Customer Shopping Behavior and the Persistence of Revenues and Earnings

Abstract

Using GPS location data from customers’ mobile devices that encompasses nearly 2.9 billion visits to over 1 million retail locations belonging to 286 U.S. firms, we develop new measures of customer shopping behavior and find these to be associated with revenue and earnings persistence. Specifically, revenues and earnings are more persistent when a firm’s customers (a) have less variable shopping patterns, (b) live closer to retail locations, (c) are repeat rather than one-time customers, (d) shop during the week rather than on weekends, and (e) spend more time in retail locations. We use these measures to explore the implications of customer shopping behavior for investors, analysts, and management. We find that investors and analysts do not fully incorporate customer shopping behavior into trading decisions and revenue forecasts, which leads to predictable stock returns and forecast errors. However, investment decisions by managers align with customer shopping behavior. Our results illustrate conditions under which revenues and earnings are sustainable and which stakeholder decisions are consistent with insights provided by customer data.

Keywords: customers, revenues, earnings persistence, market efficiency, big data

JEL Classification: M41, M31, G14, G31