When the buyer asks an AI instead of a search engine

Buyers who once typed bandsaw blade suppliers into a search box are starting to ask an assistant the whole question: which blade cuts 60mm 304 stainless bar, and who sells one?

The assistant answers in prose and cites a handful of pages. There is no page two. Either an answer engine can read, extract and cite your page, or you are not in the conversation.

What actually changes

Less than the noise suggests. Google's own documentation says its AI features build on the established SEO fundamentals, and that a page needs nothing beyond being indexed and eligible for a snippet.

What shifts is the unit. A search engine ranks whole pages; an answer engine quotes passages. A section that answers one question in plain sentences, with its numbers in a table, gives an engine something to lift and attribute. One long unbroken claim gives it nothing to hold.

Structure is now legibility

Structured data used to be a garnish for rich results. For a machine reader it is closer to a contract:

• Organization says who is speaking.
• Product, with real offers, says what is sold and at what price.
• FAQPage says which questions the page answers.
• BreadcrumbList says where the page sits.

Add the boring hygiene: one canonical URL per page, an honest sitemap and hreflang that tells the truth about which languages exist.

The polite conventions are young

llms.txt is newer and less proven: a proposed map of a site for AI systems at a well known URL, with an optional full text companion. There is no public evidence yet that any major engine reads it.

We publish both anyway. The cost is an afternoon, the file is honest whether or not anyone reads it, and our robots file names and welcomes the crawlers that do. This is a bet on a convention, not a measured channel.

The same discipline as quoting

Industrial suppliers are well placed here, because it is the discipline their trade already runs on. An answer engine choosing what to cite does what a buyer's engineer does with a datasheet: it looks for concrete figures, stated sources and claims that stay inside their evidence.

The load table with its reference, the price with its currency, the roadmap item marked as a roadmap item: those are pages a machine can safely repeat. Puffery is not just unpersuasive now. It is unquotable.

What we did to this site

• Structured data on every page, resolved to one Organization node.
• Product nodes on product pages, and the real plan prices as offers on the pricing page.
• FAQ markup that mirrors the visible questions.
• Canonicals, hreflang and the sitemap in agreement.
• A robots.txt that names and welcomes the AI crawlers.
• llms.txt as the map, and llms-full.txt carrying the blog and the docs in full, this post included.