How to get recommended by ChatGPT and Google AI
How to get your course into ChatGPT and Google AI answers: what Google and OpenAI actually say, robots.txt settings, citable pages and what you don't need.
More and more people ask an AI assistant which course to take, which tool to use or who to follow on a topic. The answers cite some pages and ignore others. For creators selling courses, being among the cited pages is a new way to be found.
This guide separates what the companies say officially from what circulates as industry advice. We read Google Search Central's documentation on AI features and OpenAI's documentation on its crawlers. Anything else we flag as our own experience.
How do AI assistants choose their sources?
Assistants with web search look for relevant pages, read them and build an answer that cites some of them. Google explains that AI Overviews and AI Mode may use a technique called "query fan-out": running multiple related searches across subtopics and data sources to build a response, which can surface a wider and more diverse set of links than classic search.
For creators, this means you don't have to rank first for one keyword: a page that answers a specific question well ("how long should a video course lesson be?") can be cited in answers to broader questions.
What does Google say about AI Overviews and AI Mode?
Google says there are no additional requirements to appear in AI Overviews or AI Mode, and no special optimisations needed. A page must be indexed and eligible to be shown in Google Search with a snippet; beyond that, there are no extra technical requirements. Google also notes that indexing and serving aren't guaranteed.
The best practices Google lists are the usual SEO ones:
- allow crawling in robots.txt and in any CDN or hosting infrastructure;
- make content easy to find through internal links;
- provide a good page experience;
- make sure important content is available as text;
- support text with high-quality images and videos where relevant;
- make sure structured data matches the visible text on the page;
- keep Merchant Center and Business Profile information up to date.
| Required | Recommended by Google | |
|---|---|---|
| Page indexed and eligible for a snippet | Yes | Yes |
| Crawling allowed (robots.txt, CDN) | Yes | Yes |
| Helpful, reliable, people-first content | Advised | Yes |
| Structured data that matches visible text | Advised | Yes |
| llms.txt or other AI files | No | No |
| Special AI markup or schema.org | No | No |
Our summary of Google Search Central, 'AI features and your website', last updated 10 December 2025.
Google is explicit about what you don't need: no new machine-readable files, AI text files or markup, and no special schema.org structured data. If someone sells you a "GEO package" built on these, ask which source it's based on.
How to limit what appears
To keep content out of AI answers, Google points to the same controls as classic search: nosnippet, data-nosnippet, max-snippet or noindex. Google-Extended, by contrast, governs whether content is used for training and grounding in some of Google's other systems, not whether it appears in Search.
What do you need to appear in ChatGPT?
OpenAI's search crawler must be able to read your pages. OpenAI documents three main agents with different purposes:
- OAI-SearchBot is for search. OpenAI says sites that opt out won't be shown in ChatGPT search answers (they can still appear as navigational links) and recommends allowing it in robots.txt.
- GPTBot collects content that may be used to train generative AI models. Disallowing it signals that your content shouldn't be used for training.
- ChatGPT-User visits pages when a user asks ChatGPT to. It doesn't crawl automatically and, OpenAI notes, robots.txt rules may not apply.
OpenAI says each setting is independent of the others and that it can take about 24 hours after a robots.txt change for its systems to adjust. An example robots.txt for a creator who wants to appear in search but not in training:
User-agent: OAI-SearchBot
Allow: /
User-agent: GPTBot
Disallow: /
Also check that your host or CDN isn't blocking these crawlers with anti-bot rules: it's a common and invisible problem.
How do you write a page that assistants cite?
This part doesn't come from official documentation but from our work on this blog, where we apply the same method. A citable page has four traits:
- A question as the section heading and the answer right below. Two sentences that make sense on their own, with no "as mentioned above".
- Verifiable facts with sources. Numbers, dates, rule names, linked to the original document with an access date.
- A recognisable author. Name, role, why they know the subject, a contact.
- A real FAQ, with complete answers: it's the closest format to how people query an assistant.
The same applies to your course page: curriculum, length, price, who it's for and refund terms written as text, not only in images or video. An assistant can't recommend a course whose basic information it can't read. The structure is in how to write a sales page for an online course.
