Methodology

What we measure, and
why it's hard.

Metricus measures AI visibility through a three-phase process: query generation, pattern recognition, and strategic output.

AI platforms give different answers every time. Ask ChatGPT the same question twice and you'll get two different brand recommendations. That's why a single query tells you nothing — and why we query each platform hundreds of times to find the real patterns underneath the noise.

Three phases

From noise
to signal

Every report follows the same three phases: measure what AI actually says about a brand, identify the patterns across hundreds of responses, and translate them into specific fixes.

Scroll to explore each phase

hub

AI Visibility Dimensions

Mention Rate

How often AI names your brand

Positioning

Where you rank vs competitors

Accuracy

What AI gets wrong about you

Citations

Which sources feed each answer

01

What We
Measure

For every brand, we measure five things across AI: how often you're mentioned, where you rank against competitors, what AI gets factually wrong, which sources it pulls from, and whether your language matches what buyers actually ask. Each dimension requires hundreds of queries — because AI gives different answers every time.

Real User Interface Simulation

Most AI visibility tools query AI models through developer APIs. The problem: API responses often differ from what real users see in the chat interface. Different model versions, different context windows, different response formatting. We simulate real user sessions — what an actual person sees when they type a prompt into ChatGPT, Perplexity, or Gemini. This means the visibility data in your report reflects your customers' actual experience, not a developer endpoint approximation.

Same question, different answers

Hundreds of queries per report

Queries across multiple AI platforms

02

Why It's
Hard

AI answers are nondeterministic — the same question produces different recommendations each time. Some answers come from real-time web search, others from the model's training data. Each platform weighs different sources differently. A brand can be visible on one AI and invisible on another. That's why we query AI hundreds of times per report — to separate signal from noise.

troubleshoot

Vocabulary Gap Analysis

link

Citation Sources

report

Error Audit

assignment

Priority Actions

03

What You
Get

Every report includes a visibility score, exact AI quotes, a competitor comparison, every factual error we found (with source tracing), and a prioritized action plan. The action plan tells you specifically what to fix, where to fix it, and why it matters — not generic SEO advice. See a real example in our sample report.

Further Reading

Go deeper

Authority

Metricus by
the Numbers

Audit Report 2026

500+

Reports Delivered

100%

Money-Back Guarantee

$99

Starting Price

See what a Metricus report delivers.