AI Overviews on 88% of tracked commercial keywords, Semrush finds

Semrush has published a 12-month study showing that Google's AI Overviews now appear on 88% of a fixed panel of 400 commercial keywords, up from 45.75% in September 2025. Every keyword carries a buying-stage modifier: alternatives, brand vs brand, best X for Y, pricing and cost, or reviews, spread across banking, software and ecommerce. Weighted by search volume, the coverage figure is 97.08%.

For an Australian business paying for SEO, the plain reading is this. The comparison page, the pricing page and the "best X for Y" article you commissioned to catch people close to a decision now sit under an AI summary on most of those searches, at least in the US desktop data Semrush measured. Ranking still matters, but whether your brand is named in the summary, and how it is described, is now a separate question that most rank reports do not answer.

August 2026 reviews near 90%

The biggest mover was the reviews group. In September 2025, searches such as "chase checking account review" triggered an AI Overview 30.14% of the time, the lowest of the five modifiers. By August 2026 that was about 90%, a jump of 59 points and the largest of the five modifier groups.

"Best X for Y" searches finished highest at 96.12%, up 53.4 points. "Alternatives" rose 34 points to almost 83%. "Brand vs brand" had the most coverage a year ago at 63.41% and grew the least, up 23.18 points to a little over 86%.

The spread between modifiers has collapsed. A year ago the range ran from 30.14% to 63.41%. Now it runs from 82.05% to 96.12%. Semrush's conclusion is that an AI Overview should now be the working assumption for almost every commercial search an SEO team goes after.

Ecommerce climbed from 32.62% to 85.11%, still trailing on price/cost

Industry made less difference than you might expect. In September 2025, coverage ranged from 32.62% for ecommerce keywords to 68% for software, a gap of more than 35 points. By August 2026 all three industries sat between 85% and 91%, with ecommerce at 85.11% and the gap under six points.

Of the 15 industry-and-modifier combinations, 12 finished at or above 80%. Three reached 100% in August 2026: "best X for Y" in banking, "alternatives" in software and "best X for Y" in software. Three stayed below 80%: banking "alternatives" at 65%, ecommerce "price/cost" at 71.43% and software "reviews" at 75%.

Even the laggards moved a long way. Ecommerce price and cost searches, such as "coffee maker under 100", went from 4.76% to 71.43%. Banking reviews went from 13.79% to 89.66%. All 15 combinations finished higher than they started.

By volume, coverage is 97.08%

The 88% figure counts each keyword equally. When Semrush weighted the same keywords by estimated monthly search volume, AI Overview coverage was 97.08%. The keywords with no AI Overview made up 12% of the panel but under 3% of its combined search volume.

In other words, the high-demand commercial searches are the ones most likely to carry an AI Overview, and the keyword-count figure likely understates how much estimated search demand now meets an AI answer first.

Klaviyo's guide, but Zapier for Mailchimp, cited on "mailchimp vs klaviyo"

Semrush's worked example is the search "mailchimp vs klaviyo". The AI Overview cites Klaviyo's own guide for the "choose Klaviyo if" part of the answer, but cites Zapier, a third party, for the "choose Mailchimp if" part. A YouTube video and a second third-party site make up the rest of the prominent citations.

That is the pattern to expect on a "[your brand] vs [competitor]" search. The answer draws on your site, your competitor's site, review platforms, Reddit threads and YouTube, and Google decides which sentence comes from where. On 19 October we covered Google's announcement of five link changes, including more inline links placed higher in AI Mode and AI Overviews, which makes who gets cited matter more, not less.

Three limits, including Search Console's missing click data

Three limits are worth stating before anyone changes a strategy on the back of this.

• US desktop only. Semrush used its US desktop data. AI Overviews run in Australia, but this study says nothing about coverage on google.com.au, and nothing about mobile, where a large share of commercial searching happens.
• A 400-keyword fixed panel. Spread across 15 combinations, that averages about 27 keywords per cell, and some cells are smaller: 4.76% is one keyword in 21, and 71.43% is 15 of the same 21. In a cell that size, one keyword gaining or losing an AI Overview moves the figure by about five points, so treat the combination-level numbers as directional.
• Presence, not clicks. The study measures whether an AI Overview appeared. It does not measure what happened to organic clicks underneath it, and Google has still not committed to AI Overview click data in Search Console.

Semrush also sells an AI Overview tracking product, and the study closes with a trial offer. That does not make the numbers wrong, but it is context.

There is a measurement wrinkle on our side of the fence too. On 21 October we reported that some desktop AI Overviews were missing the AI Overview label, which could make a tracker record an AI Overview as absent when it is on the page. The issue was first seen on 20 October, after this study's data window, but it is a reminder that automated coverage figures are only as good as the detection behind them.

SEO agency monthly reports need a "brand mentioned" column

Start with the searches that already bring you enquiries at the decision stage: "[your brand] vs [competitor]", "[competitor] alternatives", "best [your category] for [your customer]", "[your product] pricing" and "[your brand] review". Search them from Australia, on a phone and on a desktop, and note three things: is there an AI Overview, is your brand named in it, and which sites does it cite.

Semrush's recommendations follow from that audit. Track AI Overview mentions and citations alongside rankings. Map the review platforms, roundups, videos and community threads Google is pulling from. Keep comparison and pricing pages current and structured so a clean passage can be lifted, because stale details give the summary room to misdescribe you. Ask customers for reviews on the platforms that get cited, within each platform's rules. And consider the YouTube channels your buyers already watch, since video reviews show up as cited sources.

Most of this is what we described in our guide to answer engine optimisation: make your pages, and the pages that describe you, the ones an answer engine can quote accurately. The Semrush study puts a number on how urgent that has become for the keywords that lead to a purchase.

The next step is to pull your commercial keyword list from Search Console or your rank tracker, filter it to the five modifiers above, and add a column for "brand mentioned in AI Overview: yes or no". If your SEO agency's monthly report has no equivalent of that column, ask why.