Our Methodology
The Science of
AI Visibility
How AEO Pro forces LLMs to stop hallucinating and start recommending the best brands using Live Query grounding and semantic analysis.
1. Live Query Grounding
The biggest flaw in legacy SEO tools tracking AI is relying on base-model training data. If you ask ChatGPT "What is the best CRM?" without web access, it quotes a snapshot of the internet from years ago.
AEO Pro solves this. We programmatically force every language model (OpenAI, Gemini, Perplexity) to use real-time Web Search / Grounding before answering our audit queries. This ensures the Visibility Score reflects what your buyers are actually seeing today based on current SERP data, not outdated training weights.
2. Hallucination Filtering
AI models lie. To ensure your metrics are accurate, AEO Pro employs a secondary "judge" model that cross-references the primary output against your actual `Company URL` and defined `Competitors`.
If the model recommends your brand but hallucinates features you don't actually have, or invents a fake competitor, our judge model flags and filters the semantic noise before it pollutes your Visibility Score.
3. Calculating The Scores
Your dashboard provides two critical metrics derived from our multi-shot analysis matrix:
- Mention Rate (%)The raw percentage of times your brand was cited across all generated responses. A 100% Mention Rate means you were universally recommended across all models and queries.
- Visibility Score (0-100)Our proprietary weighted metric. It factors in the position of the recommendation (first vs fifth), the sentiment (highly recommended vs alternative option), and the authority of the specific LLM providing the answer.
Zero-Retention Privacy
AEO Pro enterprise accounts interact with foundation models via zero-retention API agreements. Your proprietary brand data, audit keywords, and competitive analysis are never used to train public LLM models.