A Mixture Model to Estimate Customer Value for E-Services

Abstract

This study devises a novel customer value model to predict customers' value in an Internet environment, utilizing the concept of finance in the current status to predict future value based on historical data. The proposed model makes three contributions: (1) it constructs an equation to measure customer value for Internet-based services, (2) it considers the customer and enterprise perspectives simultaneously, and (3) it observes changes in customer value for any specific Internet user. The simulated results reveal that, in a long-term simulation, customer value decreases as the predictive time moves away from now because of deviations in perception and expectation.