An example from a course page
Here's the difference on a real page. Instead of a paragraph that says "a complete path to better photos", a citable page says: "The course runs for 6 weeks, has 24 lessons of 10-15 minutes, costs $149 and is for photographers who already own a DSLR but only shoot in auto mode." If someone asks an assistant which short beginner photography course costs under $200, the second sentence already holds every fact needed to answer. It's also verifiable: the assistant can check it against the rest of the page and against what others write about the course. The first sentence, by contrast, contains nothing that can be cited. In other words, write the facts a person would ask for, in complete sentences, and put them in visible text, not only inside a video or an image. The course in this example is made up.
Why do links and external mentions still matter?
Because assistants build answers from the pages they find, and pages are found partly by following links. Google lists internal links among its best practices; links from other sites remain one of the ways search engines discover and assess pages, as we explain in SEO for creators.
For creators, the most natural ways to get mentioned are interviews and podcasts with a page that links to you, collaborations with other creators, guest articles and genuinely useful free resources (a template, a calculator) that others recommend. An affiliate program helps too: people who promote you write about you, as covered in affiliate programs for online courses.
How do you measure traffic from AI assistants?
Google counts traffic from AI features in Search Console's Performance report, under the Web search type, alongside other Google visits. For ChatGPT and other assistants, filter your site analytics for visits referred from their domains.
Bear in mind that some people read the answer without clicking. So also track what people tell you: when someone writes "ChatGPT recommended your course", note it down. Other useful metrics are in creator metrics that matter.
Which mistakes should you avoid?
- Blocking crawlers without knowing it. Host anti-bot rules or a copied robots.txt can lock you out.
- Key information only in images or video. Price, curriculum and terms must also be text.
- Paying for "AI" files or markup. Google says they aren't needed.
- Structured data that differs from the text. Google asks that it match what people see.
- Generic content written for engines. Assistants cite whoever answers best, not whoever repeats a keyword most.
Key terms. AI Overviews are AI-generated summaries at the top of some Google results; AI Mode is Google's conversational search mode. A crawler is a program that visits and reads web pages. robots.txt is the file that tells crawlers which parts of a site they may visit.
How we checked. We read Google Search Central's "AI features and your website" (last updated 10 December 2025) and OpenAI's overview of its crawlers. The section on writing citable pages reflects the method we use on this blog, not a company rule. Contact: hi@heytea.app. Editorial note: we check every figure against the sources listed below and correct mistakes openly.
Frequently asked questions
How do you get recommended by ChatGPT?
When ChatGPT searches the web it uses its own crawler, OAI-SearchBot. OpenAI says sites that block it won't be shown in ChatGPT search answers, and recommends allowing it in robots.txt. Beyond that, you need clear pages that answer your audience's questions with verifiable facts, and that other sites cite and link to.
Do AI Overviews and AI Mode need special optimisation?
No, according to Google. Google Search Central's documentation says there are no additional requirements or special optimisations: a page must be indexed and eligible to appear in Google Search with a snippet. The usual SEO best practices apply: helpful, reliable content, crawling allowed, internal links and readable text.
Do I need an llms.txt file to appear in AI answers?
Not for Google: its documentation says you don't need to create new machine-readable files, AI text files or special markup to appear in AI Overviews and AI Mode. Other services might use such files, but none of the official sources we read requires one.
If I block GPTBot, do I disappear from ChatGPT?
No, they're separate. GPTBot collects content that may be used to train OpenAI's models; OAI-SearchBot is for search. OpenAI says each setting is independent, so you can block GPTBot for training and allow OAI-SearchBot to appear in search answers.
How do you measure traffic from AI Overviews?
Google counts traffic from AI features in Search Console's Performance report, under the Web search type, together with everything else. For ChatGPT and other assistants, check visits referred from their domains in your analytics, knowing that some people read the answer without clicking.
Sources
- Google Search Central, "AI features and your website", last updated 10 December 2025. Accessed 18 October 2026. developers.google.com
- OpenAI, "Overview of OpenAI Crawlers". Accessed 18 October 2026. developers.openai.com