ChatGPT, Perplexity, and Google AI Overviews are now answering millions of questions that used to send people to search results. If your website isn't being cited in those answers, you're losing visibility you may never see on a Google Analytics dashboard. Here are the five most common reasons it's happening — and exactly what to do about each one.
Traditional SEO gets your page to rank in a list of blue links. But AI engines don't show a list — they synthesize an answer from sources they trust, then sometimes cite those sources at the bottom. Your goal with Generative Engine Optimization (GEO) is to be one of those trusted sources.
The problem: the rules for getting cited by AI are different from the rules for ranking on Google. Many sites that rank well on Google are still completely invisible to AI search — and the reasons are fixable once you know what they are.
⚠️ AI search citations aren't just a nice-to-have. When someone asks ChatGPT "what's the best [tool/service/product]" and your competitor gets cited and you don't, they get the conversion. This compounds over time as AI usage grows.
Reason 1 of 5
AI engines can only cite content they've read. If your robots.txt file blocks their crawlers, they literally cannot access your site — no matter how good your content is.
The most dangerous pattern is User-agent: * followed by Disallow: /. This blocks every bot in existence, including GPTBot (ChatGPT), ClaudeBot (Anthropic), and PerplexityBot — in a single line. Thousands of sites have this accidentally set by a plugin, a developer, or a misconfigured CMS.
Visit yourdomain.com/robots.txt right now. Check for any rules that block GPTBot, ClaudeBot, PerplexityBot, or Google-Extended. If you're blocking them, open your robots.txt and either remove those rules or replace them with explicit Allow: / directives. See our detailed guide on fixing robots.txt for AI crawlers for exact steps.
Reason 2 of 5
When an AI engine reads your page, it doesn't just see text — it tries to understand what that text means. Who wrote it? What is it about? Is it a product, an article, a FAQ, a business? Without structured data, the AI has to guess.
JSON-LD schema markup is machine-readable context embedded in your page. It tells AI engines exactly what they're looking at. Sites with rich structured data are significantly more likely to get cited in AI answers because the information is unambiguous and easy to extract.
The FixAdd JSON-LD to every page. At minimum, every site should have Organization schema on the homepage. Blog posts should have Article or BlogPosting schema. Pages with Q&A content should have FAQPage schema — this directly maps to the format AI uses to answer questions. Here's a minimal example:
<script type="application/ld+json">
{
"@context": "https://schema.org",
"@type": "FAQPage",
"mainEntity": [
{
"@type": "Question",
"name": "What does your product do?",
"acceptedAnswer": {
"@type": "Answer",
"text": "Your clear, direct answer here."
}
}
]
}
</script>Reason 3 of 5
AI engines quote content verbatim. That means your writing needs to contain clear, standalone statements that make sense when pulled out of context. Marketing copy — "we deliver world-class solutions that transform your business" — is useless to an AI. It can't cite that because it says nothing specific.
Compare these two sentences:
The second sentence is specific, factual, and complete. An AI can quote it directly in response to "what does SiteOracle do?"
The FixGo through your key pages and rewrite vague statements into direct, specific ones. Add a clear definition of what your product or service is in the first paragraph of your homepage. Add a "What is X?" section to your most important pages — this maps directly to the most common type of AI query.
Reason 4 of 5
AI engines don't just evaluate individual pages — they evaluate whether a site is a credible authority on a topic. A site with one blog post about SEO is much less likely to be cited than a site with twenty in-depth articles covering the topic from every angle.
This is called topical authority. It's not about keyword stuffing — it's about demonstrating deep expertise through comprehensive, interconnected coverage of your subject area.
The FixMap out the key questions someone would ask about your topic and create dedicated content answering each one. Interlink those pages. Think less about "targeting keywords" and more about "covering the topic completely." If someone could get every answer they need about your subject from your site alone, you have good topical authority.
Reason 5 of 5
AI models learn from the web — and the web's opinion of your site matters. If reputable external sources mention you, link to you, or cite your data, it signals to AI engines that you're a trustworthy source worth referencing. If your site exists in isolation with no external mentions, AI engines have no reason to prioritise you over a competitor who does have them.
This is similar to traditional SEO's link-building concept, but broader. Mentions in news articles, industry publications, podcasts, and social media all contribute to your authority profile for AI citation purposes.
The FixActively build your external authority: get your business listed in industry directories and databases, pitch original data or research to publications in your space, respond to journalist queries (HARO-style), and encourage satisfied customers to mention you in public-facing reviews, posts, and content. Every external mention is a vote of confidence that AI engines can pick up.
Use this checklist to audit your site against all five issues:
💡 You don't need to fix all five at once. Start with reason 1 (robots.txt) because it's the fastest win — five minutes to check, five minutes to fix. Then tackle structured data. Then work through content quality and authority over the following weeks.
Unlike Google ranking changes that can take months, some GEO improvements can show results faster — because AI models are regularly updated with new web crawl data. Fixing a robots.txt block can mean AI crawlers access your site within days. Structured data improvements may reflect in AI citations within a few weeks as crawlers revisit your pages.
Building topical authority and external signals takes longer — typically months — but the compounding effect is significant. Sites that establish deep authority now are building a moat that will be very hard for competitors to close.
SiteOracle scans your robots.txt, structured data, content quality, and authority signals — then gives you a prioritised list of exactly what to fix, starting with the issues that will have the biggest impact on your AI search visibility.
Run Free AI Visibility Scan →The most common reasons are: you're blocking AI crawlers in robots.txt, your content lacks structured data, your writing is vague rather than direct and quotable, your site lacks topical depth, or you have no external authority signals. SiteOracle can diagnose which of these apply to your specific site.
Allow AI crawlers in robots.txt, add JSON-LD structured data, write clear and direct definition statements, build topical authority by covering your subject comprehensively, and earn external mentions from authoritative sources. These five steps collectively cover the main GEO ranking factors.
Generative Engine Optimization (GEO) is the practice of optimizing your website to be cited and referenced by AI-powered search engines like ChatGPT, Perplexity, and Google AI Overviews. Unlike traditional SEO which focuses on ranking in link lists, GEO focuses on getting your content quoted or summarized in AI-generated answers.
Yes — significantly. FAQPage schema makes Q&A content directly extractable by AI. Organization schema establishes who you are. Article schema signals that content is authoritative and current. Without schema, AI engines have to infer the meaning and structure of your content, making it less reliable as a citation source.