LLM SEO: how to get named by ChatGPT, Gemini and Perplexity

What is LLM SEO?

LLM SEO, also called SEO for LLMs or LLM search optimization, is the practical side of AI visibility. It is the checklist you work through so that assistants use your brand in their answers. The goal is the same as in generative engine optimization — the name simply starts from SEO instead of from research.

A note on words. “LLM optimization” also means making models run faster or cheaper, a developer topic about inference and fine-tuning. This page is about the marketing meaning: how a model talks about your brand.

How LLMs decide which brands to name

A large language model answers from a mix of memory and lookup. Its memory is the training data — a snapshot of the web that shaped what it “knows” about your category. The lookup happens when the assistant searches the web, reads a handful of pages and writes its answer from them.

Questions about prices, current products or this year's best options are more likely to need fresh information than a general definition. So for LLM SEO you have two jobs. Make the pages a searching assistant would read reachable and clear, and make sure the wider web describes you the way you want the next model to learn it.

Let AI crawlers in

Each AI company runs its own crawlers, and several separate search from training. You can decide, crawler by crawler, whether you want to appear in answers, feed future models, or both. Check robots.txt and also your CDN or firewall rules, which can block these bots without anyone noticing.

Five user agents cover ChatGPT, Perplexity and Gemini. Look for them by name:

• OAI-SearchBot — OpenAI uses it to surface websites in ChatGPT's search features. OpenAI's crawler documentation says sites that opt out will not be shown in ChatGPT search answers, and that robots.txt changes can take about 24 hours to take effect for search.
• GPTBot — crawls content that may be used to train OpenAI's foundation models. OpenAI says each setting is independent, so you can allow OAI-SearchBot and still disallow GPTBot.
• ChatGPT-User — visits a page when a ChatGPT user's request calls for it, rather than crawling on a schedule.
• PerplexityBot — surfaces and links websites in Perplexity's results. Perplexity's documentation says it is not used to crawl content for AI foundation models.
• Google-Extended — a token that controls whether content Google crawls may be used to train future Gemini models and for grounding in Gemini apps. Google says it does not affect inclusion or ranking in Google Search.

Make your pages easy to quote

Once a crawler can reach a page, the page has to be worth quoting. Assistants lift short, self-contained statements: a definition, a price range, a “best for” line, a comparison. Write those sentences on purpose, near the top of the page, in plain text.

• Give each buyer question its own answer. “Best running shoes for flat feet” deserves a clear section, not one line on a category page.
• Say who you are for and who you are not for. Recommendations usually come with a reason, and a clear “best for” line gives the model one.
• Keep product facts in HTML text, not only in images, PDFs or scripts that crawlers may skip.
• Update facts in one place and make every page reflect them. Old pricing on a forgotten page can end up in an answer.
• Publish honest comparisons with named alternatives. Comparison questions are where assistants make recommendations.

Get described on the pages models trust

Your own site is only one voice. When an assistant answers a category question, it may draw on review sites, retailer listings, editorial roundups, forums and reference pages. Those sources also make up much of what the next model learns about you.

Start with the sources the engines already use for your questions. Perplexity lists its sources with its answers, and ChatGPT and Gemini can show theirs when they search. Get listed where you are missing, fix outdated descriptions and ask real customers for reviews on the platforms your buyers read.

ChatGPT SEO: what is different about ChatGPT

ChatGPT SEO — or SEO for ChatGPT — is LLM SEO aimed at one engine. ChatGPT answers some questions from what the model already knows and searches the web for others, and when it searches, it shows its sources. That gives you two levers: the pages it can find today and the reputation the next model will learn.

The technical part is short. Allow OAI-SearchBot, keep your pages indexable in regular search engines and put key facts in text. The content part is the same as for any assistant, but test it on ChatGPT itself — its brand list for a question can differ from what Gemini or Perplexity say.

People often ask how to check their “ChatGPT ranking”. There is no fixed ranking to check, because the brands named and their order can change between runs. What you can measure is how often you appear, how often you are recommended and where you tend to sit among the brands named.

Gemini, Perplexity and other engines

The basics carry across engines, but each has a detail worth knowing. None of them replaces the others, so plan to check each one separately.

• Gemini — Google-Extended decides whether Google may use your content for Gemini training and grounding, while your Google Search indexing stays unaffected.
• Perplexity — it shows sources next to its answers, which makes it the quickest place to see which pages shape your category.
• Google AI Overviews and AI Mode — Google says standard SEO is enough to be eligible. The practical steps are in how to rank in AI Overviews.
• Microsoft Copilot — citations of your pages in Copilot and Bing's AI summaries appear in the AI Performance report in Bing Webmaster Tools.
• Claude — worth checking too if your buyers use it. Like Copilot, it can be added to the Llamat report as a paid add-on.

LLM SEO, GEO and AEO

LLM SEO, generative engine optimization and answer engine optimization describe one goal. GEO is the umbrella term from research, AEO grew out of featured snippets, and LLM SEO is the name SEO teams give the checklist. The practical differences are small, and AEO vs GEO goes through the ones that matter.

Measure before and after

You cannot see a model's answer to your buyer unless you ask the same question yourself. Pick the category questions that matter, run each one on each engine more than once and record whether you are mentioned, recommended and cited. Then repeat after your changes, with the same questions.

The free AI visibility checker does the first round for you: up to 20 questions on ChatGPT, Gemini and Perplexity, with the competitors named instead of you, in one e-mail report. Google AI Overviews, Google AI Mode, Claude and Microsoft Copilot can be added as paid add-ons, and repeated monitoring is available on request.