The essays / AI search and SEO
What is documented.
And what is not.
Essays on how search engines and AI answer engines find, choose and cite what is on the web.
-
Answer engines · 6 minute read
How AI answer engines pick and cite sources: what is documented and what is not
What Google, OpenAI, Perplexity, Anthropic and Microsoft have documented about choosing and citing sources. And where the record stops.
-
Prompt wording · 8 minute read
Your AI visibility score depends on how the question is asked
In one LLM Scout data set, ChatGPT named the brand in 30.8% of answers to how-to prompts and in 59.3% of answers to compare prompts. But 55% of the compare prompts contained the brand's name. A visibility score needs its prompt mix reported.
Each essay is built on primary sources and ends with a list of them. One of them, on prompt wording, reports figures from LLM Scout, which I founded. It says so at the top. New essays are also published to the RSS feed. The editorial policy sets out how they are researched, reviewed and corrected. The glossary defines the terms they rely on.