تضمین خدمات و مدل های مطلوب پرداخت

This paper evaluates both the optimal service level and optimal economic payout in service payout models. A service guarantee level is explicitly taken into consideration to obtain the optimal payout. In this study, we consider a generic model to provide insights into the dynamic interaction between the service guarantee and optimal payout levels. Parametric analyses show that when the demand payout coefficient is high, the impact of the payout is positive only if the payout rate is high enough.

Service guarantees have become an important means in service industries, and even non-service-oriented suppliers, to attract and retain customers. According to Kashyap (2001), a typical service guarantee policy includes two elements: a meaningful promise of a certain service quality and a compensation or payout offer. Service failure occurs when a customer is dissatisfied with the provided service. Some service guarantee policies might also include an invocation procedure of service recovery when service failures occur. Organizations strive to avoid the occurrence of service failure. Unfortunately, service failures are inevitable due to the complexity of real situations including the varying service quality of employees, the defection of products, different comprehensions of the service guarantee policy, some unfathomable customer preferences, and internal and external accidents. For example, a security breakout in the Royal Bank of Canada on May 31, 2004 collapsed the bank service for several days and spread dissatisfaction among customers who filed a lawsuit to claim $500 for each affected customer (Laudon and Laudon, 2003). In another example, a discount brokerage firm, Quick & Reilly, promised to refund the commission fees if the customers were unsatisfied with their trades due to service quality (Dunkin, 1992). The literature on service guarantees has been rich since the seminal work by Hart (1988). Wirtz and Kum (2004) suggest that an explicit guarantee has strong impacts on customers' purchase intentions. Other studies also conclude that an appropriately designed service guarantee and payout can reduce customers' perceived risk and therefore attract new customers. In other words, a service guarantee can increase the demand of the service. Baker and Collier (2005) propose a quantitative model, the Economic Payout Model for Service Guarantees (EPMSG), for determining the optimal payout level for the service industry. In the EPMSG model, they assume that the probability of the customer retention rate is a function of the economic payout. In turn, the incremental profit depends largely on the shape of the pre-defined customer retention probability function, which could be strongly biased. Furthermore, they also assume that the unit service price and the service guarantee level are fixed. However, as an integrated dynamic service guarantee system, taking the service guarantee level and the unit service price out of consideration might result in a lack of flexibility and also lose the generality of the service industry. As Tucci and Talaga (1997) point out, in order to determine the payout, it is more comprehensive to take the cost of the service and the cost of recovery into consideration. This paper has contributed to the literature in several ways. Firstly, we explicitly introduce the service guarantee level, which incurs a service cost, as a decision variable into the EPMSG model. We also detail the shortcoming of the customer retention distribution function in the discussion. Secondly, we assume that the economic payout has an impact on the demand. The next period demand will decrease as the probability of customer retention from the economic payout decreases. This generic model provides insights into dynamic interaction between the service guarantee level and the optimal payout. Thirdly, to avoid the strictness of the customer retention distribution function in the EPMSG model, we transform the model to allow a flexible unit service price in a more straightforward manner. Fourthly, we also explore a specific situation by assuming that the payout is a percentage of the unit service price. Finally, we provide parametric analyses for the models. The remainder of this paper is organized as follows. In Section 2, we present a literature review regarding service guarantee. Section 3 introduces the base economic payout model that we will extend in the later sections. The relationship between the optimal service level and payout with a fixed demand is discussed in Section 4. Section 5 presents two generic models to examine the dynamic interaction of the optimal service level, payout, and demand. We provide parametric analyses and empirical examples in Section 6 and conclude our research in Section 7.

This paper is the first to evaluate both the optimal service level and optimal economic payout in service payout models. In the first model, we provide the optimal service guarantee level and the payout of the model by assuming that the demand is fixed. We further find that in this specific model the service guarantee level is at its highest level when the economic payout is optimized. In the second model, we relax the fixed demand assumption to allow a more flexible aggregate linear demand function. This new model shows the dynamic interaction between the service guarantee level and the payout level. We find that the revenue per service unit increases as the payout increases; the revenue per service unit decreases as service guarantee level increases when the payout is big enough. We also provide optimal solutions to the generic model. The numerical examples show that when the demand coefficient of the service level becomes higher the payout tends to decrease whereas the unit price might increase. On the other hand, if the demand coefficient of the payout is high, the service provider might want to provide a high enough payout rate; in return, the customers benefit from the high service level and become more loyal. Otherwise, it is better to discontinue the payout policy because the cost of the payout could be larger than the profit obtained from the increased demand due to the payout plan. There are several possible extensions of this research. An investigation into real world payout and service guarantee data might provide first-hand data to validate the models, although this could be difficult due to the protection of commercial data or secrets. It might also be interesting to categorize the service guarantee level into different levels to cater different customer segments.