Before worrying about whether ChatGPT recommends your brand, check something simpler: can it read your website at all? Many sites block AI crawlers by accident, or show their key text only after JavaScript runs. Here are five checks you can do in about fifteen minutes, most of them without a developer. Or run our free checker, which runs the crawler, JavaScript and page-basics checks automatically and looks for your sitemap.
Minutes 0–3: check robots.txt
Open yourdomain.com/robots.txt in a browser. This small file tells crawlers what they may read.
Look for three names, each controlled separately:
- OAI-SearchBot: OpenAI’s search crawler. OpenAI says sites that block it “will not be shown in ChatGPT search answers”. This is the one that matters most.
- GPTBot: OpenAI’s crawler for training data. Blocking it doesn’t remove you from search.
- ChatGPT-User: visits a page when a ChatGPT user asks about it.
What you want to see is either no rule that blocks OAI-SearchBot, or an explicit allow:
User-agent: OAI-SearchBot
Allow: /
Watch out for a blanket block such as User-agent: * followed by Disallow: /. If there is no specific group for OAI-SearchBot, it falls back to the rules for all crawlers, so that line blocks it too. OpenAI says robots.txt changes can take about 24 hours to take effect. Our guide to robots.txt for AI crawlers explains the rules in detail.
Also ask your developer: does our CDN or firewall block bots? Some bot-protection settings block AI crawlers before robots.txt is ever read.
Minutes 3–6: check Bing
OpenAI says ChatGPT may share search queries with “third-party search providers such as Bing”. Many Indian teams look only at Google Search Console and have never opened Bing’s equivalent.
- Search Bing for
site:yourdomain.com. If few or none of your important pages appear, Bing may not have indexed them. - Set up Bing Webmaster Tools if you haven’t. You can verify your site, submit your sitemap and see which pages Bing has found.
- Consider IndexNow, a way to tell participating search engines when a page is added or changed. Bing’s own page is clear that “using IndexNow does not guarantee that web pages will be crawled or indexed”, but it helps new pages get noticed faster.
Minutes 6–10: check your text without JavaScript
Many modern websites build the page in the browser with JavaScript. People see the full page; a crawler that doesn’t run JavaScript may see an almost empty one.
A study by Vercel and MERJ, monitoring crawler traffic on Vercel’s network, found that “none of the major AI crawlers currently render JavaScript”, naming OpenAI’s crawlers among them. The study was published in December 2024; crawlers can change, so treat it as a strong reason to check, not a permanent law.
How to check:
- Open one of your key pages, such as a product or category page.
- Right-click and choose View page source (not “Inspect”).
- Search the source for a sentence you can see on the page, such as a product name or price.
If the sentence is in the source, a crawler can read it. If it isn’t, that text only appears after JavaScript runs. Ask your developer about server-side rendering or pre-rendering for those pages.
Minutes 10–13: check page basics
On the same page, in the source, look for:
- A clear
<title>that says what the page is about in the buyer’s words. - A meta description that summarises the page.
- One main heading (H1) that matches what the page actually offers.
- Structured data (JSON-LD) for products, organisation details or FAQs, if you use it. Google advises making sure your structured data matches the visible text on the page.
None of these on its own gets you named. They make it easier for any crawler to understand what each page is.
Minutes 13–15: one page per buyer question
Finally, list five questions your buyers ask, such as “best mattress for back pain” or “term plan with critical illness cover”. For each, ask: is there one page on our site that answers this plainly, in these words?
If the honest answer is “it’s spread across three pages” or “we call it something else internally”, that’s a gap. An answer can only cite a page that exists.
If you find a problem
Most fixes are small, but they sit with different people:
- robots.txt and bot protection: your developer or hosting team. A wrong line can be fixed in minutes; OpenAI says the change takes about a day to be picked up.
- Bing indexing and sitemaps: whoever owns Google Search Console today can usually set up Bing Webmaster Tools the same afternoon.
- Text that needs JavaScript: a development task. Start with the pages that sell your main products.
- Missing pages for buyer questions: your content or product marketing team.
What about llms.txt?
llms.txt is a proposed file that lists a site’s key pages for AI tools. It’s cheap to add, but OpenAI’s crawler documentation doesn’t say its crawlers or ChatGPT search use it, and Google says you “don’t need to create new machine readable files, AI text files, or markup” for its AI features. Add one if you like; don’t expect it to change what ChatGPT says.
What this checklist can’t tell you
Passing these checks means ChatGPT can read your site. It doesn’t tell you whether ChatGPT names you when buyers ask about your category, or which other websites it relies on instead. For that you have to ask the questions, several times each, and count the answers. Our guide to running an AI visibility audit shows how.
How Citationly helps
Our free checker runs the crawler, sitemap, JavaScript and page-basics checks for any address in seconds, with plain findings and no score. When you want to know whether ChatGPT actually names you, a Citationly audit records the answers to your category’s questions. Book a demo.