What is Mention Rate?
Also called AI mention rate
Definition
Mention rate is the percentage of eligible AI answers that name a defined brand or product. It measures answer-level presence rather than citation links, recommendation strength, or the number of times the name appears inside each answer.
How does Mention Rate work?
Calculate mention rate by dividing qualifying answers that name the brand by all eligible answers and multiplying by one hundred. Decide in advance which names and aliases qualify.
A product name may count for one analysis but remain separate from its parent company in another.
An answer containing the brand three times still counts as one mentioned answer. Include unlinked references if the definition is brand presence.
Log source citations and sentiment in separate fields so an increase in mentions does not conceal a decline in recommendation quality.
Why does Mention Rate matter for B2B SaaS?
Mention rate helps B2B SaaS marketers assess whether a vendor enters conversations about relevant buying needs. It can cover category discovery, alternatives, implementation, or comparison questions.
Each prompt group reflects a different stage of evaluation and should remain identifiable in reporting.
A strong result on branded prompts may simply show that the engine recognizes the company name supplied by the user. Non-branded prompts offer a different test of discovery.
Combining the two without labels can make a brand appear more visible than it is during early research.
What is an example of Mention Rate?
Suppose a fictional vendor appears in thirty of one hundred eligible category answers. Its mention rate for that sample is thirty percent.
Ten answers may include clickable vendor links, but that would be a separate citation observation.
If five answers criticize the vendor and twenty-five describe it neutrally or favorably, all thirty still qualify as mentions. A sentiment breakdown provides necessary context.
The headline presence metric should not silently discard unfavorable references to produce a more flattering result.
How do you improve or measure Mention Rate?
- Create an alias list and document words that cause false matches.
- Keep branded and non-branded prompts in separate reporting segments.
- Repeat observations under recorded engine and collection settings.
- Audit a sample manually to check automated entity matching.
Report the numerator and denominator alongside the percentage. Improve the accuracy and relevance of the information available about the product rather than trying to force name repetition.
OpenAI’s citation documentation explains source references, but it does not establish an official brand mention-rate methodology.
Frequently asked questions about Mention Rate
Is mention rate the same as share of voice?
No. Mention rate uses all eligible answers as its denominator. Share of voice usually compares brand visibility with the visibility of a defined competitor set.
Should negative mentions count?
Yes, if the metric measures presence. Keep sentiment as an additional dimension so negative visibility remains visible rather than being excluded.
Can multiple brands appear in one answer?
Yes. Each brand can receive an answer-level mention, so mention rates across competitors do not need to add up to one hundred percent.