Introducing Experiments: measure how every LLM talks about your brand
Today we're introducing Experiments — the foundation of XLR8 AI. An Experiment is a curated set of natural-language buyer questions that we fan out across every major LLM, capture in full, and turn into structured measurement you can act on.
Two types of Experiment
Every Experiment is either Visibility (does your brand show up?) or Sentiment (how are you described when you do?). They share the same setup flow and result schema; only the scoring logic differs.
Five sub-types, one per buyer intent
Pick the sub-type that matches what you're trying to answer:
Visibility → Discovery — top-of-funnel category questions where the buyer doesn't yet know which vendors exist.
Visibility → Informational — educational queries; track whether your content is being cited as the source.
Visibility → Competitor Alternatives — "alternatives to [competitor]" queries, usually the highest-intent shape.
Sentiment → Competitor Comparison — direct head-to-head prompts that return structured pros/cons per brand.
Sentiment → Brand Sentiment — open-ended "what do people think of X" queries to track the LLM narrative.
Get started
Open the Experiments tab in your dashboard, click New Experiment, pick a type and sub-type, and the wizard takes you from a blank slate to a running experiment in under five minutes. Every Experiment automatically captures the full LLM response, the cited URLs, and every brand mentioned — and feeds the Insights Agent and Action Center downstream.
