See if ChatGPT, Google AI Overviews and Claude name your brand.
We find your category and 5 buyer questions, ask ChatGPT, Google AI Overviews and Claude each one once, and show where you are named. It takes 2 to 4 minutes. Free, no sign-up.
What this AI visibility checker does
It answers one question fast: when a buyer asks an AI assistant about your category, does your brand come up? It works in three steps.
-
1) Find your category
We read your homepage and work out what you sell and who buys it. Then we write five questions a buyer in your category would ask. We pick questions that leave your brand name out, because the test is whether the engines recommend you, not whether they know you.
-
2) Ask three engines
We ask each question once on ChatGPT, Google AI Overviews and Claude. That is 15 answers in all. Most checks finish in two to four minutes.
-
3) Score the answers
We check each answer for your brand, note your position in the list, and record the competitors named instead and the sources cited. You can open every answer and read it in full.
What your AI visibility score means
Your score is the share of the answers that came back that name your brand. One sample of each question, on each engine.
AI visibility score = answers that name your brand / answers that came back
Example: the three engines return 14 answers to the 5 questions, and 5 of them name you. Your score is 36%. A question where Google shows no AI Overview, or where an engine gives no answer, is left out, because there was nothing to be named in.
The score counts a mention in the answer text. A link to your site in the sources is shown apart, because being cited and being recommended are different things. It is the same idea as the visibility rate CitePrism tracks, from one sample instead of many. To see how your mentions compare with your competitors, read about LLM share of voice.
Why one check is not enough
Ask ChatGPT the same question twice and you can get two different lists of brands. AI answers are generated fresh each time, so a brand that appears in four answers out of ten can be missing from any single one. This checker shows you one draw. It tells you where to look, but one answer cannot tell a brand named 40% of the time from a brand never named.
To measure, ask each question many times and read the rate over days. That method is called prompt sampling. To do it by hand, follow how to check if ChatGPT recommends your brand.
Free checker vs tracking
The checker is a first look. CitePrism is the daily measurement. It runs your own prompt set on a schedule across seven engines and turns the answers into trends.
Free checker
- 5 buyer questions, written for you
- ChatGPT, Google AI Overviews and Claude
- One answer per question, once
- A score, competitors and sources
- No account
CitePrism tracking
- Your own prompt set, 15 to 25 questions or more
- Seven engines: ChatGPT, Claude, Gemini, Perplexity, Google AI Overviews, Google AI Mode and Microsoft Copilot
- Runs every day, every answer stored
- Visibility rate, share of voice and rank over time
- Gaps, sources and a weekly digest email
The free plan runs on your own API keys. Pro runs on ours from $29 a month. See pricing and how AI search monitoring works across all seven engines.
FAQ
Is this AI visibility checker free?
Yes. The checker is free and needs no account. We cap the number of checks per day to keep the cost fixed, so if it has reached the limit, try again tomorrow. To track your brand every day, CitePrism has a free plan that runs on your own API keys.
What is a good AI visibility score?
There is no fixed benchmark, because it depends on your category and on how many brands compete in it. Compare your score with the competitors named instead, not with an absolute number. A brand named in most answers is a default pick for the engines. A brand named in none is not yet linked to these questions. Confirm the score with repeated sampling before you act on it.
Does it check Google AI Overviews?
Yes. For each question we look at the AI Overview Google shows and check whether it names your brand. Google does not show an AI Overview for every question. When it shows none, the report says so, and the question is left out of the score instead of counting as a miss.
Why did ChatGPT not mention my brand?
ChatGPT names the brands it sees most often next to your category, in the pages it reads and in its training data. A brand is often missing because few independent pages, such as reviews, comparisons and lists, mention it with the category terms. A new brand can also be newer than the training data. And one answer is a single draw: ask again and the list can change. The competitors named instead and the sources cited show what the engine leans on.
Do you store my data?
We store each check and its results so the report link keeps working. We keep limited technical data to prevent abuse, and we use an email address only to send a report you ask for. We do not sell any of it.
How often can I re-check a site?
Once every 7 days with the same setup. Inside those 7 days the checker shows the last report instead of running a new one, and the report page shows when a fresh check is available. Every finished check stays listed on the report page, so you can see how the result moves over time.
How long does a check take?
Most checks finish in two to four minutes. As soon as the check starts you get a shareable link, so you can leave the page and open the report later.
Which AI engines does the free checker use?
The free checker asks ChatGPT, Google AI Overviews and Claude. CitePrism tracking covers seven engines: ChatGPT, Claude, Gemini, Perplexity, Google AI Overviews, Google AI Mode and Microsoft Copilot.
Can I check a competitor?
Yes. Enter any website domain. The checker works out the category from that site and asks the same kind of buyer questions, so you can compare two reports side by side.