RavenPack Webinar From narrative to investable signal: building datasets for alpha generation Date November 4, 2026 Time 11:30 AM EST / 5:30 PM CET Duration 30 min + Q&A (45 min total) Format Online, free to attend Turning an investment idea into a dataset means solving a few hard problems: defining exposure precisely enough to measure it, identifying which companies actually qualify, avoiding false positives, and tracking that exposure consistently over time. In this webinar, our Data Scientists Peter Hafez and Francisco Gomez walk you through the full pipeline, live: from a starting hypothesis to a validated, backtestable dataset. The full pipeline, live What you'll learn See each decision between the original narrative and a dataset that is ready to test. How a narrative is translated into a precise concept definition. How that definition is expanded into a structured search plan. How the resulting dataset is validated to confirm it measures the intended exposure before it becomes a signal. Building and analyzing datasets across a large company universe and years of history, using pricing power and market share as examples. The resulting company exposure heatmaps and backtests. The economics of doing this at scale: achieving sufficient coverage and historical depth cost-effectively. Worked examples Pricing power and market share, built end to end on a real company universe. The method generalizes to other investment themes. Built with RavenPack Annotations and Bigdata.com. Combining enriched content with scalable search and retrieval. If you want to know whether an investment idea can be turned into a reliable, cost-effective, backtestable dataset, this webinar is for you. Meet the experts Speakers Peter Hafez Chief Data Scientist, RavenPack Francisco Gomez Data Scientist, RavenPack