Semrush: AI Overviews Hit 85% of Ecommerce Keywords

Semrush: AI Overviews Hit 85% of Ecommerce Keywords
Semrush: AI Overviews Hit 85% of Ecommerce Keywords
Affiliate disclosure: This post contains affiliate links. If you buy through them, I may earn a commission at no extra cost to you. Full disclosure

Semrush says Google’s AI Overviews now appear on 85% of tracked ecommerce keywords, up from 33% in September 2025.

That lands on the pages that carry a high-ticket dropshipping store: category guides, best-of lists, comparisons and review pages. Those are the searches where a buyer weighing a $1,500 purchase starts, and the study shows an AI summary now sits above nearly all of them. This is the Ecommerce Paradise breakdown for store owners.

The study measured whether an AI Overview appeared, not whether clicks fell, so nobody can price the damage yet. Below you get the numbers with their baselines, the counterpoints, and the five changes I would make to commercial pages this week.

Organic clicks are getting harder to predict, so build the legal base under the store first: form your LLC with Northwest, the best LLC formation service for ecommerce owners. Cheap formation sites that upsell at checkout turn the renewal bill into a surprise, while Northwest keeps the renewal price the same as year one with no upsell stack. Form your LLC with Northwest →

Semrush Finds AI Overviews on 88% of Commercial Keywords

Semrush published the study on October 21, 2026. Luke Harsel wrote it, with Nikola Manojlovic contributing, according to the Semrush blog. The team tracked a fixed panel of 400 commercial keywords across banking, software and ecommerce every month from September 2025 through August 2026, using Semrush’s U.S. desktop data.

Coverage across the whole panel, weighted by keyword count, rose from 45.75% in September 2025 to 88% in August 2026. Weighted by search volume it reached 97.08%. The 12% of keywords with no AI Overview accounted for under 3% of search volume, per Semrush.

Ecommerce moved furthest. It climbed from 32.62% to 85.11%, while software went from 68% to 91% and banking finished near 89%, as PPC Land reported on October 22. The study also ran every industry against five search modifiers, and the shifts are large.

Search modifier Sept 2025 Aug 2026
Reviews 30.14% 89.04%
Best X for Y 42.72% 96.12%
Alternatives 48.44% 82.81%
Price and cost 43.59% 82.05%
Brand vs. brand 63.41% 86.59%

Of the 15 industry and modifier combinations, 12 reached at least 80%, and every one ended higher than it started. Ecommerce price and cost queries were the weakest ecommerce cell at 71.43% in August 2026, but that cell was at 4.76% a year earlier.

Semrush’s own read is blunt. Per the study, “ranking number one likely won’t lead to as much traffic as it did this time last year.”

What the study does not tell you

Semrush’s data covers U.S. desktop only and three industries. It has no click, traffic or conversion numbers. PPC Land’s read of the percentages implies roughly 20 to 30 keywords per cell, so a single keyword can move a cell by several points. Semrush also sells AI visibility tracking, and it is now part of Adobe, so the recommendations in its post serve its product line.

From 46% to 88%: How AI Overviews Took Over Commercial Search

This did not arrive overnight. Semrush’s previous study, published July 2, 2026, covered more than 600,000 keywords across 10 industries from November 2025 to April 2026. It found AI Overviews on commercial-intent results grew an average of 71% in those six months, while the share of transactional-intent results with one fell 5%.

That split mattered. Google was putting AI summaries on research queries first and leaving “buy now” searches mostly alone. The new study suggests the line is blurring, because “best X for Y” and “price and cost” queries sit between research and purchase.

Branded searches moved the same way this fall. Per Search Engine Journal, Ahrefs found AI Overviews on 5 of 98 usable brand-name results on July 1 and on 63 of 100 by September 30. DemandSphere’s series went from 26.12% of branded keywords on September 1 to 90.48% on September 27.

On the traffic side, Marie Haynes analyzed 100 websites and 12 months of GA4 data. Per Search Engine Land, ChatGPT drove about 0.62% as many sessions as Google organic search, and Google organic produced 27 times the conversions of all AI assistants combined. I covered those referral numbers in my post on ChatGPT versus Google organic traffic.

The counterpoint

Two things cut against panic. First, Semrush measured presence, and Google’s own click behavior on these results is not in the data. Second, Haynes found that ChatGPT visitors converted at 13.8% against 3.02% for Google organic, so the smaller channel is not the weaker one per visitor. Her sites also varied in size, and the report does not define how conversions were counted.

There is a further wrinkle on measurement. Ahrefs says Google alters 5% to 20% of the results it serves to suspected SEO scrapers, per PPC Land, which means your rank tracker may be showing you a slightly wrong page. I wrote that one up in Ahrefs says Google feeds SEO tools swapped results.

What 85% AI Overview Coverage Does to Your Store’s Organic Traffic

This section is my read, not reporting. My read is that the review and “best X for Y” pages are the ones taking the hit first, because those modifiers carry the biggest jumps, and those are the pages most stores built to catch buyers in the middle of the funnel.

Here is hypothetical math to size it. Say a store gets 1,000 organic visits a month to commercial pages, converts 1% of them, and averages a $1,500 order. That is $15,000 a month. If an AI summary cuts those clicks by 20%, you lose $3,000 a month without a single ranking change. If it cuts them by 40%, you lose $6,000. These numbers are invented to show the mechanism, and the real loss depends on how Google treats your specific queries.

Here is how I would set thresholds, again as a working rule and not a finding. If organic clicks to a commercial page fall less than 10% after an AI Overview appears, leave the page alone and spend your time elsewhere. If they fall 10% to 30%, rewrite the page so it answers the question the summary cannot, which for high-ticket means exact pricing, freight cost, lead time and warranty terms. If they fall more than 30%, treat that page as a lead-capture asset and stop counting on it for first-touch sales.

