The Service Industries Journal ISSN: 0264-2069 (Print) 1743-9507 (Online) Journal homepage: http://www.tandfonline.com/loi/fsij20 An innovative service quality evaluation and improvement model Li-Fei Chen, Szu-Chi Chen & Chao-Ton Su To cite this article: Li-Fei Chen, Szu-Chi Chen & Chao-Ton Su (2017): An innovative service quality evaluation and improvement model, The Service Industries Journal, DOI: 10.1080/02642069.2017.1389907 To link to this article: http://dx.doi.org/10.1080/02642069.2017.1389907 Published online: 23 Oct 2017. Submit your article to this journal Article views: 1 View related articles View Crossmark data Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalInformation?journalCode=fsij20 Download by: [University of Florida] Date: 25 October 2017, At: 02:35 THE SERVICE INDUSTRIES JOURNAL, 2017 https://doi.org/10.1080/02642069.2017.1389907 An innovative service quality evaluation and improvement model 創新的服務品質衡量與改善模型 Li-Fei Chena, Szu-Chi Chenb and Chao-Ton Suc a Downloaded by [University of Florida] at 02:35 25 October 2017 Department of Business Administration, Fu Jen Catholic University, New Taipei City, Taiwan, Republic of China; bMBA Program in International Management, Fu Jen Catholic University, New Taipei City, Taiwan, Republic of China; cDepartment of Industrial Engineering and Engineering Management, National Tsing Hua University, Hsinchu, Taiwan, Republic of China ABSTRACT ARTICLE HISTORY Importance-performance analysis (IPA) is a popular approach used by firms to focus resources on crucial attributes, reduce expenditure on non-critical ones and develop improvement and innovation strategies accordingly. However, IPA develops quality improvement plans based on inaccurate assumptions about the independence between importance and performance and lacks clear measurement standards, which may lead to inappropriate recommendations. IPA also does not account for desired versus adequate service. Therefore, this study proposes an innovative framework that integrates the advantages of IPA, the zone of tolerance concept, and Kano’s model. A case study conducted in a wealth management department in the banking industry demonstrates the effectiveness of the proposed methodology. The results indicate that the proposed approach recommends optimal service strategies to managers and outperforms traditional IPA. Received 30 March 2017 Accepted 1 October 2017 KEYWORDS Importance-performance analysis (IPA); zone of tolerance (ZOT); Kano’s model; service quality 关键词 重要性-績效分析; 容忍區 間; 狩野品質模式; 服務品 質 摘要 重要性-績效分析法 (importance-performance analysis, IPA) 可協助 企業將資源有效集中於顧客在意的關鍵屬性，減少非必要的支 出，並據以發展相關的改善與創新策略。此方法簡單易用，因此 受到業界廣泛的使用。然而，IPA 法的設計假設有部分並不適 當，如：此法假設重要性與績效是彼此獨立無關的，這並不符合 常理；它也缺乏清楚的指標衡量標準，因而可能導致錯誤的決 策。此外，IPA 法也未考慮顧客期待的服務水準與至少應符合的 服務水準。因此，本研究整合 IPA 法、容忍區間法 (zone of tolerance, ZOT)，以及狩野品質模式 (Kano’s model) 的優點，提 出一個創新的服務品質衡量模式，並利用某銀行的財富管理部門 的實際案例，以驗證本研究所提出方法之有效性。研究結果顯示 本方法能較傳統的 IPA 法提供更適當的服務改善方向與策略。 1. Introduction The importance of service quality has been widely noted in the literature; its strategic advantages in contributing to market share and returns on investment have been CONTACT Li-Fei Chen email@example.com Department of Business Administration, Fu Jen Catholic University, No. 510, Zhongzheng Rd., Xinzhung Dist., New Taipei City 24205, Taiwan, Republic of China © 2017 Informa UK Limited, trading as Taylor & Francis Group Downloaded by [University of Florida] at 02:35 25 October 2017 2 L.-F. CHEN ET AL. demonstrated in Anderson and Zeithaml (1984), Cui, Lewis, and Park (2003), and Phillips, Chang, and Buzzell (1983). Because of the rise in consumer awareness in recent years, customer demand has become more varied; today, fulfilling customer demands is a considerable challenge for every organization (Chen, Wen, & Yang, 2014; Gelbrich, Gäthke, & Grégoire, 2015; Hsiao, Chen, Chang, & Chiu, 2016). Although all companies exert effort to improve their products or services to increase customer satisfaction, determining how to allocate limited resources is also essential for achieving optimal levels of customer satisfaction (Chang, Liang, Chu, & Chou, 2012; Chen, 2012). Furthermore, not all attributes hold the same weight in satisfying customer needs. Identifying the critical factors that determine satisfaction is critical to the sustained success of any