When someone asks Claude, ChatGPT, Gemini or Perplexity to recommend a provider in your category, one name comes back first. This checks whether it is yours — and if not, whose it is.
A ranking is a position in a list of links, and the person still chooses which to click. An AI answer has no second page and nothing to scan. The assistant names two or three providers and the conversation moves on. If you are not in that sentence, you were not considered.
What earns a mention is different from what earns a ranking: self-contained explanations a model can lift without context, explicit decision guidance, named tools and prices, and structured data that states plainly what you do and who you serve. That is the work behind programmatic SEO and the answer-first restructuring we build into web development engagements.
The share of answers that named your business. We send the same buyer-intent questions to each available assistant and record whether you appear. Where you do not, we list who did.
A growing share of buying research happens inside an AI conversation rather than on a results page. There is no second page in a conversation. Either the assistant names you or the buyer never learns you exist.
They are pattern-matched out of the answer text, so treat them as a starting point rather than a definitive list. Names that recur across several answers are the meaningful ones.
Answer-engine work is built into fixed-price engagements from $999 to $15,000, delivered in two to twenty-four weeks depending on scope. Price is agreed before work starts and is never billed hourly.
