AI Search7 min read

    ChatGPT SEO: How to Get Found and Cited in ChatGPT Search

    ChatGPT SEO is first and foremost crawlability and clarity. Here are OpenAI's crawlers, robots.txt, the Bing index and what you should actually do.

    ChatGPT SEO is the work of making a website possible to find, understand and cite when ChatGPT searches the web to answer a question. Most of it is classic technical SEO: the content has to be crawlable, indexed and unambiguous. What is new is that there are more crawlers to deal with, and that the answer is a piece of text with source citations rather than a list of ten links.

    This guide sticks to what OpenAI and the other providers document themselves, and clearly marks where no official answers exist. The status is as of October 2026.

    What is ChatGPT SEO?

    When ChatGPT searches the web, it typically rewrites the user's question into one or more targeted queries, retrieves results and writes an answer with links to its sources. So there are two things you can become: a source the answer is built on, and a link the user can click.

    OpenAI states that any public website can appear in ChatGPT search, and that results are ranked using multiple factors intended to help users find relevant, reliable information. OpenAI also makes clear that placement is not guaranteed. There is no official list of ranking factors, which is why it is more productive to focus on the prerequisites than on tricks.

    OpenAI's crawlers: OAI-SearchBot, GPTBot and ChatGPT-User

    OpenAI uses several user agents, each with its own purpose. Keeping them apart matters, because each one is controlled separately in robots.txt.

    • OAI-SearchBot is used to surface websites in ChatGPT's search features. If you block it, your content will not be shown in ChatGPT search answers, according to OpenAI.
    • GPTBot is used to collect content that may be used to train OpenAI's models. A Disallow for GPTBot is the signal that the content should not be used for training.
    • ChatGPT-User is used when a user in ChatGPT or a Custom GPT triggers an action, for example by asking it to open a specific page. OpenAI notes that robots.txt rules may not apply here, because the action is initiated by a user.

    There is also OAI-AdsBot, which checks ad landing pages.

    robots.txt example

    Because the crawlers are controlled separately, you can be visible in ChatGPT search and still opt out of training:

    User-agent: OAI-SearchBot
    Allow: /
    
    User-agent: GPTBot
    Disallow: /
    

    Whether to block GPTBot is a business decision – not an SEO question. What matters for search visibility is that OAI-SearchBot has access. According to OpenAI, changes to robots.txt take around 24 hours to take effect in their systems.

    If, on the other hand, you want to keep a page out of ChatGPT search entirely, OpenAI points to the noindex meta tag.

    Crawlability: the technical checklist

    Most visibility problems in AI search are perfectly ordinary crawl problems.

    1. Check robots.txt. Look for blanket rules such as User-agent: * with Disallow: /, and for rules a security or SEO plugin has added for "AI bots". Many of them block OAI-SearchBot along with GPTBot.
    2. Check your firewall and CDN. Bot protection can reject crawlers even when robots.txt allows them. OpenAI publishes the IP addresses for each crawler (for example openai.com/searchbot.json), so you can verify the traffic and allow it.
    3. Serve the content as HTML. OpenAI does not document whether its crawlers execute JavaScript. The safe choice is to have headings, body text and links in the HTML the server sends.
    4. Return the right status codes. Pages that should be citable must respond with 200 – with no login wall, cookie wall or redirect chains in the way.
    5. Look at your server logs. Search for OAI-SearchBot to see whether, and how often, the crawler actually visits the site.

    The llms.txt file is often mentioned in the same breath, but it is not part of OpenAI's guidance for publishers. We cover its status and use in our guide to llms.txt.

    The Bing index and other search providers

    OpenAI states that ChatGPT search also partners with other search providers and that the rewritten queries are sent to them. OpenAI's help pages do not name the providers. Microsoft Bing is widely reported to be one of them, but that is not something we can confirm from OpenAI's current documentation.

    The practical consequence is simple all the same: make sure your site is indexed in more search engines than Google. Add the site to Bing Webmaster Tools, submit your sitemap and fix the crawl errors the tool reports. It is a small task that many sites have never completed, and it pays off regardless of which providers an AI assistant draws on.

    Structured data and entities

    Precision is needed here. OpenAI's documentation for publishers says nothing about structured data being used as a signal in ChatGPT search. Google similarly states that structured data is not required to appear in its generative AI features – while still recommending the markup as part of an overall SEO strategy.

