Start your free check
What the scorecard contains
- Visibility per engine — appearance rate across the prompt set, 0–100, per engine, with the exact prompts where you were and were not named.
- Cited-URL inventory — every URL the engines cited across the run, tiered (owned / authoritative earned / commercial earned / community).
- Semantic coverage map — which buyer-question clusters you appear in, and the gap clusters that form your content backlog.
- Brand cognition extract — the factual claims engines made about you, flagged correct / wrong / stale, with sentiment classification.
Why human-reviewed (for now)
AI answers are probabilistic and context-sensitive; a naive automated score misleads more than it informs. During the beta, every check is run and read by an analyst using the published methodology — three samples per prompt, majority scoring. The tradeoff is deliberate: slower, but a number you can act on.
Frequently asked questions
Which engines does the Checker test?
Five in the standard run: Google AI Overviews, ChatGPT, Gemini, Perplexity and Microsoft Copilot. Claude and Grok are added on request — their citation behavior differs enough that we report them separately rather than blending them in.
Is the Checker really free?
Yes, one full four-dimension check per brand, free, human-reviewed. Ongoing monitoring (weekly/daily tracking with alerts) is the paid GEO Monitoring service.
What if my scores are zero?
Common, and the most valuable result to get early. A zero-visibility report includes the technical reasons (crawler blocks, missing indexation, absent source graph) ranked by fix effort — most zero-scores have at least one fix that takes under a day.