What is AI Share of Model vs Share of Voice - Key Differences

Share of Voice is a familiar marketing metric: the percentage of total category conversation or advertising exposure that your brand captures relative to competitors. Share of Model is the AI era equivalent: the percentage of AI-generated responses within a defined prompt set where your brand is mentioned or cited. Understanding the difference and why you need both helps clarify your AI search measurement strategy.

Share of Voice in the AI Era

Traditional Share of Voice measures how much of the conversation in your category your brand captures across media, social platforms, and advertising. It is calculated as your brand mentions divided by total category mentions. It is a useful benchmark but has always had the limitation of equating all mentions regardless of context or influence.

In AI search, a mention in a ChatGPT recommendation response that 2.5 billion monthly users see is categorically different from a mention in a social post with 200 followers. Share of Model captures the AI-specific dimension of this with more precision.

What Share of Model Measures

Share of Model measures the percentage of your target prompt set in which a given AI model includes your brand. For example: if you track 100 prompts on ChatGPT and your brand appears in 35 of the responses, your ChatGPT Share of Model is 35%.

This metric is specific, platform-by-platform, and directly tied to a defined set of buyer queries. That specificity is its strength: you can track exactly how it changes month over month, which prompts drove the change, and whether you are gaining or losing ground relative to competitors on the same prompt set.

Using Both in Your Strategy

Share of Voice tells you your brand's presence in the broader media and social conversation. Share of Model tells you your brand's presence in AI-generated responses to buyer queries. You need both because they move independently. A brand can have high traditional Share of Voice (lots of press and social mentions) but low Share of Model (absent from AI responses) and vice versa.

In practice, track Share of Model per platform (ChatGPT, Perplexity, Gemini, Claude) as your primary AI visibility KPI. Use traditional Share of Voice as context for why your Share of Model is changing over time: a PR crisis that reduces Share of Voice will often reduce Share of Model with a lag.

Frequently Asked Questions

Is Share of Model a standardized metric?

Not yet. Different AI visibility platforms calculate it slightly differently: some weight by prompt importance, some include partial mentions, some require the brand to be explicitly cited versus merely referenced. When comparing Share of Model data across tools, confirm the methodology. AEO Vision uses prompt-level citation detection with explicit mention counting and publishes its methodology.

How often does Share of Model change?

It can change week to week as AI models update, competitor content improves, and your own content changes. Platforms with weekly model updates (most major AI systems) can shift citation patterns noticeably in a short time. Daily tracking with weekly review is the right cadence for active optimization.

What is a good Share of Model benchmark?

It depends heavily on category competitiveness and how many competitors are in your space. In a category with 5 major competitors, a 25 to 30% Share of Model would be proportionally strong. In a category with 20 competitors, 10 to 15% might be competitive. The more useful benchmark is your trend over time and your gap relative to your top 2 to 3 competitors, not an absolute number.