that a human would actually thank you for. Side-by-side specs, honest pros and cons, "who this is for" sections. Recycled manufacturer blurbs do not get cited. Genuine comparison does, because that is the exact query type AI Overviews target.
3. Earn authority, not just rankings. Sites with real topical depth and a credible backlink profile get pulled into AI answers more often. This is where consistent link building that earns citations, not just rankings does double duty across both classic search and AI surfaces.
4. Cover the entity, not just the product. Answer the surrounding questions: care, sizing, compatibility, returns, comparisons. Depth around a product cluster signals to Google that you are a trustworthy source on the whole topic, not a single thin page.
5. Measure citations, not only positions. Track when and where your brand appears inside AI Overviews and AI assistants. Rank tracking alone misses the new battlefield entirely.

If you want the deeper version of the "how do I actually get pulled into the box" question, we wrote a full guide on improving your AI Overviews citation rate and a companion piece on getting your brand into AI answers across ChatGPT, Gemini and Google. Same principles, applied to product catalogues.

Should you actually change your 2026 ecommerce strategy?

Short answer: yes, but not in the way the doom merchants suggest. Do not torch your SEO budget. Adjust where it points.

Here is the perspective that keeps us calm. People are still buying online, a lot. The US Census Bureau's most recent data, released on 18 May 2026, put first-quarter 2026 ecommerce sales at $326.7 billion, which is 16.9% of all retail sales and up 9.8% on the same quarter a year earlier. The demand is growing. The pathways into your store are just shifting.

So our advice to ecommerce clients right now breaks down like this:

• Own the channels Google cannot rewrite overnight. Email, SMS, and a base of customers who search your brand directly. Brand demand is the one asset an AI Overview cannot intercept.
• Treat AI search as its own surface. Optimising to be cited inside AI Overviews and assistants is now a discipline in its own right, not a footnote to classic SEO. It is the core of what we do in AI search optimisation.
• Defend the deep tail. Build out the specific, intent-rich product and comparison pages where AI Overviews rarely intrude and buyers are closest to purchase.
• Diversify discovery. Visual platforms matter more than ever for products. Our take on Pinterest and visual search for ecommerce covers a channel a lot of stores underrate.

One genuine frustration we will name: attribution from AI surfaces is still a mess. You can get visibility inside an AI Overview, influence a purchase, and have your analytics credit "direct" or nothing at all. Anyone promising you clean AI Overview conversion tracking today is selling you something. We measure what we can, triangulate the rest, and we are upfront about the gaps.

Where this goes next, and what we would do today

The 14% is not a ceiling. It is a checkpoint. Google did not pour billions into AI and a 50-billion-listing Shopping Graph to show summaries on one in seven searches and call it a day. The honest expectation is that the share keeps climbing, especially on research-heavy queries.

The wider trend backs that up. We charted the broader expansion in our look at how AI Overviews surged 58% across industries, and shopping is simply the latest category to join the curve rather than an outlier.

If you run an online store and you want a straight answer on where you stand, that is the work we do every day. We will pull your real AI Overview exposure, show you which money keywords are at risk, and build the offence-and-defence plan around it. Tell us your situation on our contact page and we will give you the unvarnished version, not a fear pitch.

The stores that win in 2026 are not the ones fighting AI Overviews. They are the ones who got cited inside them, fortified the queries AI ignores, and stopped renting their entire business from a single algorithm. That shift is very doable. It just rewards the people who start now instead of arguing about whether it is real.