Is there a rank tracker for ChatGPT?

Clients ask this in almost those words: "Is there a rank tracker for ChatGPT?" The short answer is yes. The longer answer starts by correcting the word "rank", because it sets the wrong expectation and then the reporting goes wrong.

Why ChatGPT has no rank

Google lists results. Position three is position three, give or take personalisation. ChatGPT writes a paragraph. Your brand might be the first name in it, the fifth, or missing, and run the same prompt twice and the order can change. Answers vary by user, by country and by the day, and ChatGPT sometimes runs web searches before answering and sometimes does not.

So what we track is a distribution, not a position:

How often the brand is named across repeated runs.
Where in the answer it appears.
Who else is named, and how often.
What sources are cited.
What sub-searches (fan-outs) ChatGPT ran.

That is what a good ChatGPT rank tracker reports.

What we look for in the tool

Real interface tracking. ChatGPT Search through its real interface produces answers that match what users see. API samples can differ.

Repetition and history. Daily runs for trends, not single snapshots.

Position in answer, share of voice and sentiment. Enough to say how we are doing against two or three named competitors.

Fan-outs. Seeing the sub-queries helps us choose what to write.

Citations. Whether our pages are cited, and which other sources are.

What happens on the site. Whether OpenAI's bots fetch our pages, and whether ChatGPT referrals convert.

How we do this on Promptwatch

We run client programs on Promptwatch. For ChatGPT work specifically:

Prompt panels run daily through the real ChatGPT Search interface, and report Visibility Score, Share of Voice, Sentiment and Position in Answer.
Query fan-outs are shown per model, and Prompt Volumes give search volume, difficulty and citation rate, which helps us prioritise prompts.
Citation analytics show which pages and domains ChatGPT cited, and Page Tracker follows specific URLs.
Agent Analytics shows IP-verified hits from GPTBot, OAI-SearchBot and ChatGPT-User, with errors and a crawl-to-citation path. These start on the Professional plan ($245 a month). The three bots do different jobs, which is why separate counts matter: one is training, one is search indexing, one is a user-triggered fetch.
Visitor analytics attribute ChatGPT referrals and conversions.

There is a free Explore tier covering ChatGPT only, with 10 prompts and 310 responses. For a first look at a client, it is often enough. Essential ($95 a month) raises that to 50 prompts and 6,000 responses.

Other trackers

AthenaHQ ($295 a month Starter, free Essential with 300 credits) and Scrunch AI ($250 a month Core) are dedicated trackers. Ahrefs Brand Radar offers Custom Prompts on paid Ahrefs plans with small daily quotas, which suits teams already in Ahrefs. Semrush offers an AI Toolkit add-on. Cheaper trackers such as GetIntel and Ceyo capture ChatGPT among other engines. They all answer "were we mentioned?" Our experience is that the next question, "why not, and what do we do?", needs citations, crawler data and traffic, and that decides our choice.

What we tell clients

We say it plainly: no tool can make ChatGPT recommend you, and none can give a true rank. What they do is show your presence, the evidence behind it and what to change. Improvements come from content that answers the questions the fan-outs reveal, credible third-party mentions, and a site that OpenAI's bots can read.

A two-week starter plan

Write 20 prompts a customer would type.
Track daily without changing the site.
Look at average mention rate, not individual days.
Read the fan-outs and citations.
Check the bot logs for errors on your key pages.
Make one content change and one technical change, then compare.

You can try the first steps on the free tier of Promptwatch before you spend anything.