How to get your brand cited in ChatGPT and Perplexity

Something is shifting in how people find information online. A growing number of searches that used to go to Google now go straight to ChatGPT or Perplexity. Not every search — but the conversational, research-oriented ones. The “what’s the best way to…” and “which tool should I use for…” queries that are often worth the most to a business.

If your brand doesn’t appear in the answer, you don’t exist for that person. There’s no page two. There’s no organic result four spots down. There’s just the answer — and whoever got cited in it.

Getting cited isn’t luck. It follows a logic. Here’s what that logic looks like in practice.

How AI systems decide what to cite

ChatGPT and Perplexity work differently under the hood, but they share the same basic goal: give the user a reliable answer. The question is what signals they use to determine reliability.

Perplexity is a live search engine at its core. It crawls the web in real time, ranks pages, pulls text, and synthesises an answer from sources it considers authoritative. ChatGPT with browsing enabled does something similar. Both are looking for pages that state things clearly, come from credible sites, and are backed up by signals elsewhere on the web.

ChatGPT’s base model also draws on training data, which means older content that existed before any given training cutoff can influence what it “knows” about your brand. That’s useful context, but not something you can directly control — so the practical focus is on the live-crawl side.

In both cases, the content that gets cited tends to share a few characteristics: it directly answers a specific question, it’s written in a way that’s easy to extract a clean passage from, and it sits on a domain that other authoritative sources link to or reference. None of that is mysterious. It’s just a more demanding version of what good SEO has always asked for.

Write content that actually answers questions

Most business content is written to impress rather than to inform. It gestures at expertise without giving anything away. That’s understandable — the instinct is to make someone want to contact you, not read your article and go elsewhere. But AI systems need extractable answers, and vague thought leadership doesn’t give them anything to extract.

If your customers search “how much does a WordPress website cost in the UK,” write a page that actually answers that question with numbers, ranges, and the variables that affect the price. If they ask “what’s the difference between Laravel and WordPress,” write the clear comparison. Give the definitive answer, not the hedged non-answer.

This is the same instinct that drives featured snippets in Google. AI citation is that principle taken further. The goal is content that a machine can confidently quote — because it’s specific, accurate, and direct.

Structure content for extraction

How you present information matters as much as what you say. AI systems parse page structure when pulling content. A wall of prose is harder to cite cleanly than a short definition followed by numbered steps or a clear FAQ section.

Concise definitions at the top of a section work well. Step-by-step guides get pulled cleanly because the structure maps directly onto how someone might ask the question. FAQ sections — properly written ones, not fluff — are among the most consistently cited formats because the question-and-answer pairing is exactly what an AI response needs.

None of this means your writing should read like a manual. It means you should think about whether each piece of content answers a specific question, and whether that answer is findable within a few seconds of a machine reading the page.

Build brand entity signals

An “entity” in search terms is a clearly defined, consistently named thing — a person, a company, a place. AI systems have absorbed a lot of knowledge graph logic, which means they understand the world in entities, not just keywords.

For your brand, this means making sure you’re described consistently everywhere you appear online. Your company name, what you do, where you’re based, and who you serve should read the same way on your website, your Google Business Profile, industry directories, social profiles, and anywhere else you’re listed. Inconsistency creates ambiguity, and ambiguity is the enemy of citation.

If your brand could have a Wikipedia article written about it — even if it doesn’t — ask yourself whether someone reading all your online presence would have a clear, consistent picture of what you are. If the answer is “probably not,” that’s worth fixing before you invest heavily in content.

Earn third-party mentions

AI systems give weight to sources that other sources trust. This is borrowed credibility — the same principle that makes backlinks valuable in SEO, except here the signal extends to being mentioned, quoted, or referenced even without a link.

Getting listed in relevant industry directories matters. Being quoted in trade press or niche publications matters. Writing genuinely useful guest posts on credible sites matters. Not because any one mention will flip a switch, but because the pattern of third-party references builds the kind of authority profile that AI systems recognise.

PR is worth thinking about differently in this context. A press mention in a publication that AI systems consider authoritative isn’t just brand visibility — it’s a citation signal. The distinction between PR and GEO strategy is blurring.

Use schema markup

Schema markup is structured data that sits in your page’s code and tells crawlers — search engines and AI alike — exactly what type of content they’re looking at. An Article schema tells a crawler this is editorial content with an author and a publication date. An FAQ schema labels each question and answer explicitly. An Organization schema declares your brand name, URL, and description in an unambiguous, machine-readable format.

