Key takeaways
- Google itself says AI Overviews now appear on roughly 50% of US queries and reach 2 billion monthly users. Whatever prevalence estimate you prefer, the AI box sits above your position-1 listing often enough that it has effectively become the real position #1.
- The CTR damage is real and measurable: position-1 organic listings lose 34.5% to 58% of their clicks when an AI Overview is present, and Pew found users click an organic result only 8% of the time when the AI box shows.
- Being cited inside the AI Overview is the counterweight. Seer Interactive's analysis of 2.43 billion impressions found cited brands earn roughly 35% more organic clicks and 91% more paid clicks than uncited brands on the same query.
- What gets cited is shifting. Promptwatch's July 2026 data shows product pages overtook listicles as the most-cited content type in AI Overviews for the first time, and video citations roughly doubled since January.
- Google's own ecosystem dominates citations. YouTube and Reddit are the top two cited domains, and google.com itself has risen 6x in citation share since January 2026.
The AI box is the new first thing people read
For fifteen years, the SEO industry argued about position 1. Title tags, meta descriptions, CTR optimization, all of it aimed at the top blue link. That argument is mostly over now, because there's a new tenant above it.
Google said it plainly in February 2026: AI Overviews appear on roughly 50% of US queries and reach 2 billion monthly users globally. Third-party trackers land in the same neighborhood. BrightEdge's commercial tracker recorded about 48% prevalence in February 2026, up 58% year over year. Conductor's Q1 2026 benchmark of 21.9 million queries found 25.11%. Semrush's keyword panel peaked near 25% in July 2025 before dipping to 15.69% in November, then recovering. Xponent21 measured 70.48% of personalized US results as of July 6, 2026.
The numbers disagree with each other because methodology differs: personalized vs. non-personalized, commercial keyword sets vs. random samples, desktop vs. mobile. But the direction is unambiguous. In January 2025, AI Overviews appeared on 6.49% of queries. Fifteen months later, Google is confirming coverage that's nearly eight times higher.
A seoClarity study put it in positional terms: 87.6% of AI Overviews on US desktop SERPs render in position 1, above the first organic result. When the AI box shows up, your number-one ranking is visually the second thing on the page. That's why people in the industry have started calling AI Overviews "the real position #1." It's not a metaphor. It's a description of what the SERP actually looks like.
One caveat worth stating honestly: these prevalence numbers move fast, and they vary hugely by vertical. Healthcare queries trigger AI Overviews 88% of the time per BrightEdge, while e-commerce has collapsed to around 4%. If you're in travel or local services, your experience of "AI Overviews everywhere" is very different from someone in health content. The aggregate stats describe an average nobody actually lives.
What the click data actually shows
Here's where it gets uncomfortable. The AI box doesn't just sit above your listing. It takes your clicks.
| Study | What they measured | Result |
|---|---|---|
| Ahrefs (Dec 2025, 300K keywords) | Position-1 organic CTR with an AI Overview present | Drops 34.5% to 58% |
| Pew Research (Jul 2025, 68,879 searches) | Users clicking an organic result | 8% with AI Overview vs. 15% without (−47%) |
| Pew Research (Jul 2025) | Users ending their session after seeing an AI Overview | 26% vs. 16% without |
| Sistrix (Germany, Feb 2026) | Position-1 organic CTR | Falls from 27% to ~11% (~60% drop) |
| Seer Interactive (2026, 2.43B impressions) | Clicks per million impressions, uncited brands | 9,445 vs. 33,500 on queries without an AI Overview |
| BrightEdge (12-month analysis) | Total impressions vs. click-throughs since AIO launch | Impressions +49%, clicks −30% |
The Seer Interactive numbers deserve a second look because they capture the whole picture. Queries without an AI Overview generate 33,500 clicks per million impressions. On queries with one, brands cited inside the AI Overview get 20,743. Brands not cited get 9,445, less than a third of what the same query would have sent before AI Overviews existed. So the AI Overview doesn't just redistribute clicks. It splits the world into cited brands and everyone else, and the gap between those two groups is more than 120%.
There's a wrinkle in the doom narrative, though. Seer also found organic CTR on AI Overview queries bottomed out at 1.76% and rebounded to 2.4% by February 2026. And cited brands get a 91% paid CTR uplift on the same SERPs. The citation halo effect is real: being named inside the AI answer appears to build enough trust that people click your ads and your listings more, not less. Pew's finding that only 1% of users click links inside the AI Overview itself misses something, then. The citation works even when the click doesn't happen inside the box.
