What is AI share of voice and how do you measure it?

AI share of voice translates the share-of-voice concept from media measurement to the AI engines: it measures how often a brand turns up in the engines’ default answers to category-level questions, relative to its peers. It is a useful signal of inclusion, but it is incomplete on its own and is best read alongside sentiment and source data.

How it is measured

Category prompts

Start from a defined set of category-level prompts, for example, “what are the leading firms in X” or “who are the major players in Y.”

Mention frequency, weighted by prominence

Measure how often each brand appears across the responses, weighting each mention by how prominently it sits within the response rather than counting all mentions equally.

A named peer set, same conditions

Compare the brand against a named set of peers running the same prompts on the same engines, so the share is measured on a like-for-like basis.

Why share of voice is incomplete on its own

Share of voice shows whether a brand is being included in the AI engines’ default category answers, but it says nothing about how the brand is described. Two brands can post identical share of voice and face very different reputation outcomes, one described favorably, the other named as a cautionary example. The metric counts presence, not tone.

How AIQ reports it

AIQ reports share of voice alongside sentiment, theme analysis, and source attribution, so the number is interpretable in context rather than read in isolation.