The fan-out effect REPORT

What happens between a query and a citation?

New research from 17k queries and 354k pages reveals what actually gets your content cited by ChatGPT and what doesn't. From retrieval rank and heading structure to freshness and domain authority, find out what matters and what doesn't.

What you'll walk away with

Rank is the dominant signal in AI search

A page at position 0 in ChatGPT's search results has a 58% chance of being cited. By position 10, that drops to 14%. A 4x gap driven by rank alone and even strong content can't overcome a weak position.

Your headings decide if AI cites you

Pages with headings that closely match the user's query are cited 41% of the time vs. 29% for weak matches. The relationship is monotonic: better heading match, more citations. Headings are the lever most teams are ignoring.

Focused pages beat comprehensive guides

Pages covering 26-50% of ChatGPT's fanout sub-queries get cited more than pages covering 100%. Be the best answer to one question, not an adequate answer to five.

Schema markup is still your hidden advantage

Pages with JSON-LD structured data are cited 38.5% of the time vs. 32.0% without. A +6.5pp advantage that holds regardless of word count, headings, or domain authority.

Long-form is losing. College-level writing is winning.

The sweet spot is 500-2,000 words. Pages over 5,000 words get cited less than pages under 500. Meanwhile, college-level writing (Flesch-Kincaid 16-17) is cited 21% more often than simple writing.

"We were able to focus our energy on the highest-value initiatives for the business in a matter of weeks."

Nick Fairbairn
SVP of Growth, Chime

“We set out to change the narrative around Sanity in AI search, not just show up more often. Staying the brand the models name most, and climbing in the categories that matter, is exactly the signal we were after, and we can see it happening week to week.”

Richard Lawrence
Principal, Agentic Operations & Growth Marketing, Sanity

“If we wanted to 10x the growth we were starting to see, we needed a system that handled SEO, AEO, content production, all of it. AirOps was the first to actually deliver results.”

Shani Mithani
Head of Growth, Venn

Built for the teams
that own search

Content and SEO leaders who need to adapt strategy for AI searchkly ChatGPT Ads are taking hold

Growth and demand gen leaders building pipeline from new channels

Marketing agencies advising clients on  AI visibility

CMOs and VPs evaluating how AI search impacts their brand

Your competitors
are already adapting.
Are you?

Join the brands using this playbook to gain  AI search visibility. Get the framework, the data, and the 90-day action plan.

Frequently Asked Questions

What is the fanout effect?

When a user sends a query to ChatGPT, the model doesn't just answer from memory. It issues its own web searches, called fanout queries, to gather information, then selects which pages to cite. This study maps what happens in that gap between the original query and the final citation, across 16,851 queries and 353,799 pages.

What are the biggest findings?

Retrieval rank is the single strongest predictor of citation. A page at position 0 has a 58% chance of being cited vs. 14% at position 10. Among content signals, heading match to the original query is what matters most, not broad topical coverage. Domain authority and backlinks show no positive correlation with citation. And the content sweet spot is 500 to 2,000 words at a college reading level, refreshed every 30 to 89 days.

Who should read this?

Content, SEO, and growth leaders who need to understand what drives visibility in AI search. If you're making decisions about content strategy, page structure, or refresh cadence, this research gives you the data to prioritize what actually moves citation rates.

Does traditional SEO still matter?

Some signals carry over. Structured data, clean heading hierarchy, and page freshness all help. But several pillars of traditional SEO don't translate. Domain authority and backlinks show no positive correlation with AI citation, and long-form "ultimate guides" underperform focused, well-matched pages. The playbook is shifting.

How is this different from other AI search research?

Scale and specificity. This study analyzed 16,851 unique queries across 10 industries and 4 query types, with each query sent to ChatGPT 3 separate times. It goes beyond "what gets cited" to map the full retrieval pipeline: fanout behavior, retrieval rank, heading similarity, content structure, schema usage, readability, and freshness, with controlled comparisons at every step.