Why Your Website Isn't Showing Up in ChatGPT (And How to Fix It)

AI crawlers are blocked in your robots.txt file

Your website has a file at yoursite.com/robots.txt that instructs automated visitors - including search bots - what they're allowed to access. Many websites were configured years ago with broad restrictions that block all automated visitors, or were set up by developers following old security guidelines.

The result: GPTBot (ChatGPT's crawler), PerplexityBot, and Google-Extended (for AI Overviews) simply cannot read your website. Everything else is irrelevant if this is the case - you cannot appear in AI search results from a site the crawler cannot enter.

In our audit data, a significant proportion of business websites have at least one major AI crawler blocked. Most have no idea this setting exists.

You have no structured data (schema markup)

Humans read your homepage and understand what you do within seconds. AI systems need machine-readable labels to reach the same understanding with confidence. These labels are called structured data or schema markup - they're code snippets embedded in your website that explicitly declare your business name, type, address, phone number, opening hours, services, and more.

Without schema markup, an AI tool reading your website must infer all these facts from your text alone - a process that's unreliable and frequently results in incomplete or incorrect representation. With schema markup, there's no ambiguity. The data is machine-readable and directly consumable.

For most businesses, LocalBusiness schema is the priority. For e-commerce, Product schema. For content sites, Article schema. The right type depends on your business, but the absence of any schema at all is one of the most consistent findings in low-scoring SearchScore audits.

Your business has weak entity signals

AI search tools don't evaluate your website in isolation. They cross-reference everything they know about your business across the entire web to assess credibility and legitimacy. A business with a well-designed website but almost no external presence - no Google Business Profile, few reviews, minimal directory listings, no press mentions - looks like a thin, unverified entity.

Entity signals are the external footprint that makes your business legible to AI systems. They include: reviews on Google, Trustpilot, and industry platforms; listings in relevant business directories; mentions and citations in third-party content; a complete and active Google Business Profile; and consistent brand information across all platforms.

Weak entity signals don't just reduce your AI search visibility - they can cause AI tools to generate inaccurate information about your business, because they don't have enough verified data to draw from.

Your content is too thin or vague for AI to extract answers

AI search tools are designed to find pages that directly answer questions. When someone asks ChatGPT "who provides accountancy services for small businesses in Manchester?", the tool is looking for a page that states clearly and specifically what the business does, where it operates, and who it serves - not a page full of marketing language about being "passionate about numbers" and "delivering exceptional client outcomes."

Thin content means content that circles around what you do rather than stating it plainly. AI tools extract direct, factual statements. If your homepage doesn't contain a sentence along the lines of "[Business name] provides [specific service] in [specific location], serving [specific customer type]", you're not giving AI anything concrete to cite.

This is one of the most common and easily fixable issues in our audit data. The fix doesn't require a full website rewrite - it requires ensuring every key page opens with a direct, specific statement of what the business does.

Your NAP data is inconsistent across the web

NAP stands for Name, Address, Phone number. AI systems cross-reference your business information across multiple sources to verify it's accurate and trustworthy. If your business is listed as "Harrison Consulting Ltd" on your website, "Harrison Consulting" on Google, and "Harrison Consulting Services" on Yell.com, those discrepancies create doubt - and AI tools default to citing sources they can verify confidently.

Inconsistent NAP data is particularly damaging for local service businesses, where location and contact details are among the primary signals AI tools use to identify and recommend you. Even small variations - different phone number formats, abbreviated versus full address, trading name versus registered name - can fragment your entity signals and reduce citation frequency.

You have no llms.txt file

An llms.txt file is a plain text document you publish at yoursite.com/llms.txt that gives AI systems a structured summary of your business - what you do, your most important pages, and how to understand your content. Think of it as a welcome note for AI visitors: instead of making the AI piece together your business context from scattered web pages, you give it a clear, authoritative starting point.

The llms.txt standard is relatively new, which means the vast majority of websites don't have one. That creates a genuine competitive advantage for businesses that add it now - AI tools reading your site will have immediate, unambiguous context that most competitors aren't providing.

An llms.txt file typically includes your business name and description, your key service pages with brief summaries, your location and contact details, and any specific context that helps AI understand your positioning and audience.