What we measure, and how it's read.
We separate mention rate from citation rate. Being named in an answer is not the same as being sourced as the answer — MARK tracks both, because they are fixed in different ways.
Every score is broken down per model. ChatGPT, Gemini, Claude and DeepSeek retrieve and trust different signals, so your position is rarely identical across them.
Scores are tracked per category and per buying question, not as one blended average, so you can see the exact queries where selection is slipping.
Interceptor brands — adjacent brands, aggregators and marketplaces AI models trust more than expected — are pulled into your competitive set automatically, so the benchmark reflects who is really taking the recommendation.
Stage: Quantify.
Share of Model is the core output of the Quantify stage — the first step in the QUEEN framework, where MARK measures exactly how you're showing up in AI models before anything gets fixed.
See the full QUEEN framework →You're losing revenue to AI you can't see.
As more of your audience researches and shops inside AI models instead of search engines, fewer of them ever reach your site to be counted — even when they would have chosen you. The revenue leaks before analytics can see it.
Share of Model turns that invisible loss into a number you can report, benchmark and defend a budget against — as in this cruise-line audit, where grounded Share of Model halved for two brands.
Share of Model, answered.
↳ What is Share of Model?
Share of Model is a measurement of how often AI models recommend your brand versus competitors and interceptors in your category. It is refreshed monthly, broken down by model and category, and benchmarked against a defined competitive set.
↳ How is Share of Model different from SEO rankings?
SEO rankings measure position on a results page a person still has to click. Share of Model measures whether an AI model names and cites your brand inside the answer itself — where there is no page of ten blue links and usually only two or three brands get mentioned at all.
↳ How often is Share of Model measured?
Monthly. Each wave re-runs your category's buying questions across every major AI model so you can see movement over time rather than a single snapshot.
↳ What's an interceptor brand?
An interceptor is a brand AI recommends in your category that you wouldn't previously have considered a competitor — often a booking aggregator, marketplace, or adjacent brand that AI models trust more than expected. MARK names them by category so you know exactly who you're really losing selection to.
Get your Share of Model benchmarked — free.
Get your Share of Model benchmarked — freeMachine Preference Index measures how often your brand surfaces.
How ready your brand is to be selected by autonomous AI agents, across both AEO and ACO.
How AI shopping agents actually experience your site — and a competitor's — and where it breaks.
What AI models actually believe about your brand — not just whether they recommend it.
Continuous monitoring of the data and content feeds AI models pull from, so drift never silently breaks visibility.