organization (Chang et al., 2012; Chen, 2014, 2015). To help firms provide efficient services to customers with optimally allocated resources, Martilla and James (1977) propose the importance-performance analysis (IPA), which is considered an economical, simple, and effective approach for prioritizing items (Arbore & Busacca, 2011). However, despite its simplicity, IPA has shortcomings such as the absence of a clear standard for measuring attribute importance and performance, and inherently inaccurate assumptions about the independence and linearity relationship of importance and performance. Therefore, companies may adopt inappropriate quality improvement strategies when prioritizing attributes by using IPA (Azzopardi & Nash, 2013; L. F. Chen, 2014; Garver, 2003; Lai & Hitchcock, 2015; Lo, Wang, Chien, & Hung, 2012). Service quality can be used to measure the suitability of the match between the service levels perceived by customers and customer expectations (Grönroos, 1984; Lewis & Booms, 1983). Parasuraman, Zeithaml, and Berry (1985) identify two levels of customer expectations: desired service (DS) and adequate service (AS). DS is the ideal service that a customer would like, and AS is the poorest service that a customer can accept. The zone of tolerance (ZOT) is the difference between DS and AS (Berry & Parasuraman, 1991; Parasuraman, Zeithaml, & Berry, 1991; Zeithaml, Berry, & Parasuraman, 1993). According to Parasuraman (2004), if the perceived service level falls within the zone, customers are satisfied; if the perceived service exceeds their DS level, customers are delighted; and if the service falls below the AS, customers will be dissatisfied. The ZOT concept provides additional valuable information that enables companies to measure customer feelings toward every service item (Johnston, 1995). Kano, Seraku, Takahashi, and Tsuji (1984) posit that attribute performance exerts asymmetric effects on satisfaction and dissatisfaction, and propose the two-dimensional quality model, which assumes that the invisible requirements regarding quality can be identified (Berger et al., 1993). The model reveals these requirements and classifies the quality attributes into five categories: ‘indifferent,’ ‘attractive,’ ‘one-dimensional,’ ‘must-be,’ and ‘reverse.’ Each category provides a specific strategy for improving a firm’s service quality based on its characteristics. Researchers have indicated that it is crucial to consider Kano’s quality categories when using IPA to avoid misinterpretations (Arbore & Busacca, 2011; L. F. Chen, 2014; Tontini & Picolo, 2010). The aim of this study is to present an innovative model for measuring service quality, called the Kano-ZIPA model, by integrating the advantages of IPA, the ZOT, and Kano’s model. The Kano-ZIPA model can enable organizations to identify improvement opportunities, so that specific quality improvement strategies can be planned accordingly. A case study of a wealth management department in the banking industry is explored to THE SERVICE INDUSTRIES JOURNAL 3 demonstrate the effectiveness of this proposed model. Moreover, the performance of the proposed approach is compared with that of traditional IPA. 2. Literature review IPA was developed by Martilla and James (1977), and is the conventional method for prioritizing improvements to service quality. It involves two dimensions and four quadrants (Figure 1) that measure the performance and importance of the service attributes. According to their positions on the matrix, the following improvement strategies can be recommended: (a) keep up the good work: attributes in this quadrant matter to customers and customers are currently satisfied with them; therefore, the performance level should be maintained; (b) concentrate here: the performance of these attributes does not reach the level of their value to customers; therefore, immediate improvement is required; (c) low priority: the attributes in this quadrant are not considered crucial, nor are they performing well; therefore, these attributes are not prioritized when managers make improvement plans; and (d) possible overkill: customers do not particularly value the attributes in this quadrant, but the performance of these attributes is high; therefore, the company may be allocating too many resources to