Skip to content

Contact Website Log In

Documentation Agencies MCP Server API Reference Research

Research notes›Writing in reranker-friendly chunks is mostly a myth

Documentation Agencies MCP Server API Reference Research

Research notes

AI bot types — training, indexing, and retrieval bots do different jobs The real risk of low-quality AI-generated content AI Overviews by country — 76% in the EU, 90% outside, 0% in France AI Overviews by query intent — commercial queries trigger it too AI Overviews by query length — longer queries trigger it more How often Google AI Overviews appears — 87% of searches AI agents read your image alt text as content How to write a paragraph AI will cite — answer shape wins AI assistants judge companies by their brand name ChatGPT fan-out targeting — domains persist, URLs churn ChatGPT fan-outs doubled in length in four months ChatGPT vs Google — the real search market share ChatGPT has its own search index — the Labrador family ChatGPT caches pages instead of fetching them live Where ChatGPT search results come from — Google, Bing, Web IQ and 8+ providers ChatGPT searches in English even when users don't ChatGPT Shopping runs on OpenAI's own product index — live A/B experiments ChatGPT site-operator fan-outs — 84% branded, Reddit tops the rest Citation rate benchmarks differ sharply by engine Citation rate by prompt intent — informational and commercial win Content types cited most by engine — listicles, product and category pages Deep Research's reading budget — navigation crowds out your content DeepSeek shared chat links are indexed by Google Fan-outs per prompt by engine — Perplexity 1.4, ChatGPT 2.1, Grok 6.8 Losing Google rankings kills AI search visibility too GPT-5.6 retrieves product pages over listicles GPT-5.6 retrieval changes — more iterations, double the sources How AI search chooses passages — retrieval, reranking, generation How ChatGPT Deep Research reads your site — Bing, three commands, no clicks Kimi's content strategy — retrieved often, recommended rarely Listicle rank effect — position in cited listicles shapes AI visibility Listicle strategy by engine — few placements beat many Most-cited domains by engine — top 5 sources per AI platform Most-cited domains in AI search — Reddit, YouTube, LinkedIn lead Optimizing for Google AI Overviews and AI Mode vs ChatGPT — 5 differences Prompt style changes how many brands AI answers show Rewording of a prompt rarely changes AI visibility until cosine similarity drops Being read is not being cited — AI agents read far more than they cite Reddit in AI search — the most-cited domain Translated Reddit is winning AI citations on Google's surfaces Writing in reranker-friendly chunks is mostly a myth Retrieval vs citation — the difference and why it matters Review sites in AI search — Yelp, G2, Trustpilot shape recommendations Self-promotional listicles still get cited — about 1 in 10 citations Server logs for AI search — diagnose crawled-but-not-cited pages Shared URLs and leaked chats in AI retrieval — rare but real What is a query fan-out (QFO) in AI search?What is AEO — Answer Engine Optimization, and how it differs from SEO Why listicles get cited by AI search — answer shape, not list bias Wikipedia in AI search — influences models on two levels Words ChatGPT secretly adds to fan-out searches — best, top, reviews YouTube in AI search — #2 most-cited domain, transcripts included

Research notes

Writing in reranker-friendly chunks is mostly a myth

Coming soon

Coming soon. This feature or research guide is not currently available from Kept AI; no release date is announced.

← PreviousTranslated Reddit is winning AI citations on Google's surfaces Next →Retrieval vs citation — the difference and why it matters

WebsiteKept AI documentation

On this page

Search documentation

Search titles, guides, and API references.

↑ ↓ Navigate ↵ Open