Understanding the metrics

AI share of voice

What share of voice measures in AI answers, and how to read it against mention and recommendation.

By the AI Native team · Updated 2026-06-11

AI share of voice is the slice of the brand conversation in AI answers that belongs to you. Where mention and recommendation ask whether you appeared, share of voice asks how much of the room you took up compared with everyone else. This guide explains what it measures and how to read it.

What it measures

When AI Native runs a scan, it reads every answer and notes which brands appeared, yours and your competitors'. Share of voice is your presence in those answers as a proportion of all brand presence across them. If five brands turn up across a set of answers and you account for a fifth of the brand presence, your share of voice is roughly twenty percent.

The important word is relative. Share of voice is a comparison, not a count. A high mention rate tells you that you show up often; a high share of voice tells you that when buyers ask, you dominate the conversation rather than sharing it evenly with four rivals. The two can move in different directions, and the gap between them is informative.

Why it is a competitive metric

Mention and recommendation are about you in isolation. Share of voice is about you against the field, which is closer to how a buyer actually experiences an answer. A buyer does not see your mention rate; they see a paragraph that names three brands, and your share of that paragraph is what shapes whether you feel like the obvious choice or an also-ran.

That makes share of voice the metric that moves when a competitor does. You can hold your own mention rate steady and still watch share of voice fall because a rival started appearing everywhere you do. Reading it alongside mention catches that, a flat mention rate with a falling share of voice means the conversation is getting more crowded, not that you got worse.

How to read it

Read share of voice as one of three numbers, never alone.

  • Low mention, low share of voice: you are mostly absent. The job is to get into the conversation at all.
  • High mention, low share of voice: you appear, but you are outnumbered. Competitors crowd the same answers. The job is to stand out, not just show up.
  • High mention, high share of voice, low recommendation: you own the room but are not being chosen. The job is to convert presence into preference.

What moves it

Share of voice rises when you are present in more of the answers buyers actually generate, and when the competitors who used to crowd those answers appear less. In practice that means being the brand the sources behind those answers point to, since AI engines draw on those sources when they decide which brands to name. The scan shows you those sources, which is where the work of moving share of voice begins.

Watching it over time

A single share-of-voice number is a snapshot. Its real value is the trend. Scanning on a recurring cadence turns it into a line you can watch, and because share of voice reacts to competitors, that line is often the earliest warning that the competitive picture in AI answers is shifting under you.

Back to Understanding the metrics or the documentation hub.