AI share of voice
Definition: AI share of voice is the percentage of AI-generated answers, for a defined set of prompts, that mention or cite your brand, compared with competitors measured over the same prompts.
Also called: AI SOV, share of model, LLM share of voice
Updated
A worked example
Say you sell project management software. You pick 50 prompts that real buyers ask, such as “best project management tool for a 10-person agency”, and run each one three times in ChatGPT and twice in Perplexity. That gives 250 answers.
| Brand | Answers mentioning it | Share of voice |
|---|---|---|
| Your brand | 60 | 24% |
| Competitor A | 115 | 46% |
| Competitor B | 40 | 16% |
The shares do not need to add up to 100%, because one answer can mention several brands.
Why the number is noisy
AI answers are not deterministic. The same prompt can produce different brands depending on the model version, the user’s location, whether they are logged in, earlier messages in the conversation and small changes in wording. Vendors also define the metric differently: some count mentions, some count only citations, some weight a brand by its position in the answer.
Our view: treat AI share of voice as a trend measured the same way each month, not as a precise figure. A swing of a few points between two runs usually means nothing.
Making it useful
- Build the prompt set from real customer questions, sales calls and support tickets, not from your brand name.
- Report mentions and citations separately; they move for different reasons.
- Record which third-party sources the answers cite. Those are the places where better coverage is most likely to change the result.
The AI visibility guide explains how to set up a repeatable tracking routine.