Type a question into Google this year and there is a fair chance you read the paragraph at the top and closed the tab. Ask ChatGPT who to hire and it gives you names, not ten sites to compare. A growing share of searches now resolve inside a generated answer, in AI Overviews, in Perplexity, in whatever assistant ships with the phone. The page that would once have ranked first is now, at best, a citation in a footnote.
The industry took a while to name this, then named it twice. AEO and GEO overlap, get used interchangeably, and are not the same thing. Here is how each arrived, and what they ask of you.
First came the answer engine
Google had been edging towards this for a decade. Featured snippets lifted a paragraph from a page and put it above the results; voice assistants read that paragraph aloud. AI Overviews finished the thought: a written summary, sourced from several pages, sitting where the first link used to be. AEO, answer engine optimisation, is the discipline that grew up around this: structuring content so an engine can lift a clean, accurate answer from it and credit you for it.
In practice it rewards writing the way people ask. A direct question as the heading, a direct answer in the first sentence beneath it, then the nuance. Structured data (FAQ, HowTo, Organisation, Product) so the machine does not have to guess what a block of text is. And specificity: "a brand identity project takes eight to twelve weeks" is an answer an engine can quote. "Timelines vary depending on scope" is not.
Then the engines started writing
ChatGPT changed the habit before it changed the technology. People got comfortable asking a full question and accepting a composed reply. Claude, Gemini and Perplexity's own model followed, and none of them lift a paragraph: they write one, from what they learned in training and what they retrieve the moment you ask. GEO, generative engine optimisation, is the work of making your brand legible to those models, so that when a founder asks "who are the good branding studios for a £10k budget in London", your name is one the model has a reason to produce.
Two engines, two jobs
AEO: the answer engine
- Extracts one answer from a page
- Rewards clear, direct statements
- Featured snippets, voice assistants
- Your page is quoted
GEO: the generative engine
- Writes its own answer from many sources
- Rewards being consistently described
- AI overviews, chat assistants
- Your brand is cited, or not
That is a wider brief than AEO. It covers what other sites say about you, whether your facts agree everywhere they appear, whether your positioning is stated in language a model can compress into one sentence, and whether you show up in the places models weight heavily: reviews, directories, press, well-linked comparison pieces. A model will not recommend a company it has no coherent picture of.
Why they rose together
Think of AEO as the on-page discipline and GEO as the off-page reputation, sharing a foundation of clarity. A page that answers a question directly serves the engine that quotes it and the model that retrieves it. A brand described the same way on its own site, on LinkedIn, in a directory and in a press piece gives a model something stable to learn. The two rose together because the same shift, people asking instead of browsing, feeds both.
What neither is: a trick. The early "GEO hacks" (stuffing pages with statistics, adding fake citations, writing in a chatbot's voice) are already being discounted the way keyword stuffing was fifteen years ago. What lasts is dull and effective: say true, specific things about what you do, in the shape the question is asked, everywhere it matters.
What to do about it
What to do about it
Three moves cover most of the ground for a small or mid-sized business.
Rewrite your key pages as answers. Services, pricing, process, about. Each should open with the question a buyer actually asks and answer it in the first line. Add FAQ sections where the questions are genuinely frequent, and mark them up.
Make your facts agree. Name, location, what you do, who for, what it costs. Check every profile and listing says the same thing. Inconsistency is the most common reason a model hedges or leaves you out.
Get described by others. A guest piece, a client case study on their site, a review on a platform models trust. Third-party descriptions carry more weight with a generative engine than anything on your own domain.
Then ask the engines. Put your category into ChatGPT and Perplexity and see who comes back. That result is your baseline, and the most honest audit of your brand you will get this year.
Frequently asked
What is AEO?
Answer engine optimisation: shaping your content so that answer engines such as Google AI Overviews, Perplexity and voice assistants can pull a direct, citable answer from it.
What is GEO?
Generative engine optimisation: making your brand, facts and language legible to large language models such as ChatGPT, Claude and Gemini so they mention and cite you when they generate a response.
Why did AEO and GEO appear now?
Because the habit changed. Once people started asking assistants full questions and trusting the written reply, a business absent from that reply was absent from the decision. SEO still matters, but it no longer reaches the moment of choice on its own.
