retrieval
7 posts
Why your own pages compete for the same AI answer
Cannibalisation moved from pages to passages. Why a second article on a topic can lower visibility, and how to consolidate at section level instead of merging.
Does AI search visibility transfer across languages?
Model knowledge crosses languages. Retrieval does not. Why your English citations do not follow you into German, and how to pick markets by pool thinness.
How to audit which pages AI engines can actually retrieve
Most AI visibility problems are retrieval problems. Three gates to test in an afternoon: can a crawler fetch it, is your text there, is any passage usable.
How much of your page does an AI engine actually read?
Engines read passages, not pages. What survives crawling, chunking and retrieval, why position in context matters, and how to write sections that stand alone.
Why AI engines get your pricing wrong
Pricing is the fact AI answers get wrong most often. Why tables flatten badly, why client-side prices are invisible, and how to write a price that survives.
what are multi-hop questions in llm search patterns
Multi-hop questions need information combined from several sources across steps. AI answers them by retrieving and stitching sources together.
API testing SEO citations in RAG frameworks
AI answer engines are RAG systems, so testing whether they cite you is a retrieval problem. Here is how to test SEO citations via engine APIs and a local RAG harness.