Dr. Ewelina Forker will be presenting:
“The Informativeness of Dark Data for Future Firm Performance”
Abstract: This study investigates whether an organization’s “dark data” can predict future firm performance. “Dark data” represents the vast amounts of information that organizations collect and store but fail to use for decision-making purposes. I examine if the aggregate sentiment from employee emails, a ubiquitous form of dark data, predicts future sales incremental to traditional information sources. I further study if differences in the information environment explain cross-sectional variation in the predictive ability of aggregate email sentiment. To answer these questions, I collaborate with a large, U.S.-based medical technology company seeking to improve its planning process. The firm provides financial data as well as access to over 200,000 de-identified emails from more than 200 employees directly involved in planning across sales, operations, and accounting functions. Using the firm’s dark email data, I estimate aggregate employee sentiment by product-month and test its relation to future performance. My results show that, even after controlling for known predictors of future sales, product-specific sentiment is significantly associated with future sales. Further, the predictive ability of aggregate employee sentiment for future sales is greater for growth and declining products for which information uncertainty is greater, and for emails sent by rank-and-file employees for which information sharing is more costly. This study provides novel evidence of how organizations can use dark data to identify inefficiencies in forecasting and planning processes.
REGISTRATION CLOSED