Share of AI Voice is a competitive metric that measures the percentage of AI-generated responses mentioning your brand versus all brand mentions in the category. It is the AI-era version of classic share of voice - the metric PR and brand marketers have used for decades to track competitive presence. It is one of three operational signals that roll up to AI visibility, alongside brand mentions and citations.
What is Share of AI Voice?
Share of AI Voice is a brand-visibility metric measuring competitive presence inside AI-generated answers. When a user asks ChatGPT, Perplexity, Gemini, Claude, or Google AI Overviews a category question, the AI's response typically names several brands as candidates. Share of AI Voice quantifies how often your brand is one of those named, as a proportion of all brand mentions across the category's responses.
The metric is a direct adaptation of classic share of voice, which has been a staple of PR and brand-marketing measurement since the 1960s. Traditional share of voice tracked brand presence in press mentions, ad impressions, and social conversation. Share of AI Voice tracks the same concept applied to the most important new measurement surface: the interior of AI-generated answers. The math carries over. What changes is where you look.
One important disambiguation up front: Share of AI Voice has a naming collision with AI voice market share, which refers to the commercial market for voice AI technology (voice assistants, speech-to-text, companies like Speechmatics or ElevenLabs). The two concepts share a word and nothing else. When CI research tooling was run on this term, two of five AI platforms initially confused them. The convention in this glossary, and in the AI-Advisors platform, is that Share of AI Voice always means the brand-visibility metric.
The formula
The math is straightforward.
Share of AI Voice = (brand's mentions across AI responses) / (total brand mentions across all responses in the category)
Worked example. A marketer tracking a CRM category runs 20 representative buyer queries ("best CRM for small teams," "CRM tools for B2B SaaS companies," etc.) across five AI platforms. Across all 100 platform-responses, AI answers mention 12 different CRM brands a total of 240 times. Their brand is mentioned 36 times. Their Share of AI Voice is 36 / 240 = 15%.
In practice the calculation has subtleties. Should a mention in the "you might also consider" list at the bottom of a response count the same as the primary recommendation? Should mentions weighted by the platform's reach (ChatGPT's 900M weekly users vs Perplexity's smaller base) count the same? Should brand mentions in negative context count? Most mature setups weight by position, platform reach, and sentiment. The simple unweighted version is the starting point; the weighted version is the advanced version.
How to measure Share of AI Voice
A measurable program has four steps.
Define the category query set
Pick 20-50 queries a real buyer would ask to discover vendors in your category. Include discovery queries ("best CRM for..."), comparison queries ("CRM A vs CRM B"), and pain-point queries ("why is my CRM losing sales data"). The query set is the single biggest determinant of how accurate and useful the metric is. A badly-chosen query set gives numbers that do not reflect real buyer behavior.
Run across platforms
Execute the query set against at least ChatGPT, Perplexity, Gemini, Claude, and Google AI Overviews. Each platform's ranking logic is different; a Share of AI Voice number aggregated across only one platform is a weak metric. Five platforms is the practical minimum; some programs add Copilot and brand-specific tools.
Count mentions
For each response, identify every distinct brand mentioned and count occurrences. Mentions of the same brand within a single response can be counted once (uniqueness) or per occurrence (frequency); most programs count uniqueness for simplicity. A mention with a citation (linked source) is a stronger signal than a mention without one and is worth a separate column in the data.
Aggregate and track over time
Calculate Share of AI Voice per platform, per category, and per query set. Track week-over-week. Single-point-in-time numbers are almost meaningless because AI platforms change their ranking behavior continuously. Trend direction matters more than any given week's value.
Share of AI Voice vs traditional share of voice
Both measure competitive presence as a percentage. Both roll up across multiple data sources. Both are useful as trending metrics rather than point-in-time scores. They diverge in three places that matter.
- Surface - Traditional share of voice measures press mentions, ad impressions, and social content. Share of AI Voice measures the inside of AI-generated responses. These are different data sources with different dynamics.
- Signal - Traditional share of voice is shaped by PR investment, ad spend, and content volume. Share of AI Voice is shaped by structured content, schema, topical authority, and how well the brand's presence in source-of-truth data (Wikipedia, review sites, industry reports) comes through into AI training and retrieval.
- Cadence - Traditional share of voice reports are usually monthly or quarterly. Share of AI Voice is worth checking weekly because AI platforms iterate continuously and the underlying ranking can shift quickly.
