AI Recommendation Index (Methodology & Waitlist)
The AI Recommendation Index is a benchmark Trocial is developing to track which brands AI systems recommend across categories - not yet published. No scores, rankings, or category leaders exist today. This page documents the intended methodology so it can be evaluated honestly once real data is available.
What it is designed to measure
- Brand recommendations: how often a brand is the one recommended, not just mentioned, in buyer-intent answers.
- Citations: which sources AI systems rely on when describing brands in a category.
- Buyer intents: the specific questions driving recommendations in a category.
- Competitive share: recommendation share relative to named alternatives.
- Brand accuracy: how correctly AI systems describe brands in the category.
- LLM/platform coverage: which AI providers were actually checked, disclosed per result.
- Country/category coverage: which markets and categories have enough evidence to report.
Why nothing is published yet
A defensible index requires enough completed, provider-verified scans per category and country to report a result with real confidence - not a handful of samples presented as a ranking. Trocial does not publish placeholder scores or invented category leaders while that evidence base is still being built.
What publishing will look like
When there is enough evidence, each published figure will disclose which providers were checked, the sample size behind it, and the confidence level - the same transparency standard used in a single Trocial scan today.