Technical SEO for AI search
AI answers are assembled from pages a machine can reach, parse and trust. This is the technical work that makes your site readable to those systems, and the measurement that shows whether it changed anything.
The six things that decide machine readability
Crawlability
Robots rules, sitemaps, render paths and status codes, so the crawlers that feed AI answers can actually reach the pages.
Structured data
Organization, Service, FAQPage and Article markup matching what is on the page, and kept in one place so it cannot drift.
Entity clarity
One consistent name, description and set of profiles, so the assistants resolve your business to the same entity every time.
Architecture and internal linking
Page structure and link paths that make the important pages obvious to a machine, not just a reader.
llms.txt
A machine-readable summary of what the business does and where the authoritative pages are.
Measurement
Prompt testing and citation spot checks, so the technical work is judged by whether visibility changed.
When this is the right starting point
It suits sites with a visibility problem that new content does not fix: pages that render only in the browser, thin or conflicting structured data, or an entity the assistants describe differently on every platform.
It is also the first stage of a wider programme for most clients, because the content and citation work depend on the crawlers being able to reach the pages in the first place.
What is AI search saying about you?
We run a free audit on the buyer questions in your market across the major AI assistants, and report where you appear, who appears instead, and what to fix first.
- 01
You send your site
Plus the questions you want to be found for, if you already know them.
- 02
We run the prompts
The buyer questions are tested across the AI assistants people actually use.
- 03
You get the findings
Where you appear, who appears instead, and what to fix first.
Frequently asked questions
Does technical SEO still matter now that AI answers the questions?
It matters more, because generative engines retrieve from the same crawled web. A page that cannot be crawled, rendered or parsed is not available to be quoted, however good the writing is.
What is llms.txt?
llms.txt is a plain-text file at the root of a site that summarises what the business does and points to its most authoritative pages. It is a convention rather than a ratified standard, and we treat it as one useful signal among several.
How do you know the technical work helped?
We re-run the same buyer prompts before and after, and count whether the site is mentioned, quoted or cited. Technical fixes are judged on that change, with crawl and index coverage as the supporting evidence.