Skip to content
CiteScore

Guide

How to check if ChatGPT mentions your brand

A practical way to see whether ChatGPT and other assistants name or link your brand: the same prompts, a log over time, robots.txt, structured data, and pages you do not control. llms.txt is optional; Google says it ignores llms.txt.

20 October 2026 · CiteScore Team · 11 min read

A mention check is a log of what an assistant said when you asked, plus a look at whether your own pages can be fetched. It is not a ranking report, and it is not a dashboard that queries ChatGPT for you. OpenAI does not publish a citation rate, a required schema type, or a word count. CiteScore does not log into ChatGPT. The steps below are things you can do in a browser, and the limits are stated when a step is only a fetch.

ChatGPT is one product. Perplexity, Claude, Gemini, Microsoft Copilot, and Meta AI are separate products with separate crawlers. Run the same prompt in each product you care about, and keep the rows separate. The published control for ChatGPT search is on how to get cited by ChatGPT. The short map of the other controls is how to get cited by AI answer engines.

Write the prompts before you open a chat

Decide the wording first. Changing one word is a different test. Four kinds of question cover most of what people mean by “does it mention us”:

Keep the list short enough that you will repeat it. Five to ten prompts is a workable set. A list of fifty becomes a log nobody updates. Use the words a customer would type, not a slogan from the homepage.

Ask in a fresh chat, and write down what came back

Open a new chat for each prompt. A thread that already talked about your product can repeat that context. A logged-in account can also carry memory from earlier chats. Neither is wrong. They are different tests from what a new person sees, so label the row if you were signed in.

For each answer, record:

Ask the same prompt once more in another new chat if you want to see whether the name keeps appearing. Two different answers are two rows, not an average. These products do not publish a share of voice, and this page does not invent one. The ChatGPT citation check describes the inferred ChatGPT row in a CiteScore report. That row is not this manual log.

Keep the same sheet over time

A spreadsheet is enough. Columns that stay useful are date, product, prompt, mentioned, cited with your URL, other names, and a short note. Pick an interval you will actually keep, such as once a month, and do not change the prompt text between runs. If the product adds a source panel or changes the model, note that in the row rather than treating the new answer as comparable to the old one without a label.

Mentioned and cited stay separate. An answer that names you and links a review site is a mention plus a third-party URL. An answer that links your pricing page is a citation of that URL, even if the sentence is short. A competitor named in the same answer gets its own cell. Do not turn the sheet into a 0–100 score. The only 0–100 number CiteScore publishes is the audit score, and that score is not this log.

Check whether a crawler can fetch your pages

A manual prompt can still name a brand whose site blocks the search crawler, because the name may have come from somewhere else, or from an older fetch. The check that matches OpenAI’s own documentation is the robots.txt group for OAI-SearchBot. OpenAI says a site that disallows that token is not shown in ChatGPT search answers. GPTBot is the training crawl. ChatGPT-User is a fetch a person started, and OpenAI says robots.txt may not apply. The three tokens, and the lines to paste, are on how to get cited by ChatGPT. The interaction between a named group and User-agent: * is on robots.txt for AI crawlers.

Open /robots.txt on the host and read the OAI-SearchBot group yourself. The AI crawler robots.txt checker points at the free check, which reads GPTBot, PerplexityBot, and Google-Extended. It does not read OAI-SearchBot. The robots.txt generator can write the OpenAI groups. A file that says Allow can still sit behind a firewall that returns 403. OpenAI also publishes IP ranges. Use the live list on OpenAI’s crawler overview rather than a copied one.

llms.txt is a reading list, not a mention switch

llms.txt is an optional Markdown file, usually at /llms.txt, that lists pages you want a reader to open. OpenAI’s crawler overview does not list it as a requirement for ChatGPT search. Google says an llms.txt file neither helps nor harms visibility or rankings in Google Search, because Google Search ignores it. That statement is Google’s, for Google Search, and the page that carries it here is Google AI Overviews optimization. A missing file does not mean ChatGPT will not mention the brand, and a present file does not mean it will.

If you keep the file, the H1 is the site name and the links are canonical URLs of pages that answer real questions. The llms.txt generator copies that shape in the browser. The llms.txt checker is what a CiteScore fetch looks for: a non-empty /llms.txt. The best-practices note is the section order and the mistakes that break the file.

Structured data repeats facts. It does not create a mention

JSON-LD is a labeled copy of facts a person can already read on the page. Organization names the publisher. WebSite names the site. Product, Article, FAQPage, and VideoObject name the thing that page is about. A parser can use those labels. The script does not make an assistant mention you. OpenAI’s crawler overview does not list a required schema type.

