Mention Rate
The mention rate measures how often a brand, product or company is named in the answers of AI assistants like ChatGPT, Claude or Perplexity, relative to a fixed number of questions asked. It shows you, as a percentage, in how many relevant queries your name appears at all. This makes it a core metric for your visibility in generative AI search.
Why the mention rate matters
In classic search, it was all about ranking positions on the results page. With AI assistants, you often get just a single, summarized answer. Anyone not named in it simply does not exist for the user; there is no longer a click on the number two spot. The mention rate translates this new reality into a measurable figure: it tells you whether you appear in the answers at all. A high rate means the AI knows you as a relevant option and offers you up. A low rate is an early warning sign that competitors are occupying the space while you remain invisible. That is why it is equally meaningful for marketing and management.
How it works
To measure it, you give an AI a fixed list of realistic questions from your field, for example "Which tax advisors in Cologne specialize in startups?". Each question is queried several times, because AI answers fluctuate. You then count in how many answers your brand was named. With 100 questions and 18 mentions, your mention rate is 18 percent. A clean baseline measurement as a starting point is important so you can compare changes later. Because different AI systems answer differently, you measure per assistant separately; your rate on Perplexity may turn out very different from ChatGPT. Regular repetition makes trends and the effect of your measures visible.
Common mistakes
The most common mistake is the one-off measurement: a single test run is chance, not a reliable value, because AI answers vary. Ask each question several times and form an average. Second, mention rate and tone get confused: being named is good, but whether it is positive or negative is only revealed by sentiment analysis. Third, many phrase only brand questions ("What can company X do?") instead of neutral need-based questions. What is decisive, though, is whether the AI suggests you on its own when someone asks for a solution entirely without your name. Fourth, some compare across different AI systems even though their answer logic is too different. Measure per system and document your question catalog so that later measurements remain comparable.
Relation to AI recommendations
The mention rate is the entry ticket, but not the goal itself. Only when an AI does not just name you but actively recommends you or cites you as a source does real business value emerge. That is why you consider it together with related metrics such as the citation rate and the share of voice, meaning your share of all named providers. If your mention rate rises but your revenue does not, it is worth looking at the quality of the mention: are you named as the first choice or just listed in passing? In practice you improve the rate through the same levers as with Generative Engine Optimization: clear, citable content, clean structured data and a distinct entity that the AI can unambiguously assign to your brand.
Example
A mid-sized bicycle retailer wants to know whether AI assistants know them. Their team asks ChatGPT 50 typical customer questions such as "Where can I buy an e-bike with good service in Freiburg?" or "Which bike shop repairs cargo bikes?". Each question runs three times. In 6 of 50 answers the shop's name comes up, making a mention rate of 12 percent. After three months with a structured FAQ section, a Google Business Profile and an advice blog, the team repeats the measurement: now 31 percent. The retailer sees in black and white that its content is reaching the AI.
Common questions
What is a good mention rate?
There is no fixed target value, because it depends heavily on industry and competition. What matters is the trajectory: a baseline measurement as a starting point and regular repetition. More important than the absolute value is that your rate rises over time and is higher than that of your direct competitors.
How does the mention rate differ from share of voice?
The mention rate shows in what share of questions you are named at all. Share of voice sets your mentions in relation to all named providers and thus shows your market share in the AI answers. The two complement each other: first visibility, then relative share.