Share of Voice explained
Share of voice (SOV) originated in advertising as a brand’s share of all advertising in its category. SEO adapted it to estimated share of clicks across a keyword set.
In AI search, SOV is measured by sampling: run a fixed set of prompts repeatedly across platforms, count how often each tracked brand is mentioned or recommended, and divide each brand’s count by the total. It’s best reported separately for mentions, recommendations and citations, with the sample size stated.
SOV reflects your prompt set; a biased set gives a biased number. It also can’t measure actual audience exposure, which platforms don’t publish.
Share of Voice: an example
Across 360 sampled AI answers, your brand appears 108 times out of 540 total brand appearances — a 20% mention share of voice.
Why it matters for search and AI visibility
Share of voice puts visibility in competitive context; 30% presence means little until you know rivals have 60%.
Last reviewed by the Prisol team.