ChatGPT search visibility is how prominently ChatGPT’s web search features show and link your pages in their answers. The feature began as the SearchGPT prototype in 2024 and is now simply ChatGPT search. For a website owner, the first lever is access: OpenAI says a site that opts out of its search crawler will not be shown in ChatGPT search answers.
What is ChatGPT search, and how does it use your site?
When a question needs fresh information, ChatGPT searches the web and writes an answer with source links. OpenAI’s help center says any public website can appear. It also says ChatGPT ranks results using several factors and that placement is never guaranteed.
Three different OpenAI user agents touch websites, and each follows its own robots.txt setting:
| Bot | What it does | Matters for ChatGPT search? |
|---|---|---|
| OAI-SearchBot | Surfaces websites in ChatGPT’s search features | Yes: this is the one to allow |
| GPTBot | Crawls content that may train OpenAI’s foundation models | No. Disallowing it only signals “do not train on this site” |
| ChatGPT-User | Visits a page when a person (or a Custom GPT or GPT Action) asks for it | No. It is not used to crawl automatically, and OpenAI says robots.txt rules may not apply |
OpenAI also lists OAI-AdsBot, which only checks landing pages submitted as ads in ChatGPT. The settings are independent: you can allow OAI-SearchBot and block GPTBot.
Why does it matter for AI search?
ChatGPT is a place where buyers now ask “which tool should I use for X” before they open Google. A cited link there arrives pre-qualified. It also costs you nothing to be eligible. Blocking the search bot by accident removes you from that surface, and most blocks are accidental: a security plugin, a CDN bot rule or a User-agent: * block meant for scrapers.
How do you make your site visible to ChatGPT search, step by step?
- Read your current robots.txt. Open
yoursite.com/robots.txtand look forUser-agent: *blocks and any rule naming OAI-SearchBot. See robots.txt and noindex for the syntax. - Allow OAI-SearchBot. Add this, then decide on GPTBot separately:
User-agent: OAI-SearchBot
Allow: /
User-agent: GPTBot
Disallow: /The second block is optional: it keeps your content out of OpenAI model training while staying visible in search. Remove it if you accept training use.
3. Check your CDN and firewall. OpenAI recommends allowing requests from its published IP ranges, listed at openai.com/searchbot.json (and gptbot.json, chatgpt-user.json). A robots.txt “Allow” does nothing if Cloudflare or a WAF returns 403 to the bot.
4. Wait for the update. OpenAI says search systems need about 24 hours to adjust after a robots.txt change.
5. Make the page worth citing. Put a direct answer at the top, name your sources, and add numbers or facts only you have. The research behind this is on the generative engine optimization page, and the format is in answer-first content.
6. Keep pages fast and server-rendered. A bot fetching a page that needs heavy JavaScript to show text can come away with an empty shell.
7. Use noindex if you want no link at all. OpenAI’s FAQ says a page blocked from crawling can still show as a bare title and link; a noindex meta tag prevents that, but the crawler must be allowed to read the page to see it.
8. Check other engines’ bots too. The full list sits on the AI crawler access page, and the Microsoft side is covered in Bing and Copilot.
How do you measure ChatGPT search traffic?
OpenAI’s FAQ says ChatGPT adds the parameter utm_source=chatgpt.com to referral links, so visits show up in Google Analytics under that source. Build a GA4 report filtered on session source chatgpt.com.
Next, read your server logs. Search for OAI-SearchBot in the user-agent field and confirm that requests return status 200. A bot that appears with 403 or 429 codes tells you the firewall is the problem. For visibility, run a fixed list of 20 to 50 customer questions in ChatGPT monthly and note which sites are cited (see AI visibility tracking).
What mistakes should you avoid?
- Blocking all “AI bots” with one wildcard rule. This removes you from search along with training.
- Mixing up the three bots. Blocking GPTBot does not remove you from ChatGPT search, and allowing it does not put you in.
- Trusting
robots.txtto stop ChatGPT-User. OpenAI says user-initiated visits may ignore it. - Paying for “ChatGPT ranking” guarantees. OpenAI states that placement is not guaranteed.
- Forgetting the firewall. Many sites allow the bot in robots.txt and still block it at the CDN.
- Hiding your best content behind logins, pop-ups or scripts.
Where can you learn more?
OpenAI’s crawler overview is short and kept current, so check it before each robots.txt change. The free digital marketing module of the Fe/male Switch startup school covers the SEO base that AI tools build on. Next steps: set up the log check above and compare it with your AI visibility tracking results.