AI visibility glossary

AEO is the practice of making a website's content and technical structure readable, trustworthy, and citable by AI answer engines (ChatGPT, Perplexity, Gemini, and Google AI Overviews) rather than only by traditional search crawlers.

Where SEO optimises for ranking in a list of blue links, AEO optimises for being the source an AI system actually pulls facts from and cites when it answers a question directly. That means AI crawlers need to be able to reach your pages, your business facts need to be exposed in structured data an AI can parse, and your content needs to directly answer the questions people are likely to ask. See can ChatGPT find my business? for a free check against all of this.

The three terms overlap heavily and are often used interchangeably across the industry, but they emphasise different things:

• SEO (Search Engine Optimization): getting pages to rank in traditional search results (Google, Bing) for keywords people type.
• GEO (Generative Engine Optimization): a broader term for visibility across generative AI systems in general.
• AEO (Answer Engine Optimization): specifically about being cited as the answer itself, not just appearing in a results list. It's the term Geogize uses because it's precise about what we check: can an AI system read, understand, and cite you.

In practice, good technical SEO (fast, crawlable, well-structured pages) is also the foundation of good AEO and GEO. They're not competing disciplines.

llms.txt is a plain-text file at the root of a website (like robots.txt) that gives AI systems a curated, machine-readable summary of what's on the site, written in Markdown, with links to the most important pages. The idea is to give an AI a lightweight, structured shortcut instead of making it crawl and parse full HTML.

It's an emerging, informal standard: not every AI crawler reads it yet, and no major AI provider has confirmed it's used for ranking or citation decisions. Treat it as a minor, low-cost signal worth having, not a guarantee of anything. Geogize's own llms.txt is a working example, or use the free llms.txt generator to build one for your own site.

Geogize's score runs 0 to 100 across four weighted categories:

• Crawl access (25 pts): can AI crawlers like GPTBot, ClaudeBot, PerplexityBot, Google-Extended, and OAI-SearchBot actually reach your pages, or does robots.txt block them?
• Structured data (30 pts): does the site expose machine-readable facts via JSON-LD (Organization, LocalBusiness, FAQ, Product), or is an AI left guessing from prose? Try the free schema generator.
• Answer readiness (25 pts): does the content directly answer the questions people are likely to ask, in a form an AI can lift and cite?
• NAP consistency (20 pts): does the business name, address, and phone number match exactly everywhere AI systems might cross-check facts?

Bands: 80+ is "ai-ready", 60 to 79 is "partially-visible", and under 60 is "likely-invisible". The score measures technical readiness to be understood: it isn't a guarantee of ranking or citation by any specific AI system. Run a free check to see your own score.