The Law and Policy of People Analytics

Leading technology companies such as Google and Facebook have been experimenting with people analytics, a new data-driven approach to human resources management. People analytics is just one example of the new phenomenon of “big data,” in which analyses of huge sets of quantitative information are used to guide decisions. Applying big data to the workplace could lead to more effective outcomes, as in the Moneyball example, where the Oakland Athletics baseball franchise used statistics to assemble a winning team on a shoestring budget. Data may help firms determine which candidates to hire, how to help workers improve job performance, and how to predict when an employee might quit or should be fired. Despite being a nascent field, people analytics is already sweeping corporate America. Although cutting-edge businesses and academics have touted the possibilities of people analytics, the legal and ethical implications of these new technologies and practices have largely gone unexamined. This Article provides a comprehensive overview of people analytics from a law and policy perspective. We begin by exploring the history of prediction and data collection at work, including psychological and skills testing, and then turn to new techniques like data mining. From that background, we examine the new ways that technology is shaping methods of data collection, including innovative computer games as well as ID badges that record worker locations and the duration and intensity of conversations. The Article then discusses the legal implications of people analytics, focusing on workplace privacy and employment discrimination law. Our article ends with a call for additional disclosure and transparency regarding what information is being collected, how it should be handled, and how the information is used. While people analytics holds great promise, that promise can only be fulfilled if employees participate in the process, understand the nature of the metrics, and retain their identity and autonomy in the face of the data’s many narratives