How to Get ChatGPT, Gemini or Perplexity to Cite You Instead of Your Competitors

Imagine this: a potential client types into ChatGPT who is the best web design agency in Valencia. The AI responds. It cites three names. Yours is not among them.

It is not that your website is poorly ranked on Google. You might be in the top 3. The problem is something else: AI engines do not use the same criteria as Google to decide who they cite. And most businesses do not know this yet.

That is exactly what we are going to cover today.

First, the numbers that change the conversation

No need to be alarmist. The data speaks for itself.

Traffic referred by AI systems grew 527% between January and May 2025 compared to the same period the previous year. At the same time, Google AI Overviews now appears in 25% of all searches, according to Conductor's analysis of 21.9 million queries. ChatGPT has 900 million weekly active users, according to OpenAI's official announcement in February 2026.

Put another way: people are still searching. They just get their answer from an AI directly now, without ever needing to visit any website.

That is where the opportunity lies. The window is still open.

SEO and GEO: they are not the same thing

Classic SEO optimises for climbing positions in a list of results. GEO (Generative Engine Optimization) optimises for getting an AI to cite you when it builds its response.

The difference seems subtle. It is not.

The good news: they are not mutually exclusive. GEO is built on top of SEO. It does not replace it.

To make it concrete: if you have an article about the best branding agencies in Spain, SEO tries to get you into the top 3 when someone searches Google. GEO tries to get ChatGPT to mention you by name when someone asks the same question in a chat. Two different games, two different fields, two different sets of rules. You can win both. But you cannot play the second one without knowing it exists.

Why each AI cites differently

ChatGPT, Gemini and Perplexity do not work the same way. It is like three different journalists covering the same story: each one has their preferred sources, their own criteria, and their own way of verifying information.

ChatGPT, Gemini and Perplexity: the same sector, three different criteria for deciding who gets cited.

Perplexity explicitly cites its sources in every response and favours recent, updated content. It is the most transparent of the three. It also has some of the highest conversion rates for SaaS products.

Gemini is deeply integrated with the Google ecosystem. If you already have good SEO on Google, you have a head start when it comes to appearing in Gemini.

ChatGPT generates 87% of all AI-referred traffic. It sends the most visits. But it cites in a less predictable way than Perplexity.

The common denominator across all three: they all learn from the mentions that exist about your brand on the web. Not from your keywords. From what others say about you.

The 4 concrete moves

No theory. This is what the brands already appearing in AI responses are actually doing.

1. Build a citable structure from the first line

AI systems read your page to build their response. If the answer to the most important question in your sector is buried in paragraph twelve, the AI will not find it. Put the direct answer in the first 200 words. Use lists. Use tables. A well-structured piece of content is easier to cite than a brilliant but dense one. Think of it like putting the executive summary at the top of a report: anyone reading quickly gets the essentials.

2. Build presence where AI learns

AI engines do not learn only from your website. They learn from everything that exists about you on the internet. Specialist forums, trade press articles, reviews, communities like Reddit. If you only exist on your own website, you barely exist as far as the AI is concerned. This is not about spamming forums. It is about having a genuine presence where your sector has genuine conversations.

3. Update with purpose, not to tick a box

Systems that use real-time search (Perplexity and Google) prioritise content with a recent update date. But updating does not mean changing four words and touching the timestamp. It means adding new data, revising what has become outdated, incorporating examples from this year. A real update sends a freshness signal that cosmetic updates do not. Between 40 and 60% of sources cited by AI rotate every month. Anyone who publishes once and disappears loses position constantly.

4. Get others talking about you on their own ground

A mention of your brand in a trade publication, in another professional's post, or in a relevant Reddit thread is worth more for GEO than five backlinks from directories. Not because backlinks are dead (they are still useful for classic SEO) but because AI weighs distributed authority: how many independent sites mention you as a reference in your field. It is the digital equivalent of old-fashioned reputation: not what you say about yourself, but what others say about you.

How GEO is measured today

The table in the previous section mentions citation rate and share of voice in AI. These are not abstract concepts. They can be measured right now with tools that already exist.

The most basic and free method: open ChatGPT, Gemini and Perplexity and ask directly about your category. Something like who are the best agencies for X in the UK, or who is considered a reference for Y in this sector. Note whether you appear, in what position, and with what context. Do this once a month. That is already a real tracking system.

What you measure in GEO is not position. It is presence. Whether the AI mentions you, in what context it mentions you, and whether that context is positive or neutral.

The content that AI cites has a specific structure: it answers directly, uses verified data, and includes sources. If you want to produce that kind of content without fabricating references, Studio analyses real sources before writing.

What still works the same way

A necessary clarification before anyone throws everything out.

Searches with direct purchase intent (buy X, price of Y, hire Z) still generate clicks at a significantly higher rate than informational content. If you have a shop or a service with a clear price, technical SEO is still your foundation.

Local SEO is not dead either. A search like emergency plumber in Manchester still leads to Google Maps, not an AI response.

And E-E-A-T (experience, expertise, authority, trust) remains the underlying criterion for both Google and AI engines. A website with shallow content will not appear anywhere, neither in search results nor in a ChatGPT response.

GEO replaces nothing. It adds a layer on top of what already works.

The window is open, but not indefinitely

In classic SEO, the first movers built domain authority that others took years to match. The same thing is happening with GEO, but at an earlier stage.

Today, most marketing teams do not have a documented strategy for appearing in AI responses. That space is unclaimed.

Citation authority in AI, like domain authority before it, compounds. Whoever starts earlier starts with an advantage. And that advantage is hard to close once others have consolidated it.

This is not about abandoning what works. It is about adding the layer that is missing.

The next step is not technical: it is checking whether the first 200 words of your most important content directly answer the question your customer would ask an AI. If you are still unclear on how to measure what happens when Google answers before you do, this analysis on zero-click in 2026 is the natural starting point.

Start there.

Sources