these attributes. Using IPA, managers can consider the costs of various improvements and develop an action plan accordingly (Bacon, 2003). The main strengths of IPA are its low cost, ease of use, and well-focused strategy suggestions; thus, IPA is considered an effective way to prioritize items (Sampson & Showalter, 1999). IPA has been applied in numerous industries such as tourism and hospitality (Albayrak, 2015; Chu & Guo, 2015), health care (Cohen, Coleman, & Kangethe, 2016; Rau et al., 2017), education (O’neill & Palmer, 2004), banking (Matzler, Sauerwein, & Heischmidt, 2003), and manufacturing (Murali, Pugazhendhi, & Muralidharan, 2016). Despite the simplicity of this model, it has the following shortcomings: (a) the discontinuity in the four quadrants means that a slight change in an attribute’s position may Quadrant II High Importance Downloaded by [University of Florida] at 02:35 25 October 2017 2.1 Importance-performance analysis Low Quadrant I Concentrate here Keep up the good work Quadrant III Quadrant IV Low priority Possible overkill Low High Performance Figure 1. Traditional IPA matrix (Martilla & James, 1977). 4 L.-F. CHEN ET AL. trigger a contradictory strategy because of the changed priority (Bacon, 2003); (b) it ignores the relationship between a firm’s performance and the expectations of its customers; (c) it only considers a firm’s own performance without comparing it to competitors (Keyt, Yavas, & Riecken, 1994); (d) no clear standard for the range of the horizontal and vertical axes, measurement scale, and placement of the crosshairs exists; and (e) several authors have proposed that performance is related, linearly or nonlinearly, to overall satisfaction with service attributes (Chen, 2014). It is clear that managerial improvement directions derived from traditional IPA can be potentially misleading. Downloaded by [University of Florida] at 02:35 25 October 2017 2.2 Service quality and ZOT Service quality refers to the gap between customers’ perceived and expected levels of service. When the expected level is higher than the perceived level, customers are dissatisfied (Lewis & Mitchell, 1990; Parasuraman et al., 1985). Parasuraman et al. (1985) outline the five gaps model of service quality. These gaps are: (1) managers’ lack of knowledge about customer expectations; (2) managers being unable to meet customer expectations because the expectations are too high or resources are too limited; (3) suppliers being unable to reach the standards set by managers; (4) firms giving an impression that creates expectations in customers that they cannot meet; and (5) customers having service expectations because of the company’s reputation and their own past experience and personal demands, but finding the perceived service level below their original expectations. Notably, the fifth gap is influenced by the other four gaps. This model indicates that if companies aim to satisfy customers with their services, they must decrease the fifth gap. Parasuraman et al. (1985) also distinguish the fifth gap from the others. Consequently, the ZOT concept (Figure 2) was developed, pointing out that customer satisfaction can be classified under two levels, DS and AS. DS refers to a customer’s preconceived ideas of excellent service, AS is the minimum level that a customer will tolerate, and the ZOT is the difference between DS and AS (Berry & Parasuraman, 1991; Parasuraman et al., 1991; Zeithaml et al., 1993). The ZOT concept has received substantial academic attention, with several researchers advocating the ZOT model to enhance performance-based models (e.g. Nadiri, Kandampully, & Hussain, 2009; Teas & DeCarlo, 2004; Voss, Parasuraman, & Grewal, 1998; Walker & Baker, 2000). Walker and Baker (2000) identify the importance of service quality by examining expectation levels and zone widths, and Teas and DeCarlo (2004) investigate the increased predictive power of a measure considering the ZOT. More recently, Chang et al. (2012) develop a systematic algorithm to prioritize which service attributes should be improved, based on a fuzzy ZOT. K.