AEO (answer engine optimisation)

Answer engine optimisation is what search engine optimisation turns into once the reader is as likely to be a chatbot as a browser tab. When someone asks ChatGPT what a service costs, or asks Gemini who does resilience keynotes in Canada, those tools go looking for a page that states the answer plainly. AEO is the practice of making sure that page is yours, and that it says the true thing.

The mechanics are less mysterious than the name suggests. An answer engine wants a clear claim, near the top of the page, backed by structured data that confirms what the page is about. It wants a real name attached to real expertise, not a wall of marketing copy. It wants the same fact stated consistently everywhere it looks, your site, your listings, your social profiles, so it can trust the fact instead of guessing.

The alternative to doing this work is not neutral. It is not that AI tools simply say nothing about you. They answer the question anyway, using whatever scraps they can find, and when the scraps are thin they fill the gap with something that sounds plausible and is wrong.

Why it matters to you

I have watched Gemini invent a speaking fee for a client because the real number was never published anywhere. I have watched it invent client testimonials and attribute them to the wrong page. Nobody typed those lies in. The AI had a question to answer and not enough real material, so it built an answer out of pattern-matching. That answer then sits in front of a buyer who has no reason to doubt it.

AEO is how you take that decision back. If your site clearly states your fee structure, your service area, and who you actually are, the AI has less reason to guess. Get this right and the organic result work you have already done starts paying twice: once for the person who clicks a link, and once for the person who never sees a link at all because the answer arrived fully formed.

How I set it up

I start by running the buyer’s actual questions through the major AI tools and reading exactly what comes back, right down to which page it cited. That tells me where the gaps are. Then I build or rewrite the pages that should be answering those questions, with the real fact stated once, near the top, in plain words, and backed by structured data so a machine can parse it as confidently as a person reads it.

I do not chase every keyword. I chase the questions that actually get asked, including the unbranded search questions where nobody types your name because they do not know it yet.

What it looks like in practice

Six weeks after a fix, the same question that used to return an invented number now returns your real fee range, sourced correctly, with a link back to your own booking page. Nobody notices the change happening. They just notice that the AI, this time, told the truth.