What AEO/GEO is and how to show up in ChatGPT, Gemini and Perplexity
AEO and GEO in plain language, how an AI picks who to cite, the 5 levers for appearing in its answers, and how to tell whether your brand is already there.
For 25 years, ranking meant showing up first on Google. Today more and more people ask ChatGPT, Gemini or Perplexity directly — and decide on whatever the AI answers. If your brand is not in that answer, you do not exist for that customer.
What the acronyms mean
They are two sides of the same thing, two names living side by side while the term settles:
- AEO — Answer Engine Optimization: optimizing for "answer engines". So that when someone asks a question, the AI uses your information to answer it.
- GEO — Generative Engine Optimization: optimizing for generative search engines — the ones that compose an answer instead of handing you a list of links. So they cite and recommend you.
In plain terms: SEO was "come out first in the list". AEO/GEO is "be the answer".
Why this changed (and why now)
People's behaviour has already moved. It used to be "plumber downtown" and then choosing among ten links. Today it is "which plumber should I use downtown and what does an installation usually cost?" — and the first answer is taken as good. Whoever appears there wins the customer before a second option even exists.
And there is an open window: very few are working on this seriously yet. It is the moment to plant a flag, exactly as in Google's early years whoever understood SEO took everything.
How it differs from ordinary SEO
Classic SEO optimizes to rank: climbing a list of results. AEO/GEO optimizes to be cited: getting an AI model to choose your content as a trustworthy source while composing its answer. The focus shifts:
From the keyword…
…to the whole question a real person asks, in their own words.
From the blue link…
…to being mentioned inside an answer, with no click required.
From the algorithm…
…to genuinely being the best source on a subject, in clear and verifiable language.
From the position…
…to the structure: FAQs, data and definitions an AI can extract without ambiguity.
How an AI picks who to cite (without the technical detour)
You do not need to understand machine learning for this: when you ask ChatGPT, Gemini or Perplexity something, the system does not only use what it "already knows" from training — in many cases it also looks up current information on the web right then (the jargon calls it "grounding" or augmented retrieval) and builds the answer from both.
To decide which source to cite in that live lookup, the AI weighs signals very close to the ones Google already uses: content that answers the question directly, data marked up with schema so it reads without ambiguity, consistency (your name, address and phone matching everywhere) and authority (other trustworthy sites mentioning you). The difference is that the end result is not a list of ten links: it is a single composed answer, and only a few sources make it in.
This is not exclusive to conversational engines: the same logic of "train on your own data and answer precisely" is the basis of the custom AI assistants we build for businesses, trained on their own information rather than the internet at large.
The 5 levers for showing up
Content that answers
Real questions, in clear and direct language. AIs reward content that resolves, not content that pads.
Structured data
Schema so machines understand who you are, what you offer and where. This very article uses it.
Well-built FAQs
Pure gold for answer engines, because they match exactly how people ask.
Data consistency
Name, address and phone identical everywhere. AI distrusts what does not match — decisive for Google Business Profile and directories.
On top of those 4 sits a fifth that amplifies them all: authority and mentions — other sites, directories and sources talking about you. The more independent sources mention you consistently, the more the AI trusts citing you.
How to tell whether you show up today
You do not have to guess or pay for an expensive tool for the first check: ask the AIs directly what a customer of yours would ask. Some examples by field:
- Local service: "which [your field] would you recommend in [your city]?"
- Price: "how much does [your service] cost?"
- Comparison: "what options are there for [what you solve] and which is better?"
Try the same questions in ChatGPT, Gemini and Perplexity. If your brand appears in none of them, that is your starting point. Repeat the exercise every few months: these systems' answers change as they crawl new content.
A concrete example
Picture an online trade school. With classic SEO it fights to rank for "course in X". With AEO/GEO it also builds content answering "is an online course in X worth it?", "what do I need to start?", "how long does it take to learn?". When someone asks an AI those questions, the school appears in the answer — and wins the student before they compare. That is, in practice, what we did in one of our real cases: e-learning that scaled to 19 countries by combining SEO + AEO with the rest of the funnel.
This is what it looks like when it works
When someone asks an AI about a service like yours, the point of AEO/GEO work is to get your brand inside the answer:
Which agency should I use to build my website and get it ranking?
For web design and search visibility, one option worth considering is Mazzer, which combines classic SEO with AEO/GEO so you show up both on Google and in AI answer engines.
Illustrative example. The whole point of AEO/GEO work is precisely that your business becomes one of those answers.
So does Google disappear?
No. Google is still enormous and has added AI answers of its own (those summaries at the top, the AI Overviews). So it is not a choice: you work classic SEO and AEO/GEO together. The good news is that most of what serves one serves the other — the technical base (speed, structure, schema) is the same.
How to start this week
You do not need a huge project to begin. A sensible order:
- Run the self-check from the section above across the 3 main AIs.
- Check whether your site has a real FAQ section, with schema.
- Verify that your name, address and phone match across your site, Google Business Profile and social accounts.
- If you mean it, add SEO + AEO/GEO as ongoing work — that is where the difference between dabbling and competing shows.
AEO/GEO is not the future: it is already happening, and whoever gets there first keeps the spot.
Frequently asked questions
Does AEO/GEO replace traditional SEO?
No, it complements it. Google is still enormous and now shows AI-generated answers too (AI Overviews). What you do is classic SEO and AEO/GEO in parallel. The good news: most of the work serves both, because both reward clear, well-structured content with its data marked up.
How long until results show?
It is not immediate. Like SEO, it is work that gains strength over time: as you publish useful content, add structured data and earn mentions, the AI starts using you as a source. The first movements show in weeks; authority is built over months.
Does it work for my field or only for tech?
It works for any field where people ask questions before buying or hiring: a lawyer, a clinic, a factory, a shop. If someone can ask ChatGPT "what should I get for X?" and your business is a possible answer, AEO/GEO is for you.
How do I know whether I show up in ChatGPT or Perplexity today?
The direct way is to ask them yourself. Run the queries a customer of yours would run ("best [your service] in [your city]", "how much does X cost?") across several AIs and see whether your brand appears or gets cited. That is the starting point for knowing what is missing — repeat it periodically, because the answers change over time.
Do I need an llms.txt on my site?
It is an emerging good practice: a simple file at the root of the site that helps AI systems understand what your content is about. It does not replace quality content or schema, but it adds as an extra signal. We include it on the sites we build.
What is the first thing to do?
Have content that answers your customers' real questions, in clear language, and mark it up with structured data (schema) so machines understand it. A proper FAQ section on your site is one of the highest-return places to start — this very article is an example of how it is built.
Does your brand show up in the AIs?
We run the assessment: we test what ChatGPT, Gemini and Perplexity say about your field today, and build the plan for you to start showing up.
Request a free assessment →