AEO and GEO: How to Get Your Website Cited by AI Search

Answer Engine Optimization (AEO) is the work of making your content the answer a search feature shows directly: a featured snippet, a "People also ask" box, or a spoken answer from a voice assistant. Generative Engine Optimization (GEO) is the newer cousin: making your content easy for AI assistants such as ChatGPT, Claude, Perplexity and Google’s AI Overviews to read, trust and cite when they write an answer.

Both start from the same idea as classic SEO: be the clearest, most reliable page on a topic. The difference is that the reader is now often a machine summarising your page for a human, so structure and plain factual language matter more than ever.

Many modern websites are single-page JavaScript apps. The HTML the server sends is an empty container, and the content only appears after the browser runs the JavaScript. Google can render that, but most AI crawlers (GPTBot, ClaudeBot, PerplexityBot) do not run JavaScript at all. To them, an unrendered React site is a blank page.

The fix is to serve real HTML: server-side rendering, static generation, or pre-rendering each page at build time. Then check your robots.txt. If it blocks AI crawlers, they cannot cite you, so decide deliberately which ones you allow.

Answer engines lift short, self-contained passages. Put the direct answer to a question in the first one or two sentences under a heading that asks that question, then expand. A paragraph that starts with "It depends" and gets to the point three sentences later is much harder to quote.

Use question-shaped headings for the things people actually ask, keep one idea per paragraph, and state facts with numbers, names and dates where you have them. Vague, padded copy reads as low confidence to a model choosing what to quote.

Schema.org markup turns your content into explicit facts: this page is an Article written by this Person, this is an FAQPage with these questions and answers, this business is at this address and opens at these hours. Search engines use it for rich results, and it removes guesswork for any system summarising your page.

Mark up what is genuinely on the page and nothing more. FAQ markup must match a visible FAQ; review markup must reflect real reviews. Misleading schema can cost you rich results rather than earn them.

AI systems build an understanding of entities: people, businesses, products. Help them by describing yourself the same way everywhere: the same name, role and location on your website, LinkedIn, GitHub and business listings. Link them together with sameAs in your Person or Organization schema.

Inconsistent facts work against you. If your site says one phone number and your directory listing says another, a model has to guess, and it may decide neither is reliable.

An emerging convention is an llms.txt file at the root of your site: a short Markdown summary of who you are and links to your most important pages, written for language models. It costs little to add and gives assistants a clean entry point.

A practical starting checklist: serve pre-rendered HTML, allow the AI crawlers you want in robots.txt, add accurate schema, answer real questions in the first sentence of each section, keep your facts consistent across the web, and publish an llms.txt. Then watch which pages assistants cite and improve the ones they skip.