How GEOX works
GEOX tracks how your brand shows up when people ask AI assistants — ChatGPT, Claude, Gemini, Perplexity — about your category. Instead of guessing whether AI recommends you, GEOX measures it: whether the AI mentions your brand, how prominently, in what tone, and which sources it cites.
The core idea
Everything in GEOX is one loop. You define questions, GEOX runs them across AI engines, stores the answers, analyzes them into metrics, and aggregates those into trends.
Key concepts
Getting started in 4 steps
The top bar
Use the brand selector to choose which brand you’re viewing (or add one inline). The date range (7 / 30 / 90 days or all time) and platform filter apply to every analytics page at once.
Your dashboard, page by page
Overview
KPI snapshot: analyzed responses, mention rate, average position, sentiment, and per-platform performance.
Brand Exposure
Mention rate and average position over time, as a chart and a table.
Platforms Analysis
How your brand performs on each AI engine.
Competitor Analysis
Your mention share versus the competitors you track.
Citations
Which source domains the AI engines cite about your category.
Questions
Every stored answer with its extracted metrics, paginated.
Good to know
My analytics are empty — why?
Adding a prompt only configures it. You have to Run a collection to generate answers. After a run, give it a moment: metrics are produced by a second analysis pass.
Why does Overview show fewer than Questions?
Questions counts raw answers; Overview counts answers that have been analyzed into metrics, which can lag slightly behind.
Citations stay empty.
Citations come from source links inside the answers. Many models rarely cite sources, so this can remain empty even when other analytics are populated.
How often does it collect data?
Manually anytime with “Run now”, or automatically when a brand has an update frequency set and the scheduler is enabled.
Ready to see how AI talks about your brand?
Sign in