How to check ChatGPT brand mentions

Ask ChatGPT a question a buyer would ask, then read the reply for your name. A mention is your brand appearing in that prose: named in a shortlist, described on its own, or compared with someone else. A citation is a page the reply leans on. Neither one is a promise that the next person who asks will see the same paragraph.

ChatGPT can omit you, describe you in the wrong category, or name you as an afterthought while a competitor gets the recommendation. Those are the same three gaps the brand-visibility guide covers. This page only shows how they look inside ChatGPT.

Use two questions and keep the wording. A branded question such as "What is [Brand]?" shows accuracy and tone when the buyer already knows you. An unbranded question such as "What are the best tools for [category]?" shows whether you appear at all. That pair is the seed of a prompt set. Trends need the rest of the set to stay still too.

Paste each question in a fresh chat so an earlier answer does not steer the next one. Note the date. The same wording can produce a different shortlist later, which is why a screenshot is a sample and monitoring is a series.

Look for four things, then stop. Were you named? Was the description fair, thin, or wrong? Who else was named that a buyer would actually compare? Which pages, if any, did the answer point at? Write those down. A blended impression ("it seemed fine") is hard to compare with next month's reply.

Definitions of share of voice and citation live in the glossary. The monitoring guide is where those signals are tracked across models, not in this how-to.

One reply on one engine cannot tell you share of voice, a trend, or what Perplexity, Gemini, and Claude say. It also cannot tell you whether a fix worked. That is the monitoring habit: the same prompt set, the models your buyers use, repeated on a cadence.

Use this how-to to learn what you are looking at. Use AI visibility monitoring to run the habit. Use AI brand visibility when you want the outcome stated on its own. Prompt Presence runs the ChatGPT read inside that wider check, together with the other three models on the audit.

A free snapshot is the first multi-model look on the plan you try. The AI Visibility Audit runs ChatGPT, Perplexity, Gemini, and Claude and delivers a written report, a 30-day action plan, and a 30-minute walkthrough. Ongoing plans are listed on the homepage. Agency is talk to us.