GEO stands for Generative Engine Optimization. It is the practice of making your website readable, understandable, and citable by AI-powered search tools. The tools in question include ChatGPT Search, Perplexity, and Google AI Overviews. Each works differently from a traditional search engine, and those differences change what you need to do to show up.
How AI search differs from traditional search
A traditional search engine returns a ranked list of links. A generative search engine produces a written answer and cites the sources it drew from. Getting onto that citation list is not the same as ranking on page one of Google. You do not need to be the most authoritative page on the web. You need to be a page the AI can read completely, understand clearly, and trust enough to quote.
Which AI tools this affects and which bots they send
Each major AI search platform crawls the web with its own bot before generating answers. Your robots.txt governs which of them can access your site:
- •GPTBot: OpenAI's training crawler. Blocking it opts your content out of training future OpenAI models.
- •OAI-SearchBot: OpenAI's search crawler. It decides whether your pages can show up in ChatGPT search, and it is governed by its own user-agent rule, separate from GPTBot.
- •ClaudeBot: Anthropic's training crawler. Anthropic runs Claude-SearchBot for search and Claude-User for pages a person asks Claude to read.
- •Google-Extended: a robots.txt token, not a separate crawler, that controls whether Gemini can train on your content and use it to ground live Gemini answers. It does not affect Google Search or AI Overviews, which run on regular Googlebot crawling.
- •PerplexityBot: Perplexity's search crawler. It surfaces and links sites in Perplexity search and is not used for training. Perplexity-User fetches a page when a person asks about it and generally ignores robots.txt.
The four signal groups that determine GEO readiness
GEO optimization is not one thing. It is a combination of four categories of signals that AI crawlers evaluate when they visit your site:
- •Bot access: whether your robots.txt explicitly permits AI crawlers rather than blocking them or leaving them unaddressed.
- •Schema readiness: structured data markup that tells AI tools who you are and what you publish. Entity schemas like Organization establish your identity. Content schemas like Article and BlogPosting say what kind of content you publish in a form machines read directly.
- •Entity clarity: whether AI tools can confidently identify your business name, your location, and named individuals from your structured data and page metadata.
- •Supporting signals: whether your sitemap and robots.txt are reachable, which affects how easily crawlers find your pages.
What schema does for AI search
Structured data tells machines, in a form they read directly, who you are and what a page is. Organization schema states your business name, address and profiles. Article and BlogPosting describe editorial content. BreadcrumbList shows where a page sits in your site. None of this guarantees a citation: no AI platform documents schema as a ranking or citation factor, and Google says its AI features need nothing beyond what an ordinary search result needs. What schema does is remove guesswork about who you are. FAQPage warrants a specific note. Google dropped all FAQPage rich results in May 2026, completing a rollback that began with a restriction to healthcare and government sites in August 2023, so it no longer produces a rich result for any site. It still makes your questions and answers machine-readable, but there is no evidence that it drives AI citations.
Where GEO and traditional SEO overlap
Many GEO signals reinforce good SEO practice. A sitemap helps Googlebot and GPTBot alike. Schema markup has always helped search engines understand content and now helps AI tools cite it. Making sure bots see real page content rather than a JavaScript loading state matters for both. The foundations overlap significantly. The audit criteria differ.
Where to start
A practical checklist for most small business sites:
- •Check your robots.txt for the search bots. OAI-SearchBot, Claude-SearchBot and PerplexityBot decide whether ChatGPT, Claude and Perplexity search can show your pages. GPTBot and ClaudeBot are training crawlers, and Google-Extended covers Gemini training and grounding, not Google Search or AI Overviews, so blocking those three is a choice about training, not about appearing in search.
- •Add Organization schema to your homepage. At minimum include your business name, URL, and description as a JSON-LD block, so machines do not have to guess who you are.
- •Add Article or BlogPosting schema to content pages, so each one says what it is, who wrote it and when.
- •Make sure your sitemap exists and is referenced in your robots.txt.
- •Check that key page content is in the HTML. AI crawlers may not run JavaScript, and in our test of 1,700 homepages, 113 of 118 sites that build their content with JavaScript sent AI crawlers an empty page.
More on GEO and AI visibility
If you want to act on any of the signals above, these guides go deeper on each part of GEO.
The specific GEO signals Stackra checks, from bot access to schema diversity, and how they map to AI citations.
A practical, non-technical checklist for getting your small business cited by AI search tools.
The most common reasons ChatGPT and Perplexity skip a business, and how to fix each one.
How to measure referral traffic from AI tools using GA4's new AI channel.
A five-minute check to confirm AI crawlers can actually reach your pages.
What GA4's AI assistant channel reports and how to read it.
The structured data that tells AI tools who you are and what you publish.