Why Your Brand Doesn't Appear In AI Overviews: Ahrefs Explains

You performed a search for your content on Google's AI Overviews, but your content isn't there. You rank in the top three for the keyword you searched for. You've earned backlinks, optimized your headers, updated the content. Still nothing. While a lot of brands out there are scratching their heads, wondering what's going on, Ahrefs just published the data that explains what's happening. The picture is more disruptive than most practitioners have been willing to say out loud. The connection between ranking well on Google and appearing in AI Overviews has broken down over a gradual period of time.

The Overlap Between Google Rankings and AI Citations Has Collapsed

In March 2026, Ahrefs researchers Louise Linehan and Xibeijia Guan published an analysis of 863,000 keyword SERPs containing 4 million AI Overview URLs, one of the largest datasets anyone has applied to this question. Here's what they found, and the number here is staggering.

Only 37.9% of URLs cited in AI Overviews also appeared in the top 10 blocks of the corresponding Google search results.

That number already tells a story. But the comparison to where things stood six months ago makes it much sharper.

According to Search Engine Journal, in July 2025, roughly 76% of AI Overview citations came from the same pages ranking in Google's organic results. By March 2026, six months later, that figure had dropped to approximately 38%. The correlation between ranking and getting cited cut roughly in half. Run those numbers forward and the direction is clear: AI Overviews are pulling from an increasingly separate pool of content than traditional organic search.

The breakdown within that 37.9% is worth looking at too.

• 37.1% of cited URLs rank in the organic top 10

• 26.2% rank somewhere in positions 11–100

• 36.7% don't rank in Google's top 100 for that query at all

More than one in three AI Overview citations come from a page that wouldn't appear on the first five pages of a standard Google search. Traditional SEO metrics do not predict this outcome. They can't. But what it can do is make you think differently about how your content is sourced, indexed, and laid out for the AI Overview.

The Mechanism: Query Fan-Out

The reason AI Overviews source content from outside the top 10 comes down to a process Google calls query fan-out.

When you run a search, Google's AI doesn't just retrieve documents for your exact query. It expands your query into a set of sub-queries, each targeting a more specific angle, related concept, or component of your original question. Those sub-queries run separately, pulling from wherever Google finds the best answer to each one. The key point: each sub-query has its own source set. The pages that rank well for your head term don't automatically rank well for the sub-queries the AI generates from it. The AI Overview you see at the top of the page is a synthesis of those multiple sub-query results, not a summary of what ranked for your original search.

Ethan Lazuk, who has published research on fan-out optimization, frames this clearly: "the unit of AI optimization is not the keyword. It's the sub-query cluster behind the keyword."

What does this mean practically? Take a query like "how to reduce customer churn in SaaS." Fan-out might generate sub-queries about onboarding, customer success workflows, product-led retention metrics, churn prediction models, and pricing psychology. Each of those sub-queries may pull from different domains. A page that ranks first for the parent keyword might address only the head-level framing, and get bypassed entirely in favor of a more specific page that directly answers one of the sub-queries.

Ranking #1 is not a guarantee to become a part of the AI Overview. Your content has to be all about being the answer.

YouTube Is the Bull in the China Shop

A lot of folks sleep on just how good YouTube is as a search engine. Ahrefs just put that on notice hard and YouTube might upend what you think you know you about traditional search.

YouTube now accounts for 5.6% of all AI Overview citations, and among pages that don't rank in Google's traditional search results, YouTube represents 18.2% of citations. Over the six months of the study, YouTube's share of AI Overview citations grew by 34%, making it the most-cited domain in AI Overviews.

A video platform that doesn't produce text-based web pages is one of the primary sources Google's AI pulls from when constructing written answers.

This reflects something structural about how Google weights its own platforms. YouTube content doesn't compete in blue-link SERPs in the same way an article does, but it does answer questions. Google's AI cites it heavily because the trust signal and entity authority associated with YouTube are already baked into the system.

For content teams, this is not a soft consideration. If your topic area can be addressed through a well-structured video (an explanation, a walkthrough, or a demonstration), YouTube is now a legitimate citation channel. Not a traffic play. A citation play. The distinction matters when you're building a GEO strategy.

