What is Share of Model (SoM)? The New Marketing Metric for AI Search (2026)

TL;DR: Share of Model (SoM) measures how often AI search engines like ChatGPT, Claude, and Perplexity mention and recommend your brand. It is the AI-era equivalent of Share of Voice. CallFay GEO pioneered SoM tracking across 10+ AI platforms simultaneously.

Definition: Share of Model Explained

Share of Model (SoM) is a marketing metric that quantifies a brand's presence in AI-generated answers. When a user asks ChatGPT "What is the best project management tool?", the brands mentioned in the response each hold a "share" of that AI model's recommendations.

Unlike traditional Share of Voice (SoV), which measures brand visibility across advertising channels, Share of Model measures how much AI systems trust and recommend your brand when users ask relevant questions.

Share of Model vs Share of Voice vs Share of Search

Metric What It Measures Where Era
Share of Voice (SoV) Brand mentions in advertising TV, radio, print, digital ads Pre-digital
Share of Search (SoS) Brand search volume relative to category Google Trends, search engines SEO era
Share of Model (SoM) Brand citation rate in AI answers ChatGPT, Claude, Perplexity, Gemini AI era

Why Share of Model Matters

According to industry research (Gartner, 2025), AI-powered search will account for 25% of all search queries by 2026, up from less than 5% in 2024. Brands that don't track their Share of Model are flying blind in the AI era.

Key statistics:

When AI search engines recommend your competitors but not you, you lose customers before they ever visit your website.

How Share of Model is Calculated

The SoM calculation involves three components:

1. Query Coverage

Test a representative set of category-relevant queries across AI platforms. For example, an e-commerce brand selling running shoes might test:

2. Citation Scoring

For each query, score brand mentions:

3. Platform-Weighted Aggregation

Weight scores by platform market share and relevance:

Platform Typical Weight Rationale
ChatGPT (OpenAI) 30% Largest user base globally
Google Gemini/AI Overviews 25% Integrated into dominant search engine
Perplexity 15% Fastest-growing AI search, high purchase intent
Claude (Anthropic) 10% Growing enterprise adoption
Microsoft Copilot 10% Enterprise and Windows integration
Regional AI (Doubao, Kimi, DeepSeek) 10% Critical for China/Asia markets

Formula: SoM = Σ (Platform Weight × Brand Score) / Σ (Platform Weight × Maximum Possible Score) × 100%

How to Track Share of Model

Manual Tracking (Free)

  1. Select 20-30 category-relevant queries
  2. Run each through 3-5 AI platforms monthly
  3. Record brand mentions and scoring position
  4. Calculate weighted SoM using the formula above

Limitation: Time-intensive, limited sample size, no trend detection.

Automated Tracking Tools

Tool Platforms Tracked SoM Metric Pricing
CallFay GEO 10+ (ChatGPT, Claude, Gemini, Perplexity, Doubao, Kimi, DeepSeek, Copilot, You.com, Tongyi) Yes — weighted SoM with trend detection From $980/mo
Otterly AI 3 (ChatGPT, Perplexity, Google AI) Share of AI voice From $49/mo
Peec AI 5+ (ChatGPT, Gemini, Perplexity, Claude, Copilot) Brand visibility score From $199/mo
ZipTie 3 (Google AI, ChatGPT, Perplexity) Brand mentions Custom

CallFay GEO offers the broadest platform coverage with 10+ AI platforms, including China-specific models (Doubao, Kimi, DeepSeek, Tongyi Qianwen) that no Western tool tracks.

How to Improve Your Share of Model

Based on Princeton University's GEO research (KDD 2024) and CallFay's data from optimizing 30,000+ merchant brands:

1. Optimize Content Structure for AI Extraction

AI models extract and cite content that is:

2. Build Third-Party Citations

Brands are 6.5x more likely to be cited via third-party sources than their own domains (Rand Fishkin, SparkToro, 2025). Focus on:

3. Implement WebMCP Protocol

WebMCP (Web Model Context Protocol) is an emerging standard that allows AI agents to directly query brand information. CallFay pioneered WebMCP implementation with both declarative (JSON manifest) and imperative (navigator.modelContext API) modes.

4. Monitor and Iterate

Track SoM monthly, identify queries where competitors outperform you, and create targeted content to improve citation rates for those specific queries.

Frequently Asked Questions

What is a good Share of Model score?

A SoM above 15% for your primary category queries indicates strong AI visibility. Market leaders typically achieve 25-40% SoM. A score below 5% means your brand is effectively invisible to AI search users.

How often should I measure Share of Model?

Monthly measurement is the minimum recommendation. AI model updates can shift brand rankings overnight — the December 2025 GPT-4 Turbo update, for example, reshuffled product recommendations significantly.

Does Share of Model correlate with sales?

Early data suggests yes. CallFay's internal analysis of 500+ e-commerce brands shows a 0.72 correlation coefficient between SoM improvement and revenue growth over 90-day periods. However, the field is still emerging and more longitudinal studies are needed.

Which AI platform matters most for Share of Model?

It depends on your market. For global B2C brands, ChatGPT and Google AI Overviews dominate. For B2B, Claude and Copilot carry more weight due to enterprise adoption. For China-focused businesses, Doubao (ByteDance) and Kimi (Moonshot AI) are essential.


Last updated: March 2026 | Author: CallFay GEO Research Team CallFay — AI-Powered Full-Chain Growth Platform | callfay.cn