Match the visible page. A name in the script and a different name in the title describes two entities. The Organization schema generator and the WebSite schema generator copy those two blocks. A watch page uses the Video schema generator. An event page uses the Event schema generator. A page that shows reviews of a product or an app uses the Review schema generator. A service page uses the Service schema generator. A course page uses the Course schema generator. A job page uses the Job posting schema generator. A recipe page uses the Recipe schema generator. A forum or support page where people can submit answers uses the QAPage schema generator. A page that names a short passage for text-to-speech uses the Speakable schema generator. An image with a visible credit, caption, or license uses the ImageObject schema generator. A book page uses the Book schema generator. A show page or an episode page uses the Podcast schema generator. A film page uses the Movie schema generator. A dataset page uses the Dataset schema generator. A vehicle page uses the Vehicle schema generator. A hotel page uses the Hotel schema generator. A restaurant page uses the Restaurant schema generator. A clinic page uses the Medical clinic schema generator. A law office page uses the Attorney schema generator. A dental office page uses the Dentist schema generator. An agent office page uses the Real estate agent schema generator. A plumbing business page uses the Plumber schema generator. An electrical contractor page uses the Electrician schema generator. The field list, and what to leave out, is the schema guide.

Look at pages you do not control

Search the web for the brand name and for the question you are testing. Note public pages that describe you and that you do not publish: a review, a docs page, a directory, a partner page, a comparison. An assistant can name the brand from one of those URLs, or name a competitor from theirs. When the product shows sources, write down which URL it used. When it does not show sources, you cannot tell which page it read. Absence of a link is not proof your site was skipped, and a link is not proof the mention will still be there next month.

What you can align is the name. The homepage, the title, and profiles you actually control should use the same string. Two public pages that describe the same product under two names give a model two entities. That is entity clarity. It is not a citation guarantee, and this guide does not tell you to buy mentions.

Free AI Answer Snapshot, then the Fix Pack

The free AI Answer Snapshot asks logged-out ChatGPT and Perplexity who they name for a local business and sends one page. AI answers vary run to run. No payment. The Fix Pack, $49 is the done-for-you work after you have seen who was named.

Run the free check on the host you care about. It reports whether the homepage responded, whether robots.txt allows or blocks GPTBot, PerplexityBot, and Google-Extended, whether /llms.txt is a non-empty text file, whether a sitemap is present, and which JSON-LD types appear on the homepage. It does not read OAI-SearchBot, and it does not ask ChatGPT whether the brand was mentioned. You still open a chat for that, and you still open the robots file for the search group. A passing row means the signal is present.

The GEO Content Template Pack ($19) is a separate $19 purchase from the check and the Fix Pack.

Questions about checking a mention

Does one ChatGPT answer tell me whether the brand is cited?

No. One chat is one observation. A new chat, a different model, a logged-in memory, and a slightly different prompt can name a different source. Write down the exact prompt, the product, the date, whether your name appeared, and whether a link to your domain appeared. Repeat the same prompt later. OpenAI does not publish a citation rate, so a single answer is not a share of voice.

Is a mention the same as a citation?

No. A mention means the brand name appeared in the answer. A citation, for this check, means the answer linked or quoted a URL on your domain. An answer can name you and link a competitor, or link you without using the product name in the sentence. Record those as separate columns. Do not collapse them into one score.

Will the free check open ChatGPT and search the brand?

No. The free check reads public files on one host: the homepage, robots.txt for GPTBot, PerplexityBot, and Google-Extended, llms.txt, a sitemap, and JSON-LD on the homepage. It does not read OAI-SearchBot, and it does not query ChatGPT.

Does blocking GPTBot stop ChatGPT from mentioning the brand?

OpenAI says the settings are independent. OAI-SearchBot is the token OpenAI ties to search results in ChatGPT. GPTBot is the training crawl. Disallowing GPTBot does not disallow OAI-SearchBot. ChatGPT-User is a fetch a person started, and OpenAI says robots.txt may not apply to it. A blocked crawler is also not the only way a name enters an answer: a page you do not control can still describe the brand.

Sources

Request a free AI Answer Snapshot

Same request as the free snapshot page. AI answers vary run to run. This is a snapshot, not a ranking, and not a promise that any assistant will name the business.

All fields are required.

We email the snapshot to this address. We do not ask for passwords, cookies, or content behind a login. We do not sell emails. Privacy.

Email this request instead
citescore@mail.grokbot.com

No mail app opened? Copy this and email it to citescore@mail.grokbot.com from any inbox.

If that send fails, email this request instead, or use the copy buttons.

Sends the business name, city, service, website, and email to citescore@mail.grokbot.com through FlowForm. This page path is included. No account and no paid checkout.

Prepared by the CiteScore Team.