Technical GEO
Why a site that ranks on Google
can be invisible to AI.
You can rank #1 and never get cited in an AI answer. The reasons are usually technical: the page renders with JavaScript the crawler doesn't run, the content isn't extractable, or it's stale. Here's how that happens, and how to fix it.
Technical GEO · Updated
For the technical lead who runs SEO in-house.
A site can rank #1 on Google and still be invisible to AI. The usual reasons are technical: the page renders with JavaScript that AI crawlers don't execute, its content isn't structured into extractable passages, or it's stale. Resonate Labs treats this as the foundation of GEO, making your pages readable, structured, and current so AI engines can actually cite them.
Ranking isn't citation
Google ranking and AI citation are different systems. A page can win one and lose the other, for technical reasons.
It's often rendering
Many AI crawlers don't run JavaScript. If your page needs JavaScript to show its content, the crawler sees an empty page.
Where Resonate fits
Resonate Labs makes the technical foundation right first: readable, extractable, current pages, then the content that earns citations.
Ranking isn't the same as getting cited
It's natural to assume that if you rank well on Google, AI engines will pick you up too. They're built on different systems, and the overlap is smaller than most teams expect. Ahrefs found only about 11% citation overlap between AI assistants and Google's top 10 results, which means roughly 89% of citation opportunities are platform-specific. A strong Google position doesn't carry over on its own.
The reason is that the signals that drive a ranking aren't the ones that drive a citation. When a page wins on Google but loses in AI answers, the cause is usually technical, not a lack of authority. The page may be unreadable to the crawler, unstructured for extraction, or simply out of date. The next section covers what actually moves citation.
What actually makes content citable by AI
The common assumption is that citation comes down to schema markup or domain authority. The data says otherwise. In Profound's analysis of more than 50,000 prompts, organic traffic explained only about 5% of AI citation and backlinks under 4%. The classic authority signals barely move it. What matters is more fundamental, and it stacks in roughly this order.
Readable. First, the crawler has to be able to read the page at all. Many AI crawlers don't execute JavaScript, so a page that assembles its content client-side serves them an empty HTML shell. This is the single most common reason good content goes uncited, and it's covered in the case study below.
Extractable. Once a page is readable, its content has to be structured so a model can lift a clean, self-contained passage out of it. Walls of text and copy that only makes sense in context are hard to cite. Clear headings and passages that stand on their own get pulled into answers, and structuring content into citable units is the craft layer that does this.
Structured. Structured data and schema help an engine understand what a page is and what entities it covers. They're worth doing, but they're necessary rather than sufficient. Schema on an unreadable or unstructured page doesn't rescue it.
Current. AI citation favors content that's kept up to date, and stale pages get displaced over time. Freshness is a maintenance discipline, not a one-time fix.
Specificity ties these together. A precise, well-structured, current answer gets cited over a vague one from a more authoritative domain. That's why the lever here is technical and editorial, not reputation.
What's different if you already run technical SEO
If you run technical SEO in-house, you have a real head start, but GEO isn't the same job. The biggest gap is rendering. Googlebot executes JavaScript, so a client-side-rendered page can index and rank perfectly well on Google. Many AI crawlers don't execute JavaScript at all, so that same page can be completely invisible to GPTBot, ClaudeBot, and PerplexityBot. A site that's technically healthy by SEO standards can still be unreadable to AI.
The optimization target is different too. SEO optimizes for a ranking position; GEO optimizes for inclusion in a synthesized answer. That rewards self-contained, extractable passages over keyword-targeted pages, and it puts more weight on freshness and specificity. Your technical SEO foundation helps, but making a site readable and citable for AI is additional work, not a byproduct of ranking well.
A case study: from invisible to cited
Here's what this looks like in practice. In one documented case, a mid-market B2B HRIS platform was producing good content and ranking on Google, but sat at about 1.3% visibility across ChatGPT, Claude, Perplexity, and Gemini, showing up in just 2 of 150 buyer-research queries.
The diagnosis was rendering. The platform was built on a client-side-rendered framework, so AI crawlers were requesting pages and receiving empty HTML shells. GPTBot, ClaudeBot, and PerplexityBot were all hitting JavaScript none of them execute. The content team was writing well, and the models couldn't read a word of it.
