Understanding Customer Retrials in Call Centers: Preferences for Service Quality and Service Speed

Problem definition: Customers are likely to initiate retrial calls when their previous contact with a call center fails to deliver a satisfactory resolution. According to industry reports, retrials are listed as a top annoying issue for customers and hurt call centers’ profits. Though recognizing this problem, call centers find it challenging to reduce retrials without overshooting their operating expenses. Our research aims to empirically understand the mechanism of customer retrials and then provide economically feasible solutions to reduce retrials. Academic/practical relevance: Little empirical research has been done to understand customers’ strategic retrials, and theoretical research studies retrials by assuming the degree to which pickup speed and service quality impact retrials. Our research empirically investigates the mechanism of customer retrials by studying whether speed and quality truly matter and, if so, how strong the impact is from each of them and whether the impacts are different across various customer segments. The quantified mechanism can then guide service providers to reduce retrials cost-effectively. Methodology: We use a random-coefficient dynamic structural model to characterize customer decisions in pursuing a satisfactory resolution and estimate the parameters from call-by-call records of a uniquely designed call center. Our model tracks customer decisions in the online waiting stage, in which customers are waiting for an agent but weighing whether to abandon, and in the off-line waiting stage, in which customers are not directly connected but are actively debating whether to retry. Utilizing the hybrid system that sequentially places customers into queues for three distinct quality service groups, we disentangle the effects of pickup speed and service quality on customers’ abandonment and retrial decisions. Results: Our estimations confirm that high service quality and quick pickup speed reduce retrials. Moreover, we discover that private customers are more sensitive to quality but less sensitive to speed compared with business customers. We suggest two service designs to reduce retrials cost-effectively by tailoring services to customer preferences. One reallocates the service groups for different customer segments without expanding the system, and the other adjusts the staffing ratios by hiring low-cost, ordinary-quality agents. Under the two tailoring designs, business customer surplus increases by up to 14.4% and private customer surplus by up to 14.9%. Managerial implications: First, our research highlights the importance of recognizing customers’ off-line decisions, which are impacted by online service offerings and, in turn, affect future online service operations. Neglecting customer retrials leads to suboptimal service designs. Second, by understanding the mechanism of customer retrials empirically, our research guides call centers to reduce retrials cost-effectively with speed–quality balance. Third, our research develops a practical analysis framework for service providers to quantify customer preferences and design tailoring services.