AI search services
Technical AEO
Make sure AI crawlers and search engines can reach, render and read the pages that hold your best answers.
Ideal client
Sites built on JavaScript frameworks, large catalogues, or sites that have blocked crawlers by accident.
Request an AEO AuditIf a page cannot be read, it cannot be quoted
Every AI answer starts with retrieval. Before an engine writes a word, it fetches pages, either from its own index, from a search partner’s index, or live at the moment someone asks. Technical AEO makes sure that step works for you. It is the least glamorous part of the work and often the quickest win, because a single misconfigured rule can hide a whole site from an engine.
What we check
Crawler access
Different AI companies use different crawlers for different jobs, and each can be allowed or blocked separately in robots.txt. OpenAI, for example, documents separate agents for model training (GPTBot), for surfacing sites in ChatGPT search (OAI-SearchBot) and for pages fetched when a user asks (ChatGPT-User), and says each setting is independent. Perplexity, Anthropic, Microsoft and Google have their own agents and controls. Google’s AI Overviews and AI Mode rely on normal Googlebot crawling, while the separate Google-Extended token controls use of content for Gemini training and grounding. We review your robots.txt, CDN and firewall rules so that what is blocked is blocked on purpose, and nothing you want cited is shut out by accident.
Rendering
Many AI crawlers do not run JavaScript the way a browser does. If your key content, prices or navigation only appear after scripts run, some engines may see an empty page. We test what each type of crawler receives and recommend server-side rendering, pre-rendering or simpler templates where needed.
Indexability and snippet eligibility
Google states that to appear in its AI features a page must be indexed and eligible to show a snippet, and that there are no additional technical requirements. That makes classic indexing problems, such as stray noindex tags, canonical errors, blocked resources and thin duplicate pages, AEO problems too. We find and fix them.
Structured data
Structured data does not guarantee citations or rich results, and Google says no special schema is needed for its AI features. It does help machines understand facts about your organisation, services, products, articles and people when it matches what is visible on the page. We implement and test Organization, Service, Product, LocalBusiness, Article, BreadcrumbList and FAQ markup where it is accurate. Our schema guide lists the types that matter most.
Page structure and internal links
Engines extract passages, not whole pages. Clear headings, direct answers near the top of each section, real HTML tables for comparisons and descriptive internal links make the right passage easier to find. We also make sure your most important answers are linked from the pages engines already crawl most.
Optional files such as llms.txt
llms.txt is a proposed convention for giving language models a curated map of a site. Some sites publish one, and it is cheap to add. However, Google says it does not need special AI text files, and adoption by other engines is not confirmed. We treat it as an optional extra, never as a substitute for crawlable pages.
Typical deliverables
- AI crawler access review covering robots.txt, CDN and firewall rules
- Rendering tests for priority templates
- Indexability and canonical audit
- Structured data implementation and validation
- Heading, page structure and internal linking recommendations
- A short technical brief your developers can act on
Technical AEO overlaps with our wider technical SEO service. If your site also has speed, architecture or migration issues, we scope both together.
Questions about technical AEO
Should I block AI crawlers?
That is a business decision. Blocking training crawlers and allowing search crawlers is a common middle path, because it keeps you visible in AI search while limiting use of your content for model training. We explain the trade-offs for each agent and implement whatever you decide.
Does page speed matter for AI search?
Fast, reliable pages are easier to crawl and fetch live, and page experience is part of Google’s general guidance. Speed alone will not earn citations, but slow or failing pages can lose them.
Do you change our code directly?
On WordPress and similar platforms we can implement many fixes ourselves with your permission. On custom builds we write specific tickets for your developers and check the result.
Next step
Discuss your visibility strategy.
Tell us about your website and what you want more of. We will reply with what we would look at first and how an audit would run for your business.