For brands that historically invested heavily in PR and ad share of voice, the surprise is that those investments do not automatically flow to Share of AI Voice. A brand can be a leading share-of-voice holder in traditional media and a mid-pack holder of Share of AI Voice, because the signals AI platforms weight are different from the signals press and ad buys produce.
Why marketers track Share of AI Voice
The business case sits in three places.
Competitive benchmarking
It answers the question "are we growing or losing ground versus specific rivals in the category?" A mention-rate increase without a Share of AI Voice increase means competitors are growing faster. A Share of AI Voice increase without a mention-rate change means you are capturing share from someone specific. Both inform different strategic decisions.
Investment pacing
Share of AI Voice is a leading indicator for AI referral traffic. A category where a brand's share is growing week-over-week will see traffic growth over the subsequent 30-90 days, as the AI platforms' training and retrieval systems catch up. Marketers use the leading indicator to pace investment decisions.
Validating AEO work
Technical AEO investments, topical authority work, third-party coverage: all of these are supposed to raise the brand's competitive position. Share of AI Voice is the outcome metric that confirms whether they did.
Our free AI Visibility Checker provides a basic Share of AI Voice snapshot for a single brand query. The full Answer Engine Insights module tracks Share of AI Voice across a configured query set with weekly deltas.
Common misconceptions
Higher Share of AI Voice always means better performance
Context matters. A brand mentioned in 30% of responses in a negative tone ("X, which had the outage last quarter") has high Share of AI Voice and poor brand health. Sentiment-weighted Share of AI Voice is the more rigorous metric. Unweighted numbers can mislead if the sentiment distribution is skewed.
Share of AI Voice is just rank on ChatGPT
Rank on a single platform is not Share of AI Voice. The metric is a category-wide, cross-platform composite. Tracking only one platform's rank misses the competitive picture and can produce misleading numbers when that one platform's ranking shifts.
You can buy Share of AI Voice with ChatGPT Ads
Paid placement in conversational ads and organic Share of AI Voice are distinct measurements. Paid placements may show a "Sponsored" label and typically do not count in Share of AI Voice calculations unless the measurement setup specifically includes them. Paid placement is a separate channel with its own metrics.
Frequently asked questions
#What is Share of AI Voice in simple terms?
Share of AI Voice is a competitive metric that measures how often your brand gets mentioned in AI-generated answers, as a percentage of all brand mentions in the category. If ChatGPT's answer to "best CRM for small teams" mentions 8 vendors and one of them is you, your share of AI voice in that response is roughly 12.5%. Aggregated across many queries and platforms, it becomes a competitive benchmark that replaces traditional share-of-voice reporting for the AI era.
#How is it different from Share of Voice?
Classic share of voice tracks brand presence in traditional media: press mentions, ad impressions, social impressions. Share of AI Voice tracks presence inside AI-generated responses. The math is the same (brand mentions divided by category total), but the measurement surface is different (AI platforms instead of press and social). Brands that dominated classic share of voice do not automatically dominate Share of AI Voice - the metric is sensitive to different signals.
#Is Share of AI Voice the same as AI voice market share?
No. This is a naming collision. AI voice market share refers to the commercial market for voice AI technology (voice assistants, voice-generation models, companies like Speechmatics or ElevenLabs). Share of AI Voice is a brand-visibility metric measuring how often a brand gets mentioned by AI platforms in category responses. The two concepts share a word but operate in completely different domains. This glossary entry is about the brand-visibility metric.
#How do you measure Share of AI Voice in practice?
Define the category (a set of representative queries users might ask to discover vendors). Run those queries across at least five AI platforms: ChatGPT, Perplexity, Gemini, Claude, and Google AI Overviews. Count, per response, which brands are mentioned. Aggregate across queries and platforms. Calculate your brand's mentions as a percentage of all brand mentions. Track week-over-week. Our AI Visibility Checker automates the basic version of this for brand audits.
#Is Share of AI Voice the best metric to track?
It is one of three complementary signals, not the single metric. Mention rate measures whether you get named at all. Citation rate measures whether you get cited as a source. Share of AI Voice measures your competitive position. A complete AI visibility program tracks all three - each answers a different operational question. The umbrella term AI visibility covers the combined measurement.
