چارچوب جدیدی برای استراتژی های خدمات مشتری محور: یک مطالعه موردی از یک رستوران زنجیره ای

Importance–performance analysis (IPA) is a popular customer-driven tool that enables companies to understand market competition and identify improvement priorities for various attributes of products and services. Despite the widespread use of IPA, previous studies have identified specific deficiencies. For example, the managerial improvement directions derived from IPA are potentially misleading because they ignore the asymmetric and nonlinear relationships between attribute performance (AP) and customer satisfaction (CS). Furthermore, the relationship between AP and importance is erroneously assumed to be independent. By contrast, the Kano model offers useful insight into quality attributes based on the asymmetric and nonlinear relations between AP and CS. In this study, a customer-driven framework is proposed, integrating the advantages of traditional IPA and the Kano model to elucidate the market competition position of each service and product attribute, providing strategic improvement guidelines for managers to design service activities. By conducting a case study of a restaurant chain, we demonstrate the effectiveness of the proposed approach.

The importance–performance analysis (IPA) was introduced by Martilla and James (1977) and has been a popular customer-driven tool among researchers and practitioners, elucidating the market competition of companies and facilitating the identification of improvement opportunities and strategic planning (Azzopardi and Nash, 2013, Garver, 2003 and Oh, 2001). Typically, IPA can be implemented by scoring the importance and performance of specific product or service attributes based on the voice of customers. These data were plotted on a matrix comprising four quadrants (Fig. 1). According to their positions on the matrix, the following improvement strategies can be recommended: (a) keep up the good work; (b) concentrate here; (c) low priority; and (d) possible overkill. IPA is an appealing tool because it is simple and easy to use, allowing the managerial implications of IPA to be intuitively interpreted (Arbore & Busacca, 2011). Thus, IPA has been applied in numerous industries such as tourism and hospitality (Chang et al., 2012 and Deng, 2007), health care (Yavas & Shemwell, 2001), education (O'Neill & Palmer, 2004), and banking (Matzler, Sauerwein, & Heischmidt, 2003). Despite its widespread use, the specific limitations of IPA have been criticized in extant literature. For example, various methods of calculating importance or performance may lead to different interpretations and subsequent means of correcting perceived problems (Garver, 2003 and Oh, 2001). In addition, a slight difference in the position of an attribute could cause its inferred priority to change dramatically (Bacon, 2003). Another critical problem of IPA is that ignoring nonlinear and asymmetric relations between attribute performance (AP) and customer satisfaction (CS), and erroneously assuming that the relationship between AP and importance is independent, could cause the improper commitment of scarce resources to misguided improvement efforts (Bacon, 2003, Mikulić and Prebežac, 2008 and Oh, 2001). Since its introduction in the 1980s, the Kano model has become a popular model for evaluating quality attributes, and has been applied numerous industries. The Kano model facilitates exploring the nonlinear and asymmetric relations between AP and CS, classifying quality attributes into the following categories: (a) must-be; (b) one-dimensional; (c) attractive; and (d) indifferent (Kano, Seraku, Takahashi, & Tsuji, 1984). The performance level of different quality attributes results in varying effects on the perception of CS and customer dissatisfaction (CD). When the CS is proportional to the level of performance, it is considered a one-dimensional factor. The increasing level of performance of a must-be factor does not increase the CS, but any decrease in this factor causes CD. Conversely, an increase in the level of performance of an attractive attribute enhances CS, but a low level of performance does not specifically cause CD. Regardless of the level of performance of an attribute, if it results in neither CS nor CD, an indifferent factor is attained (Chen, 2012). To avoid misinterpretations when using IPA, it is crucial to consider the Kano's quality categories (Arbore and Busacca, 2011, Mikulić and Prebežac, 2008 and Tontini and Picolo, 2010). For example, when customers rate a must-be factor as highly important, then its corresponding improvement strategy is either “keep up the good work” or “concentrate here.” However, managers should consider the possibility that further improvement might be unnecessary if an increase of this attribute would not create a significant improvement in CS. By contrast, when customers rate an attractive factor as unimportant, then its corresponding improvement strategy could be “low priority” or “possible overkill.” However, because an attractive factor can generate substantial customer delight, enlarging differentiation, a company can lose competitive opportunities by overlooking that item. Nevertheless, the Kano model possesses certain deficiencies that must be addressed. For example, it cannot identify relative importance of attributes in the same category, e.g., one-dimensional attributes (Bi, 2012). Therefore, quantitative measures must be developed to evaluate the asymmetric impacts on CS/CD. Furthermore, without emphasizing the current performance levels of product and service attributes, the Kano model is limited in identifying improvement opportunities (Tontini & Silveira, 2007). Despite the debate in the extant literature regarding IPA and the Kano model, scant studies have attempted to address these problems by integrating both models (Tontini & Silveira, 2007). The purpose of this study is to develop a quality–performance analysis (QPA) method that provides a customer-driven framework for identifying strategic service positions and providing quality improvement guidelines. The proposed QPA approach integrates the advantages of the Kano model and IPA, allowing managers to plan service activities. In addition, a signal-to-noise ratio (SNR) approach is designed to measure how AP asymmetrically affects CS and CD. This approach can be used to classify the Kano quality categories and define priorities for improvement. By using a case from the food and beverage industry, we show the effectiveness of the proposed QPA approach, comparing between the proposed QPA and the traditional IPA. Finally, specific methods are selected to compare the power of the SNR approach for classifying the Kano's quality categories.

Organizations must understand customer needs to elucidate their market competition, identify improvement opportunities, and conduct strategic planning. However, the application of traditional IPA for positioning improvement strategies has theoretical limitations that fail to provide acceptable analyses. We propose a customer-driven framework that integrates the advantages of traditional IPA and the Kano model to assist in planning service strategies. The restaurant case study indicates the effectiveness of the proposed approach. Certain aspects of the proposed method can contribute to the field of research. First, a QPA approach was proposed, identifying effective service improvement directions and providing strategic guidelines to assist managers in designing service activities. Next, AP was divided into three zones instead of using the midpoint method applied in traditional IPA; this facilitated the generation of precise improvement strategies. Third, the proposed SNR approach proved suitable for classifying the Kano's quality categories. The statistical results of the case show strong classification power that is similar to that of the functional/dysfunctional questionnaire developed by Kano, significantly outperforming other classification methods. Fourth, the SNR measures provide the relative importance of the Kano's quality categories, allowing managers to define improvement priorities accordingly. Furthermore, the proposed QPA approach is easy to implement using the data obtained from a typical CS survey. We recommend that more empirical testing be applied to additional industries to confirm the validity of the proposed approach.