Glossary
Share of Model
The percentage of AI answers that mention your brand relative to your competitors — the AI-era successor to share of voice.
Definition
Share of Model is the percentage of AI-generated answers, over a set of tracked prompts, in which a brand is mentioned relative to its competitors. Where share of voice measured presence in media coverage, Share of Model measures presence in the sentences AI engines write when a buyer asks for a recommendation.
The denominator matters: the metric is a share, so it only exists when the same prompts have been run for the brand and its competitors. A "Share of Model" quoted without a competitor set is just a mention rate wearing a grander name.
How it is measured
- A tracked prompt set — real buyer-style questions, run on a schedule, not one-off spot checks
- Multiple engines — ApexGEO dispatches the same prompts across 7 core AI engines, because shares differ sharply per engine
- Named-competitor recording — every answer is parsed for the brand and for who gets recommended instead
- A share, not a count — mentions divided by total competitive mentions over the same prompts and period
The honest limit
Without competitor co-mention data there is no denominator, and any percentage would be fabricated. ApexGEO renders "not comparable" in that case. This is deliberate: an unmeasured value is displayed as unmeasured, never as a placeholder number. When you see a Share of Model figure in ApexGEO, a real competitive prompt set sits behind it.
Frequently asked questions
- What is Share of Model?
- Share of Model is the percentage of AI-generated answers, across a set of tracked prompts, in which a brand is mentioned relative to its competitors. Where share of voice measured presence in paid and earned media, Share of Model measures presence in the answers AI engines actually write.
- How does Share of Model differ from share of voice?
- Share of voice counts brand appearances across advertising or media coverage. Share of Model counts brand appearances inside AI answers to buyer-style prompts. A brand can dominate traditional media and still hold near-zero Share of Model if AI engines cite competitors instead.
- How is Share of Model measured?
- By running the same buyer-style prompts against multiple AI engines on a schedule, recording which brands each answer mentions, and computing each brand's share of those mentions. The metric is only meaningful when competitor co-mention data exists for the same prompt set.
- Why would Share of Model show as 'not comparable'?
- When there is no competitor co-mention data for the tracked prompts, a percentage would be fabricated. ApexGEO renders 'not comparable' in that case rather than inventing a denominator — an unmeasured value is shown as unmeasured, never as a placeholder number.
Measure your Share of Model
ApexGEO tracks who AI engines recommend for your buyers' prompts — you, or your competitors — and shows the share honestly, including when it cannot be computed. See how the measurement works.
7-day free trial, no card required — on the self-serve plans.