Fan-out query analysis: ChatGPT, AI Overview and AI Mode
ChatGPT, Google’s AI Overview and AI Mode, given the same prompt, answer differently, and search differently. A pilot study of the queries behind the answers. ChatGPT runs a single long web search query, mostly reusing the prompt’s own wording, and cites 4.1 pages. Each Google surface runs three or four short head terms and cites 13–17. Ranking explains little of what any of them cite: 10% of cited pages are a top-10 result for the primary query, and the fan-out’s own searches don’t account for most of the rest. YouTube is the most-cited domain on both Google surfaces. AI Mode and AI Overview share just 58 of the 770 URLs they cite between them, and just four URLs were cited by all three.
Latest from the field
Browse the AI SEO Blog (45) →Fan-out query analysis: ChatGPT, AI Overview and AI Mode
ChatGPT, Google’s AI Overview and AI Mode, given the same prompt, answer differently, and search differently. A pilot study of the queries behind the answers. ChatGPT runs a single long web search query, mostly reusing the prompt’s own wording, and cites 4.1 pages. Each Google surface runs three or four short head terms and cites 13–17. Ranking explains little of what any of them cite: 10% of cited pages are a top-10 result for the primary query, and the fan-out’s own searches don’t account for most of the rest. YouTube is the most-cited domain on both Google surfaces. AI Mode and AI Overview share just 58 of the 770 URLs they cite between them, and just four URLs were cited by all three.
Site Architecture for AI Crawlers: Optimize for Search & Human UX
Site architecture benefits both AI search crawlers and human visitors by utilising a pillar-cluster structure, clean URLs and clearly contextual links. This content structure is important for searchability through AI and also for citability.
Claude Watermark: Anthropic’s AI Detection Explained
Created by Anthropic, the Claude watermark embeds a statistical signal in content. This mark lets AI search and publishers identify AI-generated works as a separate type from the rest.
Content Decay in AI Search: Why Your Old Posts Disappear Faster Now
Content freshness is a major priority of AI search engines, which means decaying and visibility loss is sometimes faster for older posts. Quarterly refreshes are crucial to ensure AI citations.
Most read
Most read on the AI SEO Blog →- 01 5 min read
- 02 4 min read
- 03 5 min read
- 04 64 min read
- 05 4 min read
Topics
All topics →AI SEO Agent
Research
View all →Fan-out query analysis: ChatGPT, AI Overview and AI Mode
ChatGPT, Google’s AI Overview and AI Mode, given the same prompt, answer differently, and search differently. A pilot study of the queries behind the answers. ChatGPT runs a single long web search query, mostly reusing the prompt’s own wording, and cites 4.1 pages. Each Google surface runs three or four short head terms and cites 13–17. Ranking explains little of what any of them cite: 10% of cited pages are a top-10 result for the primary query, and the fan-out’s own searches don’t account for most of the rest. YouTube is the most-cited domain on both Google surfaces. AI Mode and AI Overview share just 58 of the 770 URLs they cite between them, and just four URLs were cited by all three.
ChatGPT web search uses Google Search and NOT Bing Search (with proof)
TL;DR: Why this is a big deal? 1. Impact on SEO This means the whole SEO / Answer Engine Optimization (AEO) narrative has come full circle from “Google is dead” -> “rank in Google to show up on ChatGPT”! 2. Shady Practice on OpenAI’s part to route large volumes of private data without explicit declaration […]
How does ChatGPT Search the Web: Data-Driven Investigation
– Abhishek Iyer (@distantgradient), o3, Opus, ACME.BOT TLDR takeaways: I analyzed 50 real ChatGPT conversations by intercepting network traffic to uncover the patterns behind when and how ChatGPT searches the web. With that TLDR out of the way, let’s take a step back and try to understand what is happening under the hood. The Three […]
Try ACME.BOT for free today!
Automate your business processes with the power of AI agents.
Sign Up For Free TrialInsights & Guides
View all →Site Architecture for AI Crawlers: Optimize for Search & Human UX
Site architecture benefits both AI search crawlers and human visitors by utilising a pillar-cluster structure, clean URLs and clearly contextual links. This content structure is important for searchability through AI and also for citability.
Claude Watermark: Anthropic’s AI Detection Explained
Created by Anthropic, the Claude watermark embeds a statistical signal in content. This mark lets AI search and publishers identify AI-generated works as a separate type from the rest.
Content Decay in AI Search: Why Your Old Posts Disappear Faster Now
Content freshness is a major priority of AI search engines, which means decaying and visibility loss is sometimes faster for older posts. Quarterly refreshes are crucial to ensure AI citations.
Audit Your Content Portfolio for AI Citability
AI search is revolutionizing the discovery process, and if you want to continue driving traffic, you need to consider the citability of your content. Audit your content for uniqueness, SEO, query coverage, and section clarity for AI citation.
Structured Data for AI Search: What Still Works Post-FAQ Rich Results
Google removed FAQ rich results. While AI search is a visibility driver, structured data has limited impact on AI citations. Instead, topical depth and a well-structured content piece might result in AI visibility.
AI Search Visibility Metrics: Track Beyond Traditional SEO
As AI search becomes more prevalent, new metrics are needed to measure AI search visibility beyond SEO rankings. Instead of relying on unreliable citation counts, track impressions for AI overviews, visibility for fan-out queries, and conversions.