AI crawlers don’t need schema to read your content, but it helps them classify it correctly and confidently. The Author schema is worth particular attention — establishing authorship with a real, named person creates the kind of E-E-A-T signal (Experience, Expertise, Authoritativeness, Trustworthiness) that both Google and the AI systems drawing on Google’s authority assessments respond to.

If you’re on WordPress, most good SEO plugins handle the basics. The FAQ and Author schemas are worth implementing explicitly if your site doesn’t already have them.

Make sure AI crawlers can actually access your site

This is the most avoidable problem in the list, and it happens more often than it should. If your robots.txt file blocks AI crawlers, none of the above matters — they can’t read your content at all.

OpenAI’s crawler is called GPTBot. Perplexity uses PerplexityBot. There are others. Check your robots.txt file and make sure you’re not blocking them. Some security plugins and certain CDN configurations add crawler restrictions that were never intentionally aimed at AI, but catch them anyway.

It’s also worth checking that your most important pages aren’t accidentally noindexed, that your site doesn’t require JavaScript rendering before any readable content appears, and that your page load times are reasonable. Slow pages that time out during crawling get dropped from the index. The same principles that apply to Googlebot apply here.

What not to do

There’s a tempting shortcut that’s worth naming directly: publishing large volumes of AI-generated content in the hope that sheer presence creates citation authority. It won’t. The economics feel appealing — fifty pages in a week, all targeting questions your audience asks. But the quality ceiling is obvious, the content is hard to differentiate from thousands of similar pages, and AI systems are getting significantly better at identifying low-quality or derivative sources.

Thin content at scale was already a diminishing return in SEO. In GEO it’s actively counterproductive, because AI systems are built to synthesise quality sources — and a high volume of low-quality content tells the same story as one bad page. It signals that the site isn’t producing genuinely original information.

The answer is fewer, better pieces of content. One genuinely authoritative page on a question your customers actually ask is worth more than thirty AI-padded versions of the same thing.

Where to start

Pick the three questions your customers ask most often before they buy from you. Write one clear, direct, well-structured page for each. Add FAQ schema. Make sure your robots.txt isn’t blocking AI crawlers. Get listed in two or three industry directories you’re not already in. That’s a reasonable starting point — not exhaustive, but real.

Frequently asked questions

How do I get my website to appear in ChatGPT answers?

ChatGPT draws on training data and, when browsing is enabled, live web results. To increase your chances of being cited, focus on writing content that directly and specifically answers questions your audience asks, structuring that content clearly with headings and FAQ sections, and building authority signals through third-party mentions and backlinks. Make sure GPTBot isn’t blocked in your robots.txt file, and use schema markup to help the crawler understand what your pages are about.

What is brand citation in AI search?

Brand citation in AI search means your website or brand name being referenced as a source in an AI-generated answer — for example, when ChatGPT or Perplexity cites your page when responding to a user’s question. Unlike a traditional search result where your link appears in a list, a citation in an AI answer places your brand directly inside the response, which can be more visible and more trusted by the person reading it.

Does SEO help with ChatGPT and Perplexity?

Yes, substantially. The signals that make a website rank well in Google — clear content, authority backlinks, good technical health, structured data — also influence how AI systems assess source quality. Perplexity in particular draws heavily on live web search, so strong SEO performance and AI citation authority reinforce each other. The main difference is that AI systems place even more weight on direct, extractable answers, so content written purely for engagement or brand awareness tends to perform less well than content written to answer specific questions.

How does Perplexity decide which websites to cite?

Perplexity runs a live web search, ranks results using signals similar to a search engine, then pulls text from pages it considers authoritative and relevant. The key factors are whether the content directly answers the query, whether the page comes from a credible domain with good authority signals, and whether the text is structured in a way that can be extracted cleanly. Pages that are slow to load, poorly structured, or blocked in robots.txt are less likely to be crawled or cited.

How long does it take to get cited in AI search?

There’s no guaranteed timeline, and the honest answer is that it depends heavily on how competitive your topic area is and what authority your site currently has. For a site with reasonable existing authority, well-structured new content targeting specific questions can start appearing in Perplexity results within weeks. Building the broader entity signals and third-party mentions that influence ChatGPT’s base model is a longer-term project — measured in months, not days. The same patience that SEO requires applies here.