The fix was to serve identical server-rendered HTML to bots and humans, so crawlers received real content instead of an empty shell, with no re-platform required. Citation tends to follow quickly once a page can actually be read. Profound found a median of about 6.81 days from publishing a page to its first citation by ChatGPT or Claude, with 90% of pages cited within roughly 37 days, across about 900 newly cited pages. Here, visibility moved to about 8% by the week-four re-audit and 16% by week six. Nothing about the content strategy changed in that window. The only difference was that the pages became readable, which is the whole point: when AI can't see your site, the best content in your category can't help you.
Getting the technical setup right
Whether you fix this in-house or hand it to a vendor, the technical requirements are the same. At a minimum, a site that AI engines can read and cite needs:
- Crawler access. An explicit allow-list for the AI crawlers that matter, GPTBot, ClaudeBot, PerplexityBot, and the others your buyers' engines use, rather than a blanket block or a managed setting that quietly disallows them.
- Server-side or pre-rendered HTML. Crawlers should receive real content, not a JavaScript shell they can't execute.
- Extractable structure. Content organized into clear headings and self-contained passages a model can lift into an answer.
- Structured data and schema. Appropriate markup so engines understand what each page is and which entities it covers.
- A freshness cadence. A maintenance rhythm that keeps priority pages current.
- A clean, accurate sitemap. So new and updated pages are discoverable.
To audit your own site against these requirements, the technical GEO readiness checklist turns them into a self-audit with a quick way to check each one. If you're putting them on a vendor instead, the GEO vendor RFP and scorecard turns this list into a checklist you can hand them and grade responses against.
Where Resonate Labs fits
Resonate Labs treats the technical foundation as the first step of GEO, not an afterthought. Before content can earn citations, the pages have to be readable, extractable, and current, so the work starts by making sure AI crawlers can actually see your site. From there, the monthly AI visibility audit tracks whether the changes are moving citation across ChatGPT, Claude, Gemini, and Perplexity.
It's a standard we hold ourselves to. This site is served as static, server-rendered HTML with an explicit AI-crawler allow-list, so the engines read the same content you do. If you want to know where your own site stands today, a GEO Snapshot shows what AI engines can and can't see.
Frequently asked questions
Why does a site that ranks on Google not get cited by AI?
Ranking and AI citation are different systems, and the gap is usually technical. The most common cause is rendering: many AI crawlers don't execute JavaScript, so a page that builds its content with client-side JavaScript serves them an empty HTML shell. The crawler reads nothing, no matter how well the page ranks on Google. Other causes are content that isn't structured into self-contained, extractable passages, and pages that are stale. Strong Google rankings don't carry over on their own, because the signals that drive ranking aren't the ones that drive citation.
What makes content get cited by LLMs, schema, page structure, or domain authority?
All of them matter, but not equally, and it isn't mainly domain authority. In Profound's analysis of 50,000+ prompts, organic traffic explained only about 5% of AI citation and backlinks under 4%, so the classic authority signals barely move it. What matters most is being readable (served as rendered HTML a non-JavaScript crawler can read), then extractable (structured into clean, self-contained passages), then supported by structured data and schema, which help but aren't a silver bullet, and kept reasonably current. Specificity beats vague authority.
We run technical SEO in-house, what's different for AI crawlers?
The biggest difference is rendering. Googlebot executes JavaScript; many AI crawlers do not, so client-side-rendered pages that Google indexes fine can be invisible to GPTBot, ClaudeBot, and PerplexityBot. Beyond that, GEO optimizes for inclusion in a synthesized answer rather than a ranking position, which rewards self-contained, extractable passages over keyword-targeted pages, and it values freshness and specificity. Your technical SEO foundation helps, but it doesn't automatically make a site readable and citable for AI.
What technical requirements should we put on a GEO vendor?
Require explicit AI-crawler access (an allow-list for GPTBot, ClaudeBot, PerplexityBot, and the other engines your buyers use), server-side or pre-rendered HTML so crawlers receive real content, content structured into self-contained extractable passages, appropriate structured data and schema, a freshness cadence, and a clean sitemap. Resonate Labs builds these into the technical foundation, and the GEO vendor RFP and scorecard turns them into a checklist you can hand a vendor.
Next step
Find out what AI can actually see.
A free GEO Snapshot maps your category and shows where you stand across the four engines, starting with whether AI crawlers can read your site at all.
- Whether AI crawlers can read your pages today
- Where you're visible, cited, or absent across the four engines
- What the first 30 days would move