How can we use text to tell us what is happening in the real world? Text-driven forecasting is the challenge of making concrete, testable predictions about future events and trends from publicly available text data. Text-based modeling methods make it possible to discover the agendas and attitudes behind the words people use. In this panel, we consider some recent success stories that use various kinds of text (expert-written analysis, blog posts, tweets) to tell us interesting things about the future and about the people behind the texts in various domains (finance, political discourse, and public opinion polls). Session co-organized by the McCombs School of Business and the Center for Research in Electronic Commerce (CREC).