What is AEO, and how do you get your business cited by AI search?

A growing share of searches no longer end on a list of blue links. Someone asks Google a question and gets an AI Overview at the top. Someone asks ChatGPT or Perplexity and gets a written answer with a handful of cited sources underneath. If your business is not among those sources, you are invisible in that moment, even if you rank well in the traditional results below.

Answer Engine Optimisation, or AEO, is the practice of making your content the thing those answer engines quote. It overlaps heavily with good SEO, but it is not identical. Here is how the engines actually choose, and what you can do about it.

How AI search picks its sources

AI answer engines work in two broad steps. First they retrieve a set of candidate pages, usually using a search index. Then a language model reads those pages and writes an answer, citing the ones it leaned on.

That second step is the new part. The model is not ranking ten links. It is reading passages and deciding which ones clearly and credibly answer the question. So the winners tend to be pages that:

• State the answer directly, near the top, in plain language.
• Are structured so a specific passage can be lifted out and stand on its own.
• Come from a source the model has reason to trust.
• Are machine-readable, with clean HTML and structured data.

If your page buries the answer three paragraphs into a wall of marketing copy, the model has to work to find it, and it often picks a competitor who made it easy.

Write so a single passage can be quoted

The most practical AEO habit is to answer the question, clearly, in the first sentence or two of a section, then expand.

Think about how you would want to be quoted. If someone asks “how long does a bespoke website take to build”, the page that wins is the one that says “A bespoke small-business site usually takes four to eight weeks” and then explains the variables, not the one that opens with a paragraph about your studio’s philosophy.

Use clear headings phrased the way people ask questions. Use short paragraphs and lists where they fit. Each section should make sense even if it is the only part someone reads, because in an AI answer, it might be.

Structured content and schema

Schema markup is structured data added to your page that tells search and AI tools exactly what they are looking at: an article, an organisation, a product, an FAQ, a how-to. It does not change what visitors see, but it removes ambiguity for machines.

For AEO, the most useful types tend to be Organization (so tools know who you are), Article or BlogPosting, FAQPage for question-and-answer content, and Breadcrumb for structure. Clean, valid markup makes your content easier to retrieve and to trust.

This is also where a bespoke site has an advantage. On a template you get whatever schema the theme decides to output. When the HTML is hand-coded, you control it precisely.

E-E-A-T: why trust signals matter more now

Google describes the qualities it looks for as Experience, Expertise, Authoritativeness and Trustworthiness, shortened to E-E-A-T. AI engines lean on similar signals, because a model citing a source is, in effect, vouching for it.

In practice that means:

• Make it obvious who wrote the content and why they are qualified.
• Show real experience: specifics, examples and numbers beat vague claims.
• Be consistent across the web, so your name, details and positioning line up.
• Earn mentions and links from other credible sites.

You cannot fake your way to being trusted. But you can make your genuine expertise legible, and a lot of businesses simply do not.

Let the AI crawlers in

AI engines use their own crawlers to gather content. The main ones are GPTBot from OpenAI, ClaudeBot from Anthropic, and PerplexityBot from Perplexity, alongside Google’s existing crawlers that feed AI Overviews.

If your robots.txt blocks these crawlers, you opt out of being cited by those tools. Some businesses choose to block them on purpose, and that is a valid decision. But many block them by accident, copy a restrictive robots.txt from somewhere, and then wonder why they never appear in AI answers. Check what yours allows.

The llms.txt convention

A newer idea is llms.txt, a plain text file at the root of your site that points AI tools to your most important, cleanest content. Think of it as a curated map for language models, similar in spirit to a sitemap.

Adoption is still early and no engine treats it as a guarantee of anything. It is low effort to add, so it is worth doing as part of a wider AEO setup, but do not expect it to do the heavy lifting on its own. The content and structure matter far more.

A practical checklist

If you want to be cited by AI search, start here:

1. Answer real questions directly, near the top of each section.
2. Structure content so individual passages stand alone.
3. Add valid schema markup for your organisation and articles.
4. Make authorship and expertise clear on the page.
5. Confirm your robots.txt allows GPTBot, ClaudeBot and PerplexityBot.
6. Keep the site fast and clean, because retrieval still favours pages that load and parse easily. (This is where a bespoke build quietly earns its keep.)
7. Add an llms.txt once the above is in place.

None of this is a trick. It is the same goal as good SEO, pursued one layer deeper: be the clearest, most credible answer to the question your customer is asking. Do that, and both Google and the AI engines have every reason to point at you.

Want to know where you stand today? Our free AI-Ready check measures your load speed and asks a live AI search whether it can find your business at all. It takes about a minute. And if the answer needs fixing, that is what we do.