When a buyer asks ChatGPT, does it recommend you or a competitor?

Enter your domain in the free checker above. The tool works out what you sell, builds the real recommendation questions buyers ask in your category, then runs them against ChatGPT with web search on and parses the answers for whether ChatGPT recommended you, cited your URL, mentioned your name, or skipped you for a competitor. It repeats each prompt to handle the fact that ChatGPT answers change run to run, then gives you a score, the competitor it named instead, and the sources it trusted. The paid monitor keeps doing this on a schedule and alerts you when your standing changes.

A strong Google rank no longer guarantees an AI mention. ChatGPT does not simply read Google's results: when it searches the web it leans heavily on Bing, and it favors pages that are easy to quote, well structured, and backed by named sources and third-party signals. Onely reports that roughly 73% of brands get zero AI mentions despite ranking on Google page one, so being eligible in search and being named in the answer are two different things. Getting crawled is also not the same as getting quoted: an AirOps study covered by Search Engine Land found only about 15% of the pages an AI retrieves end up cited in its final answer.

Yes. When a buyer asks ChatGPT something like the best tool for a job or which vendor to choose, it routinely returns a shortlist of named brands, sometimes with a single confident pick. With web search and browsing on, those names are increasingly drawn from live pages it retrieves rather than only its training data. If your category gets asked about and your brand is not in that shortlist, the recommendation is going to a competitor every single time, and you have no way of knowing unless you track it.

When ChatGPT search or browsing is active it issues web queries, retrieves a set of candidate pages, and grounds its answer in a small subset it judges most relevant and trustworthy, often relying on Bing as the underlying index. It then cites only a fraction of what it pulled. The AirOps analysis covered by Search Engine Land put that at roughly 15% of retrieved pages making it into the citations, so clear structure, a direct quotable answer near the top, named statistics, and corroborating third-party mentions all raise your odds of being the page it keeps.

ChatGPT answers are non-deterministic, so the same question can return a different shortlist of brands minute to minute, and they shift again whenever the model or its search layer updates. That is why a single chat is a snapshot, not a measurement. The tool repeats each prompt and reports an honest range with a confidence interval instead of one fragile number, so you do not celebrate noise or panic over a normal wobble. When the sample is too small to be statistically meaningful, it declines to show a score rather than guess.

Citation behavior varies sharply between assistants, which is exactly why a ChatGPT-only check misleads you. Perplexity cites sources on almost every answer, while ChatGPT cites far more sparingly. One analysis from Superlines reported via Onely put the brand citation rate in ChatGPT at around 0.6%, well below the rates other assistants show. A brand that dominates Perplexity can be a ghost in ChatGPT, so the tool measures ChatGPT alongside Perplexity, Gemini, Google AI Overviews and Copilot to give you the full read.

Yes. The instant check needs no signup and no card: enter a domain and get a real score for ChatGPT and the four other engines, your per-engine breakdown of recommended, cited, mentioned or invisible, and the top competitor ChatGPT named. The paid plans add continuous monitoring, repeated-sampling precision, tracked prompts, history, alerts, and the full prescriptive fix list so you can keep watching ChatGPT over time.

Make your pages easy for ChatGPT to reach and quote. Allow the GPTBot and OAI-SearchBot crawlers in robots.txt, because a page ChatGPT cannot fetch is a page it cannot cite. Lead each section with a direct, liftable answer of roughly 40 to 80 words. Back claims with named statistics and sources, the single most citation-friendly thing you can do. Use question-style headings, add schema, publish an llms.txt, and earn mentions on the third-party pages ChatGPT already trusts in your category. The full playbook is in our guide on how to get cited by ChatGPT.

No. You never touch the OpenAI API, install a pixel, or write a line of code. You enter a domain and the tool handles the prompting, sampling, and parsing for you. It frames the recommendation questions, runs them across ChatGPT and the other engines, and returns a plain-language report, so a founder or marketer can read where they stand in ChatGPT in about thirty seconds.

Once is a snapshot. Because ChatGPT answers drift as the model and its search layer update, and as competitors publish new content, the signal you can trust comes from repeated sampling over time. The free check is ideal for a one-off read. The paid monitor samples continuously, keeps history, and alerts you the moment ChatGPT changes who it recommends or a competitor's share of voice in your category moves.