fameproof
Glossary Metrics & Analytics

LLM Share of Voice (SoV)

The percentage of AI-generated responses within a category that explicitly cite or recommend a specific brand compared to competitors.

Detailed Definition

LLM Share of Voice (SoV) measures the relative market presence of a brand inside generative model outputs. Unlike traditional media Share of Voice (which tracks ad spend or press coverage), LLM SoV measures actual model response inclusion across standardized prompt sets.

Formula / Calculation StandardLLM SoV (%) = (Successful Samples Mentioning Target Brand / Total Category Prompt Samples) × 100
Why It Matters for Brands

High LLM Share of Voice ensures that when potential buyers ask AI models for recommendations, top alternatives, or market leaders, your company is consistently listed first.

How Fameproof Measures It

Fameproof tracks LLM SoV across multiple prompt domains (e.g., enterprise software, sportswear, automotive) using repeated sampling to filter out random model variance.

Read Full Evaluation Methodology →

Related Terms

Generative Engine Optimization (GEO)The methodology of optimizing brand presence, canonical entity representation, and authority to increase mentions in generative AI answers.Unaided Prompt Mention RateThe frequency with which an AI model organically mentions an entity when asked neutral, category-level questions without referencing the entity by name.Prominence CreditA scoring weight awarded based on an entity's ordinal placement (e.g., #1 vs. #8) in numbered or bulleted AI lists.

Related Research Articles

State of AI Brand Visibility 2026: Benchmark Data Across Apple, Microsoft, Google, and TeslaWe analyzed benchmark visibility data across top global technology brands. Explore why Apple and Microsoft rank highest in unaided AI responses, and how prompt packs affect score stability.Why ChatGPT, Claude, and Gemini Recommend Different Brands (And How to Audit Them)Why does ChatGPT list OpenAI and Apple first while Claude emphasizes Anthropic partners or open-source leaders? Learn the architectural and pre-training reasons behind model variance.