Some stores will read this and decide to chase AI citations on every page. I would not. Semrush’s July study found AI Overviews and Google Ads now appear on the same results page roughly twice as often as a year earlier, which means paid and organic are being squeezed together. Spreading effort across fifty pages for a citation you cannot measure is a worse bet than fixing your ten highest-revenue pages and tracking calls and form fills.

There is also a competitive angle. Most of your rivals will not read this study. A competitor who keeps publishing thin “best of” pages with affiliate-style blurbs will lose clicks to the summary and stay confused about why. A store that publishes real price tables, real shipping terms and a straight comparison gets quoted, gets the call, and takes the order.

Position stops being the metric that matters. A page ranked first under an AI box can earn fewer clicks than a page ranked third a year ago. So the useful questions are whether your page is the source the summary cites, and whether the visitor who does click is ready to buy.

The Mailchimp versus Klaviyo example in the study is worth studying. For that query Google’s summary cited Klaviyo’s own guide for one option and a third-party site for the other, plus a YouTube video, per PPC Land. That is one example, not a panel finding, but it shows the pattern: whoever publishes the clearest comparison on its own domain gets quoted.

For a high-ticket store this is a mixed picture. Buyers spending four figures still want a phone call, a warranty answer and proof the seller is real, none of which a summary box can provide. That is why my guide on turning Google Shopping clicks into sales for high-ticket products still applies to organic clicks too. Fewer clicks make each one more valuable, and the conversion system behind them matters more, as I explain in why high-ticket Google Ads need a complete conversion system.

The same logic argues for owning more of your demand. A store that depends on one commercial-page channel is exposed, and I laid out the broader plan in how to scale a high-ticket dropshipping store without cutting prices. If you would rather hand the rebuild to a team that does this every day, my turnkey done-for-you service builds and runs the store for you.

Want to work through your commercial-page rewrite with other store owners and me? Join the Skool community →

Five Changes to Your Commercial Pages This Week

You do not need a site rebuild. You need to find which pages sit under an AI summary and make each one harder to replace. Do these five in order, and block out one afternoon for the first two because they decide where the rest of your time goes.

  1. Pull your top 20 organic landing pages from Google Search Console and tag each one by modifier: reviews, best X for Y, alternatives, price and cost, or brand vs. brand. Then check which of those keywords now trigger an AI Overview in Semrush. Cross-check a sample in Ahrefs given the scraper-filtering caveat above. Search the keyword yourself in a clean browser window too, since that is the one view no vendor can alter.
  2. Fix your price and cost pages first, since that was the ecommerce cell with the most room at 71.43%. Put real prices, shipping and freight terms, financing options and warranty length on the page in plain text. Use KWFinder to find the exact cost phrasing your buyers type.
  3. Publish your own head-to-head comparison for every brand you carry, with a spec table and a clear pick for each use case. The Mailchimp example shows the owner of the clearest comparison gets cited. Use AlsoAsked to pull the real questions buyers ask, and answer them in short blocks near the top of the page.
  4. Judge pages by conversions from organic landing pages in GA4, not by position, and keep a weekly rank check in SE Ranking only as a secondary signal. Then capture the visitors you do get, because a click now costs more to earn. An Omnisend popup offering a buying guide on your top five pages is a fast start.
  5. Put your phone number and a lead form above the fold on every commercial page, and route those leads fast. My guide on high-ticket sales follow-up covers what to say when the call comes in. For the tracking side, I show how to set up HubSpot free for a high-ticket store. A summary box cannot take a call, and high-ticket buyers still make them.

If you are still choosing what to sell, pull my free high-ticket niches list before you write a single review page. The category math behind it is in my breakdown of the 150+ best dropshipping niches for 2026.

Explore Turnkey Builds for service details and pricing. Email me with questions before getting started.

Frequently Asked Questions

Do AI Overviews mean my organic traffic will drop 40%?
No study here measured clicks, so no one can give you a number. Semrush measured how often the summary appears. Pull your own Search Console data for the pages in question and compare clicks before and after the summary showed up. My guide to answer engine optimization for AI shopping discovery covers the next layer.

Which of my pages are most exposed?
Review pages and best-of pages. Semrush found “best X for Y” at 96.12% coverage and reviews at 89.04% by August 2026, the two highest modifiers across the whole panel.

Should I stop investing in SEO?
No. Haynes’s data shows Google organic still produced 27 times the conversions of all AI assistants combined. The work shifts toward being cited and converting the click, not abandoning search. My high-ticket Black Friday checklist covers the conversion side before peak season.

Can I trust the Semrush numbers?
Treat them as direction, not precision. The panel is 400 keywords on U.S. desktop, the cells are small, and Semrush sells tools that benefit from this story. The trend matches the Ahrefs and DemandSphere branded-search data from this fall.

Does Google Shopping still work if organic shrinks?
Nothing in this study touches paid Shopping, so keep it running while you fix organic pages. I covered how Google’s AI features are changing ads in Google AI Mode ads run through your existing campaigns.

Can AI tools help me rewrite these pages faster?
They can draft Q&A blocks from your support tickets, but a human should check every spec and price. I walk through the setup in Claude for ecommerce.

If you want to try it on one page first, start with Claude and your ten most common support emails.

Does my Shopify theme matter for any of this?
Only for page speed and clean markup. The change here is about Google’s results page, not your theme. A clean Shopify build just makes the page edits above faster.

Would you rather have my team build and run a high-ticket store that is ready for AI-era search? See the turnkey done-for-you service →

I will keep tracking this one as Google’s click data and the next Semrush panel come in. Subscribe to the YouTube channel for daily breakdowns. More breaking news coming through the day.

Related Articles

If this was useful, these go deeper: