Share of Model

Home > SEO Agency > SEO Glossary > Share of Model

Definition

SoM, also referred to as “Share of Model” is an early-stage metric that marketers view as the next step in the evolution of the “share of” metrics that marketers have relied on for decades. It reflects how often a brand is mentioned or recommended by LLMs, such as ChatGPT, Google Gemini or Claude when asked category-related questions.

Unlike older visibility metrics that were built around human media consumption, SoM focuses on a new audience: the AI systems that increasingly stand between a consumer and the brands they eventually choose.

Quick access

On top of that, SoM can also provide insights into how a brand is positioned within AI models by analyzing the positive and negative associations LLMs create around a brand compared with its competitors. This means that SoM isn’t just a frequency count but rather a tool to understand brand perception at a moment when that perception is being shaped by machines rather than editors, journalists, or reviewers.

However, frequent mentions do not necessarily mean positive ones. A brand might appear frequently in AI answers while still being described in negative terms, which is why raw mention counts can only tell half of the story.

Why does SoM matter?

In the age of LLMs, AI recommendations can influence users before they even visit a website or become aware of a brand, which is why being visible and recommended by AI is an increasingly important part of a brand’s online presence.

Traditionally, a consumer’s journey started with awareness, moved through consideration, and ended in a purchase decision – and brands could track and influence each of those stages through advertising, SEO and content marketing, as well as search. Now, a new intermediary has embarked on that journey: the AI assistant that a consumer consults before they’ve even formed an opinion on their own.

Additionally, SoM could provide an early sign of a brand’s future market position, although the link between AI visibility and actual market share is still not fully understood. It’s tempting to treat a strong SoM score as a leading indicator, the way search volume or social mentions have historically been treated, but because this is such a new phenomenon, marketers should be cautious about drawing firm causal conclusions.

Beyond visibility, SoM can also reveal how AI systems perceive and position a brand by showing which positive or negative associations they connect with it – information that traditional metrics were never designed to capture.

From Human Attention to AI Consideration

Whereas Share of Voice (SoV) is the visibility you rent within the traditional media ecosystem, Share of Model cannot simply be bought through advertising. Instead, a brand needs to be perceived as relevant and authoritative enough to be considered for a mention or recommendation by AI systems.

In that way, it also differs completely from Share of Search (SoS), which captures consumer behavior rather than media exposure, by assessing whether consumers actively search for a brand, even though the main goal of SoS is also brand awareness.

This changes the way brands need to think about visibility. While SoV is primarily about generating human attention and exposure, SoM reflects whether AI systems consider a brand relevant enough to include when answering questions from potential customers.

ASPECT	SHARE OF MODEL (SoM) ★	SHARE OF VOICE (SoV)	SHARE OF SEARCH (SoS)
Focus	Visibility in AI-generated answers	Visibility in traditional media and advertising	Share of search queries in the category
Goal	Be considered and recommended by AI systems	Generate human attention and brand exposure	Raise brand awareness and customer interest
Visibility	Cannot simply be bought; depends on relevance, authority and available information	Can be actively increased through advertising and media spending	Can be improved through relevant content, SEO, local visibility and digital PR
Audience	AI systems acting as an intermediary for consumers	Human consumers	Human consumers
Mechanism	AI-generated mentions, recommendations and citations	Ads, media placements and other forms of communication	Search volume
Customer journey	Can influence consideration before a consumer has a specific brand in mind	Primarily supports awareness and attention	Reflects brand awareness and customer interest
Key question	"Does AI recommend us when buyers ask for advice?"	"How much attention are we generating?"	"Are consumers searching for us?"

How can you measure SoM?

Building

When trying to measure the visibility a brand achieves within LLM-generated answers, the first step is to build a realistic set of buyer prompts that are increasingly used and relevant to the specific category.

Testing

The third step is to track the different levels of visibility generated by analysing how often a brand is being mentioned, recommended or cited as a source by the different models.

Tracking

Next, these prompts have to be tested repeatedly across multiple AI assistants, as their outputs can vary depending on which model is being used.

Comparing

Finally, the results have to be compared over time and against competitors. Rather than relying on just one snapshot, you want to make sure that they reflect reality and can be counted on.

Putting SoM into practice

Let’s make this more tangible. Imagine a small business owner turning to an AI assistant with the simple question:

“What’s the best domain registrar for a small business?”

At this point, the customer may not have a specific brand in mind and is simply looking for a solution, thus giving the AI the opportunity to shape their consideration set.

Now, imagine asking the same question across ChatGPT, Gemini and Perplexity. After some time you could also use a broader set of prompts, such as:

“What’s the cheapest domain registrar?”

“How do I transfer my domain to a new registrar?”

“NameSilo vs. GoDaddy: which one is better?”

These questions cover different stages of the customer journey, from discovering a solution to actively comparing brands. By tracking the answers over time, brands can see how consistently they appear in AI-generated recommendations and how their visibility compares with competitors.

A simple SoM calculation

The formula below is commonly used to calculate your brand’s share of all brands mentioned in AI-generated answers.

Start by counting how many times your brand is mentioned or recommended in relevant AI responses.

Next, divide this number by the total number of brand mentions across all competitors in the same category, then multiply the result by 100.

The higher the percentage, the larger your brand’s share of AI mentions compared with its competitors.

For example, if a brand appears in 7 out of 10 relevant prompts, its SoM would be 70%. However, the real value does not lie in one number, but in seeing how that number changes across prompts, AI models and competitors over time.

Best practices to improve your SoM

Make your content AI-friendly

Generative Engine Optimization (GEO) can play an important role in improving SoM visibility. Start by identifying the questions your target audience is likely to ask AI and create clear, evidence-backed content that answers them directly. FAQ pages, comparison pages, and topic-specific guides can provide AI systems with concise information that is easier to extract and reference. Adding structured data such as an FAQ page, organization, or product schema can further help AI systems understand the context of your content.
Content alone, however, is not enough. Brands should also strengthen their authority by earning mentions and backlinks from trusted industry publications, authoritative directories, or relevant “best of” and comparison articles. Publishing original research, statistics, or industry data can be particularly valuable, as it gives AI systems information they can reference that is not available elsewhere.
Consistency also matters: information about a brand should remain aligned across its own website and third-party sources. This helps AI systems build a clearer picture of the brand and reduces the risk of outdated or conflicting information.

Go beyond the surface: understand AI perception

However, improving SoM is not only about increasing the frequency of mentions. Brands should regularly analyze how AI systems describe them, which sources are used, which topics they are associated with, and how they compare with competitors. This can reveal content gaps, inaccurate information, or weak brand associations that need to be addressed.
Customer insights can help identify the questions and topics that matter most to your audience. Brands can then create useful, evidence-backed content around these gaps, for example, by answering frequently asked questions, publishing detailed comparisons, or providing original data. At the same time, monitoring reviews across relevant platforms can help strengthen the signals AI systems use to characterize a brand.

So, what's the catch?

As SoM is still in its early stages, it has not yet become a standardized method.

This means that different measurement tools can produce different and potentially non-comparable results, making it difficult to determine how reliable or meaningful these figures actually are.

More importantly, it is still unclear whether an increase in Share of Model actually translates into, or even predicts, an increase in market share. On top of that, there’s also no universally accepted benchmark for what makes a “good” Share of Model.

Instead, more meaningful indicators include a brand’s relative position compared with its competitors, the consistency of its performance across different AI models, and sustained growth over time.

Finally, more AI mentions do not automatically mean more business. The real question is whether greater visibility in AI actually helps a brand build relevance, strengthen its position and ultimately win more customers.

And the bottom line?

As AI is becoming increasingly more powerful and often frequented by users when they’re searching for things online, it’s also becoming a new layer between brand awareness and the final purchase that’s being made.

Therefore, brands have to start thinking beyond traditional search rankings and media visibility. Instead of asking “How visible are we?”, the more important question becomes “When buyers ask AI, does it recommend us?”

Most popular definitions

seo
seo amazon
seo camp
seo e-commerce
freelance seo
seo friendly
international seo
local seo
multilingual seo
off page seo
on page seo
technical seo

vocal seo
youtube seo

Notez ce page