Who and what Google actually cites
If citations are the new ranking, the obvious question is what earns them. Promptwatch's first-party data, drawn from billions of analyzed citations, gives the clearest picture available.
The shift toward commercial content
The most surprising finding of 2026: product pages overtook listicles as the most-cited content type in AI Overviews. In Q1 2026, listicles averaged 26.1% of citations and product pages just 9.7%. By late July 2026, product pages led every single day, ending the month at 17.9% versus 16.2% for listicles. ChatGPT shows the same pattern, which suggests this is a broad AI search shift toward brand-owned commercial pages, not a Google quirk.
The practical implication is straightforward. For years, SEO teams treated product pages as conversion assets and poured editorial effort into listicles and guides. That division of labor is now backwards. Structured specs, clear pricing, and fresh availability data on product pages now drive citations directly.
Video is the other quiet winner. Video citations climbed from about 2.7% in January to over 6% by late July, roughly doubling. Most brands still have no competing video assets. That's an open lane.
Google cites itself, a lot
In June 2026, YouTube was the single most-cited domain in AI Overviews at 4.16% of all citations, Reddit second at 2.72%, and google.com third at 2.30%. The google.com figure is the striking one: it was 0.37% in January. That's a sixfold rise in five months. Google properties (YouTube plus google.com) now account for roughly 6.5% of all AI Overview citations, a concentration Promptwatch notes has no equivalent in ChatGPT's citation mix.
Search Engine Journal reported separately that 43% of AI Overview citations point back to Google properties. Whatever the exact figure, the direction is clear: Google's AI answers increasingly reward presence within Google's own ecosystem. Your YouTube channel, your Business Profile, your Merchant Center feed, and your help content are GEO assets now, not side projects.
More citation slots than ChatGPT
One genuinely useful nuance from Promptwatch's cross-engine data: Google AI Overviews cite roughly 10 sources per answer, about double ChatGPT's 5. If you're a newer or mid-authority domain trying to break into AI answers, AI Overviews is the more forgiving engine, with twice as many citation slots per response and classic SEO signals still carrying real weight. ChatGPT's citation mix is more concentrated and more volatile.
How to compete when position #1 moved
Get cited, then measure branded search
The uncomfortable truth from the Quora thread on this exact question: informational content that can be summarized in two sentences is mostly dead as a traffic strategy. The pages still pulling clicks are the ones AI can't compress. Detailed comparisons, opinionated takes, case studies with real numbers, product reviews with photos and context.
And the citation itself has long-tail value even without a click. When someone reads an AI Overview and wants to dig deeper, they search for the brand they remember being mentioned. Track branded search volume monthly. It's the metric that captures the value of citations that traditional analytics miss entirely.
Optimize product pages like content pages
Given the citation shift toward product pages, the highest-leverage move for most e-commerce and SaaS brands is treating product pages as citation targets: structured specifications, transparent pricing, availability data, comparison tables, and FAQ sections that answer the literal questions buyers ask. Google's own AI optimization guidance says to ignore "AEO/GEO hacks" like content chunking and unnecessary llms.txt files, and to prioritize non-commodity content with unique viewpoints. The boring fundamentals carry more weight than the hacks.
Build the Google ecosystem presence
With YouTube, Reddit, and google.com holding over 9% of combined citation share, the math favors brands with presence on those platforms. Mine the Reddit threads that AI Overviews cite in your category for content briefs. Invest in video. Keep your Business Profile and Merchant Center data clean and current.
Measure what matters
Google rolled out Search Console's Generative AI performance reports to all websites worldwide by August 31, 2026, showing impressions from AI Overviews, AI Mode, and Discover AI features. No click data yet, but it's the first first-party window into AI visibility Google has ever given site owners. Pair it with an AI visibility platform. Tools like Promptwatch track your citation share, which pages get cited, and how AI crawler behavior explains your visibility, going well beyond what Search Console offers.
For a broader look at the category, the GEO software directory at bestgeosoftware.com covers the full range of tracking and optimization platforms.
Where this leaves you
The position-1 listing you spent years earning still matters. But it's now the second thing on the page for half of US queries, and the data says the brands cited inside the AI Overview are the ones collecting the remaining clicks, the branded searches, and the paid CTR uplift. The strategy didn't die. It moved up the page, and it diversified across YouTube, Reddit, and your own product pages. The teams treating citation share as the new ranking metric are the ones whose traffic graphs don't look like everyone else's.