Your number one ranking now sits under an answer Zapier helped write

On October 21, Semrush published the number that has been in every SEOSEOSearch Engine Optimization: the practice of improving your pages' natural (unpaid) rankings in search engine results pages to attract more organic traffic.View full definition → deck since. Semrush sells SEO software and now calls itself an Adobe company, so treat its numbers as vendor data. It tracked 400 US commercial keywords for 12 months, and Google showed an AI Overview for 88% of them in August 2026, up from 45.75% in September 2025, and for 97.08% when the keywords were weighted by search volume. The fastest growth came where buyers make up their minds. "Best X for Y" queries reached 96.12% coverage, up 53 percentage points, and review queries climbed from 30% to nearly 90%, the biggest increase of any modifier. Semrush's own conclusion: "ranking number one likely won't lead to as much traffic as it did this time last year."

Most people read this as "rankings are dead, citations are the new position one." I think that reading is half right, and the wrong half will cost money. Ranking now gets you into the room. It no longer closes the deal. And on commercial queries, the sentence Google writes about your product often quotes someone else. A CMO who moves budget out of SEO and into AI-visibility dashboards will miss both points. The money belongs in two places: protecting the rankings that make you eligible to be cited, and owning the third-party sources Google quotes when it describes you.

The citation-first consensus on AI Overviews, stated fairly

The consensus has a strong case, and it rests on more than one vendor's study.

Click loss is well documented. Ahrefs, which also sells SEO tools, found that AI Overviews now result in a 58% lower average clickthrough rateclickthrough rateClick-Through Rate (CTR) is the percentage of people who click a link, ad, or call to action out of those who viewed it.View full definition → for top-ranking pages, up from 34.5% just eight months earlier, and that pages in position two lost about half of their clicks, while even pages ranking tenth saw drops of nearly 20%. Pew Research Center's independent browsing panel from 2025 points the same way: visits ended in a click 8% of the time with an AI summary, versus 15% without.

Citation also clearly pays. Seer Interactive is a search agency, so it has a stake in these findings, but its dataset is large: 53 brands, 5.47 million queries and 2.43 billion impressionsimpressionsThe total number of times an ad or piece of content is displayed, regardless of clicks. Each display counts as one impression, even to the same person.View full definition → from January 2025 to February 2026. It found roughly 3.3% CTRCTRClick-Through Rate (CTR) is the percentage of people who click a link, ad, or call to action out of those who viewed it.View full definition → with no AI Overview, 2.1% when an AI Overview cites the page, and 0.9% when it doesn't.

The legal exit has closed. A US judge dismissed the lawsuit Penske Media brought against Google over traffic and revenue lost to AI Overviews. Penske had argued it was unable to block Google from using its content in AI Overviews while still appearing normally in search results, and Judge Mehta concluded that Penske had not plausibly alleged that Google Search and AI Overviews are separate products under antitrust law. If the owner of Variety and Rolling Stone can't opt out, neither can your brand.

Boards have noticed too. According to Forrester's Marketing Survey, 2026, executives or boards are questioning 49% of respondents about how their brand or content is represented in AI-powered search results and genAI tools.

So the consensus recommendation follows: track citations alongside rankings, restructure content so it can be extracted, and buy tools that do both. That's coherent. It's also incomplete in four ways.

Does ranking first still matter on commercial keywords?

Yes. It qualifies you to be cited, and the citation is what earns the click. Treating the two as substitutes is the main error in the consensus, and the evidence for it sits inside the same studies people quote.

The 88% counts appearances, not lost revenue

PPCPPCPay Per Click (PPC) is a digital advertising model where you pay only when a user clicks your ad, not when it is merely displayed.View full definition → Land's review of the Semrush study points out that it measures appearance only, with no click or traffic data, on US desktop, with fewer than 27 keywords per industry and modifier cell, and that Semrush sells the tools its recommendations call for. That doesn't make the trend false. It means the 88% tells you how often the box shows up. It doesn't tell you what the box costs you.

The second-order effects are more interesting. In Seer's data, searches without AI Overviews are getting more valuable, with CTR on those queries rising from 2.8% in early 2025 to 3.8% by February 2026. Paid search held up too: when Google showed an AI Overview, paid CTR rose slightly from 14.6% to 16.2%, and when it didn't, CTR fell from 26% to 21.8%. Click loss on AI Overview queries also wasn't a straight line down: after bottoming out at 1.3% in December 2025, CTR on AI Overviews climbed to 2.4% in February 2026.

Ranking is how you get cited

According to Google's guidance as summarised by one agency, a page must be indexed and eligible for a snippet in Google Search to appear in AI Overviews, and no special markup or files are needed. Even HubSpot, which sells an AEOAEOThe practice of making your brand and content visible and citable inside AI-generated answers from tools like ChatGPT, Gemini and Perplexity.View full definition → product (so it gains when you buy one), notes that pages cited in AI Overviews tend to have strong topical coverage, clear authorship signals, and real referring domains pointing to them. That's a description of a page that ranks. The work that earns a ranking is the same work that earns a citation. Cut the SEO budget to fund "answer engine optimizationanswer engine optimizationThe practice of making your brand and content visible and citable inside AI-generated answers from tools like ChatGPT, Gemini and Perplexity.View full definition →" and you remove the thing that qualifies you.

The source Google quotes about you is often someone else's page

This is the blind spot that matters most. Semrush's own worked example is the query "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, while a YouTube video and a second third-party site make up the rest of the prominent citations. Klaviyo wrote its own case. Mailchimp's case was written by a software directory.

