How to Rank in AI Search; A Practical Guide to GEO
Picture a customer asking ChatGPT, “What’s the best project management tool for a small agency?” The answer names three products and links to a few sources. Your site ranks on page one of Google for that topic, yet you’re not mentioned anywhere.
That gap is why generative engine optimization, or GEO, has become a hot topic. It’s the practice of making your content easy for AI search tools to find, trust, and cite.
This guide keeps it practical. You’ll learn what GEO is, what actually works, what’s hype (yes, we’ll talk about llms.txt), and how to measure your AI search visibility.
What Is Generative Engine Optimization?
Generative engine optimization improves how often, and how favorably, your content appears in AI-generated answers. Those answers come from tools like ChatGPT search, Google’s AI Overviews and AI Mode, and Perplexity.
The term comes from a 2023 research paper by Aggarwal and colleagues, presented at the KDD 2024 conference. In their tests on a benchmark of queries, adding citations, quotations from relevant sources, and statistics boosted visibility in generated answers by up to 40%. That was a controlled experiment, not a promise for every site, and results varied by topic.
GEO vs SEO: What’s Different?
Traditional SEO | GEO | |
|---|---|---|
Goal | Rank high in a list of links | Be cited or mentioned inside an answer |
Result | Ten blue links | One synthesized answer with a few sources |
Key signals | Relevance, links, page experience | Clear answers, evidence, trusted brand mentions |
Measure | Rankings, clicks | Citations, mentions, referral visits |
Here’s the reassuring part: they overlap heavily. Google says its AI Overviews and AI Mode use the same foundations as regular Search, and that they don't require additional requirements or special optimizations to appear. So think of GEO as SEO plus a few extra habits, not a replacement.
How AI Search Picks Its Sources
Each platform works differently, but the pattern is similar. The AI breaks a question into related searches, retrieves pages, reads them, and writes an answer with citations.
Google describes this as “query fan out”: AI Overviews and AI Mode can run multiple related searches across subtopics to build one response. That’s why covering related questions on your page matters. To be shown as a supporting link, a page must be indexed and eligible to appear in Google Search with a snippet.
Step 1: Get the Technical Basics Right
If AI tools can’t reach your pages, nothing else matters.
Make sure your pages are indexed. Check Google Search Console for errors.
Don’t block useful crawlers by accident. Review your robots.txt file and security plugins.
Keep pages fast and mobile friendly, with real text in the HTML rather than text hidden inside images.
Use clear headings and normal semantic HTML. Google says the same structure works for people and AI.
Step 2: Write Answer First, Citable Content
This is where GEO differs most from old school keyword writing. Aim to be the page that makes an answer easy to quote.
Lead with the answer. Put a clear two sentence answer under each question heading, then explain.
Add real evidence. Use statistics, dated facts, and named sources. Never invent them, since AI systems and readers can both spot weak claims.
Quote and cite experts where it helps the reader.
Share original information. Your own data, case studies, pricing, or process notes give AI something other pages don’t.
Cover related questions so fan-out searches find you.
Keep content fresh. Update key pages and show the date.
A caution: Google’s guidance says you don’t need to chop content into tiny chunks or write in an “AI-friendly” style. Write clearly for people. Good structure helps both audiences.
Before and After: A Quick Example
Here is how one section can change. Imagine a page for a small accounting firm.
Example | |
|---|---|
Before | We offer a wide range of tax services tailored to your needs. Contact us to learn more about how we can help your business grow. |
After | Most small businesses must file a yearly tax return and keep clean records. We prepare both, and we list our fixed fees on this page. |
The second version answers a real question, names specifics, and gives an AI tool something concrete to quote. Notice the bracket: fill it with facts you can prove, never made up numbers.
Step 3: Build Trust Beyond Your Website
AI tools learn about you from the whole web. Google’s guidance notes that AI systems look at brand mentions across the web and can tell real mentions from fake ones.
Keep your business name, address, description, and services consistent everywhere.
Earn genuine reviews, directory listings, and mentions from respected sites in your field.
Publish an About page and author bios that show who is behind the content.
Skip fake mentions and bought links. They don’t fool the systems and can hurt you.
Step 4: Use Structured Data Sensibly
Schema markup is not required to appear in Google’s AI features. It still helps your pages qualify for rich results in normal search, and it labels details like products, reviews, and FAQs clearly. Add it where it fits your content, but don’t expect it to unlock AI answers by itself.
What About llms.txt?
llms.txt is a proposal from Jeremy Howard of Answer.AI, published in September 2024. It’s a Markdown file at your site’s root that lists your key pages for AI tools. It was designed mainly to help AI coding assistants read documentation.
For AI search visibility, the evidence is weak:
Google says Search doesn’t use llms.txt, and that creating one won’t help or harm your visibility in Google Search.
Reports on crawler logs show AI bots rarely request it. One study counted 84 requests to the file out of more than 62,000 AI bot visits.
Major AI companies have not publicly committed to using it for answers.
Verdict: it’s harmless to add, and useful if you run developer documentation. But don’t expect it to get you cited, and don’t spend your week on it.