    The value lies elsewhere: in unambiguity. An AI assistant that is going to mention your company has to be able to work out who you are, what you offer, and whether information across sources refers to the same company. That takes consistency:

    • The same facts everywhere. Name, address, company registration number, services and prices should match on the site, in the markup and in your profiles.
    • One coherent identity. An Organization with a stable @id and sameAs references to your official profiles, which the rest of the site's markup points to. This is the core of entity governance.
    • Markup that reflects the visible text. Structured data that says something different from the page creates doubt instead of clarity.

    Correct schema markup also helps the search engines AI assistants retrieve results from. You will find the fundamentals in our guide to structured data, and we go deeper into the interplay with agents in structured data for AI agents.

    Content that can be cited

    An answer from ChatGPT consists of short, concrete statements with a source behind them. Content is easier to use when it:

    • answers the question directly in the first lines of a section
    • uses headings that describe what the section is about
    • contains specific facts, figures and definitions rather than general phrases
    • has a clear author or publisher and a last-updated date

    These are the same qualities that make content useful to people. Google's guidance states that you neither need to write in a specific way for generative AI nor break your content into tiny pieces.

    Other AI assistants

    The principle is the same with the other providers, but the user agents and rules differ:

    • Google AI Overviews and AI Mode: No special requirements. The page has to be indexed and eligible to be shown with a snippet in Google Search. Read more in our guide to Google AI Overviews.
    • Perplexity: PerplexityBot is used to surface and link websites in search results and respects robots.txt. Perplexity-User fetches pages on behalf of a user and, according to Perplexity, generally ignores robots.txt.
    • Claude: Anthropic uses ClaudeBot for training data, Claude-SearchBot for search and Claude-User when a user asks Claude to fetch a page. According to Anthropic, all three respect robots.txt.

    The pattern repeats: one crawler for search, one for training and one for user-initiated fetches. If you want to be visible, the search crawlers are the ones that need access.

    How to measure traffic from ChatGPT

    ChatGPT automatically adds the parameter utm_source=chatgpt.com to the links users click. That lets you see the traffic as a separate source in your analytics tool without any extra setup. Combine it with your server logs, so you can see both whether the crawler visits the site and whether those visits turn into clicks.

    Bear in mind that clicks are only part of the picture. A company can be mentioned in an answer without the user clicking through.

    Frequently asked questions

    What is ChatGPT SEO?

    ChatGPT SEO is the work of making a website possible to find, understand and cite in ChatGPT's search answers. In practice it is about crawlability, access for OAI-SearchBot, indexing in search engines and unambiguous content with a clear publisher.

    Do I need to allow GPTBot to appear in ChatGPT?

    No. GPTBot collects content for model training, while OAI-SearchBot is the crawler used for ChatGPT's search features. They are controlled separately in robots.txt, so you can allow search and opt out of training.

    How quickly does a robots.txt change take effect?

    OpenAI says it takes around 24 hours for a robots.txt change to take effect in their systems. There is no guarantee as to when your content then shows up in answers.

    Does ChatGPT use Bing for its searches?

    OpenAI states that ChatGPT search partners with other search providers but does not name them in its help pages. Bing is widely reported to be one of them. Either way, it is a good idea to make sure your site is indexed in Bing via Bing Webmaster Tools.

    Does structured data help in ChatGPT?

    OpenAI does not document structured data being used as a signal. The markup does make your facts unambiguous and consistent, and it helps the search engines AI assistants retrieve results from. Treat it as part of the foundation, not as a shortcut.

    Can I guarantee a place in ChatGPT's answers?

    No. OpenAI states that placement in ChatGPT search is not guaranteed and that results are ranked using multiple factors for relevance and reliability. What you can control is access, indexing and the quality of your content.

    Summary

    ChatGPT SEO is first and foremost technical hygiene: give OAI-SearchBot access in robots.txt, make sure your firewall and CDN are not blocking it, serve the content as HTML, and get indexed in more search engines than Google. On top of that comes unambiguity – consistent facts, a clear publisher and structured data that reflects the visible text. Nobody can guarantee a place in the answers, but without those prerequisites the opportunity is not there at all.

    ChatGPT SEOAI searchrobots.txt