What This Means for Your Content Strategy

The Ahrefs data demands three strategic shifts. None of them require scrapping your existing SEO work. Traditional SEO, according to Google's new GEO guide, should be considered more of a foundational thing, not just something you do intermittently.

1. Build for sub-queries, not just head terms

If AI Overviews are synthesizing answers from fan-out sub-queries, your content needs to address those sub-queries explicitly. This means going deeper on component topics within a subject area, not just optimizing a single comprehensive page for a head keyword. A cluster of focused, specific pages that each fully answer one sub-question will get cited more consistently than a single thorough guide.

2. Treat YouTube as a citation asset

If your marketing or content team has video assets, those should be structured and published on YouTube with the same optimization discipline you apply to written content. Descriptions, titles, transcripts, and timestamps that directly answer specific questions create citation surface area in AI Overviews. This is not about driving YouTube views. It's about being present in the citation pool Google draws from. Think about the content flywheel we always talk about here. Creating blog content, which can tie into video content, podcast content, the works. Feed them into one another, and you create that ecosystem

3. Organic Ranking Data Isn't the Only Meter Stick

The 37.9% overlap figure means that organic ranking position isn't the only weak predictor of AI Overview citation. A team reporting organic rankings as a proxy for AI visibility is not getting the full picture. This is where the lack of a standardized AI visibility measurement framework is genuinely costly, if you're not actively tracking where you appear in AI-generated answers, you have no baseline, no trend line, and no way to prove to stakeholders that your GEO work is producing results.

How to Check Whether Your Content Is Getting Cited

Before adjusting strategy, you need a baseline. Here's where to start.

Manual audit: Take the 10–20 keywords most important to your business. Run each one in Google while logged out (or in a private window). Note whether an AI Overview appears, which domains are cited, and whether your domain is among them. Do this weekly for four weeks to track changes. If you have an active Google Search Console, this is a great piece of data to have. Use the "queries" in GSC and you can see your position in Google. Note that even if you're in position 1, it's not a guarantee that you're going to be seen in the Overview.

Tool-assisted tracking: Ahrefs Brand Radar, the same tool used in the study, lets you track AI Overview citation presence at scale. Otterly.ai tracks brand mentions across ChatGPT, Perplexity, and AI Overviews. Semrush has added AI Visibility tracking to its toolkit. We at Xponent21 also have a tool, known as CARL Intelligence, which we'll be rolling out pretty soon.  

What to watch for: Citation doesn't always mean visibility. Growth Memo's April 2026 analysis found that 61.7% of LLM citations are "ghost citations”, the domain is linked as a source but the brand name never appears in the answer text. If you're being cited but not named, you're generating AI trust signals but zero brand awareness. That distinction shapes how you measure and what you optimize for.

Traditional SEO/AEO/GEO All Want The Same Thing

What does AI search and Allen Iverson have in common? It's all about being The Answer.

The implication of the Ahrefs data is broader than any single content strategy decision.

The assumption that has guided most content teams for the past year: rank well, and the AI visibility will follow. It's almost like a case of the tech bro Pied Piper. You've seen from all these YouTube videos and Reddit threads that if you ranked high in search, the Overview was a foregone conclusion. And the 76% figure from July 2025 gave people enough reinforcement to feel that the assumption was true. It seems indeed that it's not the case at all.

AI Overviews are operating on a different ranking system than traditional Google search. The signals that predict AI citation: sub-query relevance, YouTube presence, third-party entity authority, content structure for extraction. They overlap with traditional SEO signals but are not the same as them. Teams treating them as identical are likely to keep producing content that ranks well and doesn't appear in AI answers.

The practitioners building AI visibility right now are running a separate playbook. Not a replacement for SEO, but a parallel track instead. One that starts with understanding how AI reformulates queries, maps the sub-query landscape behind your important head terms, and tracks citation presence as a distinct metric from ranking.

If your team doesn't have that playbook yet, this data is a reasonable starting point for the conversation about why you need one.

Discover AIO is not only a learning platform, but we're a growing community hub. If you're looking to become a part of a community of marketers who are looking to gain an edge in the Age of AI, join us.

We also have out first webinar coming to you guys live on June 4th, 2026 at 10am EST, where we'll be covering the State of Search, tools that Xponent21 is building, and a Q&A with Will Melton, CEO of Xponent 21. If you're interested in registering, click here.