That pattern is structural. Semrush's 2026 AI Visibility Index found that between January and April 2026, Google's AI Overviews cited YouTube 83.2 million times, with Reddit next at 35.8 million and Wikipedia at 32.4 million. An earlier Semrush analysis found Zapier ranks as the #1 cited source in digital technology but only #44 in brand mentions. So "be the source" can't just mean "make our own domain citable." Often the source is a review platform, a creator or a comparison publisher. Forrester's 2027 content predictions push in the same direction: brands will shift investment toward earned influence, creator ecosystems, and evidence-backed claims.

Your measurement is breaking at the same time

Teams are being told to manage a new metric just as the old ones stop being reliable. Ahrefs says Google is returning altered search results to suspected search-data collectors, with 5% to 20% of results affected depending on the provider. Semrush confirmed that Google is returning inaccurate search results to some automated ranking tools, and Google has not confirmed that it deliberately serves altered results. Citation trackers built on scraped results pages carry the same risk. Analytics doesn't help much either: clicks from AI Overviews and AI Mode usually carry a google.com referrer and show up as Google organic.

ChatGPT is a rounding error in traffic for now

Much of the "answer engine" budget is going to ChatGPT visibility. SEO consultant Marie Haynes analysed a year of GA4 data and found that ChatGPT referred only 0.62% as many visits as Google organic search across 100 websites, covering more than 145 million sessions, rising to 0.73% when every AI chatbot was counted. ChatGPT visitors converted at 13.8% versus 3.02% for Google organic visitors on the 55 sites that tracked conversions, but Google organic still produced 27 times more conversions than all AI chatbots combined. That sample undercounts AI traffic that lands as "direct." Even so, for commercial demand in 2026, the answer engine that matters most is Google's own.

A citation-share rule for commercial keywords

Here's the rule to put on the whiteboard:rank to qualify, get cited to earn the click, own the description to win the deal. Spend where the sentence about you is written.

Apply it to each money keyword with three questions, and put the next dollar on the first "no".

• Do we rank on page one for this query? If not, it's a classic SEO problem, and no AEO tool will fix it.
• If we rank, does the AI Overview cite our page? If not, it's an extraction problem. Prices, comparison criteria and "who this is for" statements need to be in plain, current text on the page.
• If we're cited, whose words describe us? If it's Zapier, a G2-style review page, a Reddit thread or a YouTube reviewer, the next dollar goes to that source: review volume, creator relationships, a briefing for the comparison publisher. A new blog post on your own domain won't change that sentence.

Most teams fund question one by habit and question two because vendors sell dashboards for it. Question three is the one that decides Mailchimp's half of the Klaviyo comparison, and almost nobody has a budget line for it.

Worked example: what citation is worth

This is a hypothetical, arithmetic only. It applies Seer's CTRs and Haynes's conversion rates to an imaginary cluster of 100,000 monthly impressions on "brand vs brand" queries. The two studies use different samples, so read the output as orders of magnitude.

• With no AI Overview, 3.3% CTR gives 3,300 clicks. At 3.02% conversion, that's about 100 conversions.
• With an AI Overview that cites you, 2.1% CTR gives 2,100 clicks and about 63 conversions.
• With an AI Overview that doesn't cite you, 0.9% CTR gives 900 clicks and about 27 conversions.

Getting cited wins back about 36 of the 73 conversions an uncited AI Overview takes away. That's roughly half. The other half is gone whatever you do. Now add ChatGPT: at Haynes's 0.62% ratio, the cited scenario implies about 13 ChatGPT sessions, and at 13.8% conversion, about 2 conversions. For a CMO setting budgets, the order of priorities follows. Stay cited on Google first. Make sure the cited sentence is favourable second. Treat chatbot visibility as a brand-monitoring task, sized to its traffic.

There's a paid-search corollary too. If paid CTR held steady or rose on AI Overview queries in Seer's data, then brand-defense bids on "alternatives" and "vs" queries deserve a fresh look in the same planning cycle. They shouldn't be the first thing cut to fund AEO software. Teams that need torebuild the content budget around money keywords should run that review with paid search in the room.

What to do on monday about AI overviews

• Export your commercial queries from Search Console and filter to the five Semrush modifiers: alternatives, best X for Y, price or cost, reviews, and brand vs brand. Keep the top 20 by revenue, not by traffic.
• For those 20, check each AI Overview by hand in a clean browser profile and screenshot it. Given the 5% to 20% contamination Ahrefs reports, don't trust scraped data alone for the baseline.
• Add three columns to the sheet: our rank, cited yes or no, and whose words describe us. Mark every row where a third party or a competitor writes your side of the comparison.
• Rewrite your pricing and comparison pages so prices, plan limits and "choose us if" criteria sit in plain text with a visible "last updated" date.
• Pick the three third-party domains cited most often across the 20 queries and give each an owner this week: a review-request programme, a creator outreach list, or a correction request to the comparison publisher.
• Before reacting to any ranking drop in a third-party tool, check it against Search Console. Sudden ranking changes in third-party tools may reflect inaccurate search results rather than actual changes in Google rankings.
• Create a custom GA4 channel group for AI referrers such as chatgpt.com, perplexity.ai, gemini.google.com and claude.ai, and write down that AI Overview clicks will still show up as Google organic. Startfixing how AI-influenced visits show up in your analytics before the board asks for the numbers.

The keyword that matters most next quarter is the one where Google describes your product using a reviewer's or a competitor's sentence. Find those sentences across your 20 money queries and assign an owner to each before you sign for another visibility dashboard.