Audit methodology
Every GEO analysis follows the same documented, repeatable method so results can be compared over time.
GEO-AI is a Generative Engine Optimization (GEO) platform. It measures and improves how ChatGPT, Claude, Gemini, Perplexity and Grok understand, describe and recommend your brand. It is not a geospatial or mapping tool. Available worldwide, for every sector and language.
1. Questions
125 buyer-style prompts (25 per engine) adapted to your country, region, language and category: identity, comparisons, prices, buying, after-sales and use cases.
2. Engines
ChatGPT (OpenAI), Claude (Anthropic), Gemini (Google), Perplexity and Grok (xAI). The Free plan uses Gemini and Perplexity.
3. Five weighted pillars
- AI Product Understanding — 25%: does AI describe what you sell correctly?
- Generative Search Visibility — 25%: how often are you mentioned and recommended?
- Commerce Actionability — 20%: can AI cite real prices, plans and how to buy?
- Customer Interaction Quality — 15%: sentiment, support and trust signals.
- Operational Readiness — 15%: structured data, crawlability and entity data.
4. How to read the score
- 0-39: AI barely knows or confuses the brand.
- 40-59: known but incomplete or inaccurate.
- 60-79: consistently described and often recommended.
- 80-100: category reference in AI answers.
- A ±3 point change between close runs is normal model variation; real improvements take 4-12 weeks to show.
5. Deliverables
Score per engine and pillar, the real questions and answers, hallucination flags, quick wins and a 30/60/90 action plan with projected score per phase. Delivered in under 5 hours.