RavenPack

RavenPack turns unstructured text from 40,000+ news, social, and regulatory sources into structured, time-stamped sentiment and event data. It provides 20+ sentiment indicators and 80+ descriptive fields, with relevance scoring, novelty tracking, and impact analysis for detected entities and events, in 13 languages.

Analysts use it to build event-driven factors — its NICE model combining event-based sentiment has shown predictive power months out. Applications span equities, fixed income, crypto, ESG, and risk. RavenPack is a data feed for alpha generation, not a trade executor.