AI visibility: GEO and AEO
AI visibility is the work of getting your brand cited and recommended in the answers of ChatGPT, Gemini, Perplexity and Google AI Overviews. It combines GEO, for generative engines, and AEO, for direct answers, on top of a solid technical SEO foundation, with prompt testing and monthly monitoring.
In short
- AI visibility means being cited as a source, or recommended as a brand, in the answers of ChatGPT, Gemini, Perplexity and AI Overviews.
- GEO optimises for generative engines; AEO optimises for direct answers such as featured snippets and “People also ask”. Both rest on SEO.
- An assistant only cites what it can crawl, understand and trust: crawler access, citable content, entity schema and external mentions.
- The work starts with a prompt audit and continues with monthly monitoring, alongside SEO consulting.
- The person who audits is the person who implements: Carlos Gonçalves, certified SEO consultant, working in SEO since 2015.
Your customers no longer search only on Google. They ask ChatGPT which supplier to choose, ask Perplexity for a comparison and read Google’s summary before clicking any result. This service covers that new search layer on four fronts: GEO, AEO, AI Overviews and AI Mode, and an LLM visibility audit, which is the starting point.
What is AI visibility?
It is your brand’s presence in the answers generated by language models. When someone asks an assistant for the best shop, clinic or consultant for a problem, the answer mentions a few brands and cites a few sources. Being on that list is AI visibility. Being left out means the decision is made before the customer ever learns that you exist.
The difference from classic SEO is the format. A ranking has ten results and the user chooses. A synthesised answer has one text and a handful of citations. Ranking still matters, because many of these tools search the web before answering, but the goal becomes being chosen as a source.
What is the difference between SEO, GEO and AEO?
| SEO | GEO | AEO | |
|---|---|---|---|
| Where it shows | Organic results on Google and Bing | ChatGPT, Perplexity, Gemini, AI Overviews and AI Mode | Featured snippets, “People also ask”, voice assistants |
| Goal | Rankings and clicks | Being cited and recommended | Being the direct answer |
| Unit of work | Page and keyword | Entity (brand) and prompt | Question and answer block |
| How it is measured | Search Console, GA4, rankings | Prompt tests, citations, mentions | Snippets won, presence in PAA |
In practice these are not three projects. A fast, crawlable site with clear answers and a well defined brand works for all three at once. What changes is what you measure and where you look for gaps.
How do ChatGPT, Gemini and Perplexity choose their sources?
None of these systems publishes the formula, but the observable behaviour is consistent. When a question needs current information, the assistant searches the web, reads a set of pages and summarises them. To get into that set, a page has to meet four conditions:
- Be accessible. AI search crawlers such as OAI-SearchBot, PerplexityBot or Googlebot itself cannot be blocked in robots.txt or at the firewall.
- Be easy to extract. Questions as headings, answers in the first lines, lists and tables. A vague marketing paragraph is rarely cited.
- Be consistent with the rest of the web. The same name, address, services and facts on the site, on the Google Business Profile, in directories and on social profiles.
- Show trust signals. Identified authorship, real experience, mentions on industry sites and reviews.
Beyond real time search, models also carry training knowledge. That is where reputation built over years counts: brands mentioned often in credible sources tend to be remembered even without a search.
What does the AI visibility service include?
- LLM visibility audit: prompt tests in ChatGPT, Perplexity, Gemini and AI Overviews, crawler access, schema and entity consistency. It ends with a prioritised plan.
- GEO: citable content, Organization, Person and Service schema, external presence and a stronger brand entity.
- AEO: direct answer blocks, featured snippets, “People also ask” boxes and structured FAQs.
- AI Overviews and AI Mode: eligibility, snippet controls and reading the data in Search Console.
- Monthly prompt monitoring: the same set of questions, tested every month, with a record of who gets cited.
How does the work run?
It follows the same method as the SEO consulting, with AI presence included from day one:
- Diagnosis (weeks 1 and 2): technical audit, analysis of the SERP and of AI presence, and a 30 day plan.
- Strategy (week 3): a roadmap ordered by impact and effort.
- Implementation (15 day sprints): done by me or with your team, in WordPress, with Rank Math, Cloudflare and, where it makes sense, AI agents through Novamira with human supervision.
- Optimisation (monthly): reporting in Looker Studio with GA4 and Search Console, prompt test results and a monthly meeting.
The GEO and AEO extension runs quarterly or monthly and can be added to ongoing SEO consulting. Technical fixes usually take effect in 4 to 6 weeks; traffic consolidates in 3 to 6 months. In AI answers progress is less predictable, so it is measured with repeated tests rather than promises.
Illustrative example, not a measured result. For Alicouro, an online shop for leather footwear and accessories, a test prompt would be “Where can I buy men’s leather boots made in Portugal?”. The goal is for the answer to cite alicouro.pt.
The groundwork is already there: between October 2022 and October 2023 the SEO strategy took organic traffic to 4,382 visits (+116%) and organic keywords to 463 (+128%), according to SEMrush, with schema, Google Business Profile and industry links. Current work includes connecting WooCommerce to AI through MCP.
Which businesses is it for?
It makes most sense for businesses where customers research before buying and compare options: online shops with their own products, professional services, tourism, private healthcare and B2B. If your site still has indexing or speed problems, the first step is a technical SEO audit. Without that foundation, no GEO work holds up.
How much does AI visibility cost?
The LLM visibility audit has a fixed fee, set out in a written proposal after the free review. The fee for the monthly or quarterly extension depends on the number of prompts monitored and on whether SEO consulting is already in place.
Frequently asked questions
Can I pay to appear in ChatGPT?
There is no way to buy citations in ChatGPT’s organic answers. Appearing depends on the content being accessible, clear and trustworthy, and on the brand being mentioned in other sources.
Does GEO replace SEO?
No. Most assistants search the web before answering, so a page that does not rank or is not crawled is unlikely to be cited. GEO is added on top of SEO.
How long does it take to appear in AI answers?
Access and content fixes can show up within weeks in engines that search in real time. Changes in reputation and training knowledge take months and are not guaranteed.
How do I know whether my brand already appears?
By testing a fixed set of real customer questions across several assistants and recording who gets cited. It is the first step of the LLM visibility audit.
Do you only work with WordPress?
WordPress and WooCommerce are the speciality, where I implement directly. The AI visibility audit can be done on any site.