How to Rank in ChatGPT
Many people search for how to rank in ChatGPT. There’s no secret switch, but there is one concrete control. OpenAI uses separate crawlers for separate jobs:
Crawler | What it does |
|---|---|
OAI SearchBot | Surfaces websites in ChatGPT search results. Sites that opt out won’t be shown in ChatGPT search answers, though they can still appear as navigational links. |
GPTBot | Controls whether content may be used to train OpenAI’s foundation models. It is independent of search. |
ChatGPT User | Used for certain user-triggered actions. Robots.txt rules may not apply, and it isn’t used to decide search inclusion. |
OpenAI recommends allowing OAI-SearchBot in robots.txt if you want to appear in ChatGPT search. Here is an example that stays visible in search while opting out of training:
User-agent: OAI-SearchBot
Allow: /
User-agent: GPTBot
Disallow: /
Whether to block GPTBot is your call. Blocking it does not remove you from ChatGPT search, but blocking OAI-SearchBot does.
How to Track AI Search Visibility
Measurement is still messy, so keep it simple:
Test prompts monthly. Ask ChatGPT, Google AI Mode, and Perplexity 10 to 20 questions your customers ask. Record whether you’re mentioned and cited.
Check Search Console. Google says traffic from AI Overviews and AI Mode is counted in your overall Search Console data.
Watch analytics referrals from chatgpt.com, perplexity.ai, and similar sources.
Review server logs for OAI SearchBot and other AI crawlers to confirm they can reach you.
A 30 Day GEO Action Plan
Week | Focus |
|---|---|
Week 1 | Check indexing and robots.txt. Allow OAI SearchBot if you want ChatGPT search visibility. Record your baseline prompt tests. |
Week 2 | Rewrite your top five pages with answer first sections, a related questions block, and sourced facts. |
Week 3 | Fix brand consistency: About page, author bios, business listings, and profile details. |
Week 4 | Add a few useful pieces of original data or examples, then repeat your prompt tests and compare. |
Myths to Skip
“Guaranteed AI rankings.” Nobody controls AI answers. Treat such promises as a red flag.
“llms.txt is the new robots.txt.” Robots.txt is honored by crawlers. llms.txt isn’t.
“You need to rewrite everything for AI.” Clear writing for people is the foundation.
“SEO is dead.” AI features still rely on the same search fundamentals.
References
Frequently asked questions
What is generative engine optimization (GEO)?
Generative engine optimization is the practice of improving how often your content is cited or mentioned in AI-generated answers from tools like ChatGPT, Google AI Overviews, AI Mode and Perplexity. It builds on SEO and focuses on clear answers, evidence and trusted brand signals.
What is the difference between GEO and SEO?
SEO aims to rank pages in a list of links, while GEO aims to get your content cited inside an AI-written answer. They overlap heavily, and Google says the same SEO fundamentals apply to its AI features.
How do I rank in ChatGPT?
There is no guaranteed method, but you can make your site eligible. Allow OAI-SearchBot in your robots.txt, keep pages indexable and publish clear, well-sourced content that answers real questions.
Does GEO work?
The original GEO research found that adding citations, quotations and statistics boosted visibility by up to 40% in tests on a benchmark. That was a controlled study, so real-world results vary by topic and platform.
What is llms.txt?
llms.txt is a proposed Markdown file listing a site’s key pages for AI tools, suggested by Jeremy Howard in September 2024. It is a community proposal, not an official web standard.
Does llms.txt help AI search visibility?
The evidence says no for AI search. Google states Search does not use llms.txt, and studies of crawler logs show AI bots rarely request it. It can still be useful for developer documentation read by coding assistants.
Should I block GPTBot?
That depends on how you feel about your content being used for model training. Blocking GPTBot does not remove you from ChatGPT search, but blocking OAI-SearchBot does.
What is the difference between OAI SearchBot, GPTBot and ChatGPT User?
OAI SearchBot supports ChatGPT search results, GPTBot relates to training foundation models, and ChatGPT User handles certain user-triggered actions. OpenAI says each setting is independent.
How do I improve my AI search visibility?
Make sure pages are indexed, lead with clear answers, add sourced facts and original information, build consistent brand mentions across the web, and update key pages regularly.
Do I need structured data for AI Overviews?
No. Google says no special structured data is required to appear in AI Overviews or AI Mode, though schema markup can still help with rich results in regular search.
How can I track whether AI tools mention my brand?
Run monthly test prompts in ChatGPT, Google AI Mode, and Perplexity; check Search Console and analytics referrals; and review server logs for AI crawlers.
Does GEO replace SEO?
No. AI features in Google rely on core Search systems, so strong SEO remains the foundation. GEO adds habits aimed at being cited in generated answers.
Is GEO worth it for a small business?
Yes, mostly because the basics are cheap: clear content, accurate business information and crawler access. Avoid agencies that guarantee AI rankings, since no one controls AI answers.
Written by
Rehan Naeem
Spotted a mistake? Report a correction · How we handle corrections