AIMentionly

Weekly AI visibility

Are you the product AI recommends?

Buyers ask ChatGPT “what’s the best tool for this?” Checking by hand takes an afternoon, and the answer changes next week. AIMentionly asks those questions on ChatGPT and sends the brief every Monday. Perplexity and Gemini are coming soon.

  • ChatGPT
  • Perplexity Coming soon
  • Gemini Coming soon

ChatGPT is sampled via the OpenAI Responses API with web search. Perplexity and Gemini are not sampled until those API keys are set. The consumer apps are not scraped.

Run a free check

One product, up to three competitors, and about five buyer questions.

What happens after the check

01

We ask the buyer’s question

Not “does anyone mention us?” The prompt is the question a customer would type, with web search turned on.

02

Each question is sampled twice

Answers move around. A mention rate — named in 1 of 2 samples — is more honest than a single screenshot.

03

Monday’s brief names three fixes

Who got the mention, which pages were cited, and where a competitor is listed without you.

Pricing that matches a weekly brief

The free check is one sitting. Solo and Pro repeat it every Monday.

Free check

$0

  • One product and 3 competitors
  • About 5 buyer questions
  • Sampled once on ChatGPT
  • Perplexity and Gemini coming soon, on the same brief
  • No Monday brief
Run a free check

Solo

$19 / month

  • 25 tracked buyer questions
  • Weekly ChatGPT sampling and an email brief
  • Mention history for each live engine
  • Three fixes from the citations
Log in to subscribe

Pro

$49 / month

  • 100 tracked buyer questions
  • Everything in Solo
  • Perplexity and Gemini included when they launch
  • Shareable report link
  • PDF export of the latest brief
Log in to subscribe

Checkout is on Whop. Solo stays $19 a month and Pro stays $49. Manage billing

Questions founders ask first

Do you log into ChatGPT?

No. ChatGPT results come from the OpenAI Responses API with web search. Perplexity and Gemini use their official APIs too, and each one turns on when its key is set. Until then the brief marks them coming soon. The report says “sampled via API” because the consumer apps are not scraped.

Why isn’t the number stable?

These models do not return the same shortlist every time. We sample each question two or three times and report a rate, not a verdict.

Will this get expensive?

The default models are gpt-4.1-mini, Sonar, and Gemini Flash. Identical questions are cached for the ISO week, and every run logs an estimated cost.