Share of Model vs Share of Voice: Measuring What Matters in 2026

Share of Voice (SoV) has been the standard brand visibility metric for decades. It measures your brand\'s proportion of total visibility in a market — originally in advertising impressions, later extended to organic search visibility, social mentions, and media coverage. A brand with 15% Share of Voice in its category is visible in 15% of the touchpoints where customers encounter brands.

According to Joel House, founder of MentionLayer and author of AI for Revenue, "Share of Voice worked perfectly when all search happened on Google and all discovery happened through media. But now 37% of consumers start product research with AI, and those AI answers cite sources, make recommendations, and pre-qualify brands in ways that Share of Voice does not capture at all."

[Share of Model (SoM)](/blog/share-of-model-metric) is the AI-era equivalent. It measures the percentage of buying-intent prompts where AI models mention, recommend, or cite your brand. If you test 100 category-relevant prompts across ChatGPT, Perplexity, Gemini, and Claude, and your brand appears in 22 of those responses, your Share of Model is 22%.

The critical distinction: SoV measures how often humans see your brand in traditional channels. SoM measures how often AI recommends your brand when users ask questions in your category. Both matter. But they measure fundamentally different things, they respond to different optimization tactics, and they are converging as AI search grows.

The differences go deeper than the definition. Each metric tracks different signals, responds to different tactics, and tells you different things about your competitive position.

"The biggest difference is that Share of Voice is mostly a volume game — more spend, more content, more impressions equals more visibility. Share of Model is a consensus game — more independent sources validating your brand equals more recommendations," says Joel House. "You can\'t buy your way to a high Share of Model the way you can with Share of Voice through ad spend."

Another critical difference: SoV operates on a strict competitive basis within paid channels (one brand\'s ad impression displaces another\'s), while SoM is partially cooperative. AI models routinely recommend 3-5 brands in a single response. Your competitor being recommended does not necessarily prevent you from being recommended too — the consensus layer can include multiple brands that meet its confidence threshold.

Dropping Share of Voice in favor of Share of Model would be a mistake. The traditional search and media ecosystem still drives the majority of customer touchpoints. But treating SoV as sufficient would mean ignoring the fastest-growing, highest-converting discovery channel.

The two metrics serve different strategic purposes:

Share of Voice answers: How visible is our brand across traditional channels? Are we gaining or losing visibility relative to competitors in organic search, paid media, and social? How effective is our advertising spend?

Share of Model answers: How visible is our brand in AI-generated recommendations? When customers ask AI for advice in our category, do we appear? How do we compare to competitors in the AI recommendation layer?

The most valuable insight comes from comparing the two metrics. A brand with 30% SoV but 5% SoM has a major gap — it is highly visible in traditional channels but nearly invisible in AI recommendations. This brand risks losing market share as AI search adoption grows. A brand with 10% SoV but 35% SoM has the opposite pattern — strong AI visibility that may signal a competitive advantage in the emerging search paradigm.

The healthiest position is proportional: a brand should aim for roughly similar percentages across both metrics. The AI visibility audit measures your SoM as the AI presence pillar (30% of the composite score). Comparing it against your SoV data from tools like Semrush or Ahrefs reveals the strategic gap.

For agencies managing client portfolios, presenting both metrics in monthly reporting creates a complete picture. The MentionLayer platform tracks Share of Model across all four major AI platforms, providing the SoM side of the equation to complement traditional SoV tools.

Measuring Share of Voice: SoV measurement is well-established. Tools like Semrush, Ahrefs, Moz, and SEMrush provide organic search visibility scores that approximate SoV for SEO. For paid advertising, your ad platform\'s impression share metric gives you paid SoV. For media, tools like Meltwater, Cision, and Brandwatch track media mentions and sentiment.

The standard SoV formula: (Your brand\'s visibility / Total category visibility) x 100. A brand ranking for 500 out of 2,000 tracked keywords at an average weighted position has approximately 25% organic search SoV.

Measuring Share of Model: SoM measurement is newer but increasingly standardized. The process: 1. Build a library of 20-50 buying-intent prompts for your category 2. Test each prompt across ChatGPT, Perplexity, Gemini, and Claude 3. Score each response: mentioned (1 point), recommended (2 points), linked (3 points) 4. Calculate: (Your total points / Maximum possible points) x 100

For consistent tracking, test the same prompts weekly using the same model versions. The monitoring guide covers the complete setup, including prompt design templates and tracking spreadsheet formats. Paid tools like MentionLayer, Otterly, and Peec AI automate this testing at scale.

Tracking both together: Create a monthly dashboard that shows SoV trend alongside SoM trend. The most actionable insight is the divergence — when one metric improves while the other declines, it signals a channel-specific issue. If SoV is rising but SoM is flat, your traditional SEO efforts are not translating to AI visibility. If SoM is rising but SoV is flat, your GEO work is succeeding but your broader search presence needs attention.