Here's a stat that should change how you think about SEO.

Only 38% of pages cited in Google's AI Overviews also rank in the traditional top 10 search results.

Seven months ago, that figure was 76%. Google's AI is rapidly expanding the pool of sources it draws from, and the websites it cites look less and less like the websites that dominate the blue links below.

38%Of AI-cited pages also rank top 10Ahrefs 2026
76%Same figure just 7 months earlierAhrefs 2025
62%Decline in citation-ranking overlapALM Corp analysis
AI citation overlap with traditional top-10 rankings: shrinking fast % of AI-cited pages that also appear in Google's organic top 10
Jul 2025
76%
Oct 2025
62%
Jan 2026
50%
Mar 2026
38%

Source: Ahrefs / ALM Corp citation overlap analysis. The trend is clear: AI citation and traditional ranking are becoming separate games.

What this means in practice

If your SEO strategy is built entirely around getting to page one, you're playing by rules that are becoming less relevant by the month. The sites being cited in AI Overviews share a different set of qualities than the sites ranking in the top 10.

They tend to have structured data that helps AI systems understand the content. They have topical depth, covering subjects thoroughly rather than skimming the surface. And they use clear, direct language that AI models can extract and paraphrase confidently.

None of this is about keyword density or backlink volume. It's about making your content machine-readable and authoritative.

What traditional SEO rewardsWhat AI citation rewards
High domain authoritySpecific, detailed answers
Many backlinksClear, structured content
Keyword densityTopical depth and expertise
Page-one rankingMachine-readable schema markup
Site ageRecency and freshness
Site speed (basic)AI-parseable architecture

The opportunity for small businesses

This is actually good news if you run a small business. The old SEO game was expensive. Competing for page-one rankings against businesses with bigger budgets and more content was an uphill battle for most SMEs.

AI citation doesn't work the same way. Google's AI doesn't rank sources by domain authority the way traditional search does. It looks for the most relevant, most structured, most clearly authoritative answer to a specific question. A well-built page from a small business can be cited alongside, or instead of, content from a major publisher.

💡
The equaliser: A small design studio in Brisbane with a well-structured FAQ page, proper schema markup, and specific service content can be cited by Google's AI on the same query where a major agency is also cited. Domain authority and ad spend don't decide AI citations. Content quality and structure do.

What you can do right now

Your AI visibility quick-start checklist

Search your main service keywords in Google — do you appear in any AI Overview?
Search the same keywords in Perplexity and ChatGPT — are you mentioned?
Check your site for FAQ schema markup (use Google's Rich Results Test)
Check for Article schema on any blog content
Audit your service pages: do they answer specific questions in detail, or just describe services broadly?
Check that your llms.txt permits AI crawlers at your domain root

Where GEO fits in

This is the problem Generative Engine Optimisation (GEO) is designed to solve. Where traditional SEO focuses on ranking in search results, GEO focuses on getting your business cited by AI systems. It's a different set of practices: structured data, topical authority, content architecture that AI crawlers can parse, and ongoing monitoring of how your business appears across AI platforms.

At Sonder Digital, we build GEO into every website from day one. If your site was built before AI Overviews existed, it almost certainly isn't optimised for them.

Book a free discovery call

Sources

  1. Ahrefs / ALM Corp: AI Overview citation overlap analysis (2026)
  2. Semrush: AI Overview tracking data (2025-2026)