-Y. Chen (2014) proposes the analytical framework of the competitive ZOT service-quality-based IPA (CZIPA), which evaluates service quality based on a competitive ZOT; in that study, competitors are benchmarked in the hot spring hotel industry. The CZIPA eliminates measurement bias and the problem of crosshair placement, and adds up the standard for the range of the horizontal and vertical axes and measurement scale. Although the concepts adopted into IPA could improve the issue of measuring performance, they still involve improper assumptions about the independence and linearity relationships of importance and performance, and can result in inappropriate decisions by companies. Downloaded by [University of Florida] at 02:35 25 October 2017 THE SERVICE INDUSTRIES JOURNAL 5 Figure 2. Nature and determinants of customer expectations of service (Zeithaml et al., 1993). 2.3 Kano’s model Kano et al. (1984) propose the concept of a two-dimensional quality model (Figure 3) by exploring the nonlinear and asymmetrical effects of attribute performance on customer satisfaction. They classify quality attributes into the following categories: (a) one-dimensional, (b) must-be, (c) attractive, (d) indifferent, and (e) reverse elements. The concept is outlined as follows: (a) One-dimensional quality elements (O): When these elements are sufficient, the customer feels satisfied, and the more sufficient the elements are, the more the customer enjoys them; however, when they are insufficient, the quality is unacceptable. (b) Must-be quality elements (M): If these elements are sufficient, the customer believes things must be this way or the situation would be unacceptable. 6 L.-F. CHEN ET AL. Satisfied Attractive quality One-dimensional quality Indifferent quality Not fulfilled Fulfilled Reverse quality Downloaded by [University of Florida] at 02:35 25 October 2017 Must-be quality Dissatisfied Figure 3. Kano’s two-dimensional quality model (Kano et al., 1984). (c) Attractive quality elements (A): When these elements are sufficient, the customer is satisfied, and the more sufficient the elements are, the greater the enjoyment (which can exceed the excitement for the one-dimensional elements). Notably, even when they are insufficient, the customer still finds the quality acceptable. (d) Indifferent quality elements (I): Whether or not these elements are sufficient, customer satisfaction remains the same. (e) Reverse quality elements (R): When these elements are sufficient, the customer feels dissatisfied; when they are insufficient, he/she feels satisfied. Kano et al. (1984) also designed a functional/dysfunctional questionnaire to classify these quality elements. Respondents are asked how they would feel if a particular attribute were present or fulfilled, and they select one of the following answers: (1) satisfied, (2) it should be that way, (3) I am indifferent, (4) I can live with it, or (5) dissatisfied. Subsequently, they are asked how they would feel if that attribute were absent or unfulfilled, and select one of the abovementioned responses. By combining the two answers, each attribute can be classified according to Kano’s evaluation table (Table 1). 3. Methodology This study proposes a Kano-ZIPA model that combines the advantages of IPA, the ZOT, and Kano’s model to enable managers to identify a strategic position and develop a service Table 1. Evaluation table for classifying quality attributes using the Kano questionnaire. Attribute is not fulfilled Attribute is fulfilled Satisfied It should be that way I am indifferent I can live with it Dissatisfied Satisfied It should be that way I am indifferent I can live with it Dissatisfied R R R R I I I R I I I R I I I R M M M Q Notes: M: must-be, O: one-dimensional, A: attractive, I: indifferent, R: reverse, Q: questionable. THE SERVICE INDUSTRIES JOURNAL 7 improvement plan for each service attribute. The proposed Kano-ZIPA model comprises the following six stages: (1) defining service attributes and collecting customer data, (2) identifying Kano’s quality categories for each attribute, (3) evaluating AI and AP based on the ZOT concept, (4) discriminating between high and low groups for both AI and AP, (5) establishing the ZIPA matrix, and (6) constructing a Kano-ZIPA portfolio and planning quality improvement strategies for each attribute according to the position on the Kano-ZIPA matrix. The model concept is shown in Figure 4, and its stages are detailed as follows: Downloaded by [University of Florida] at 02:35 25 October 2017 3.1. Stage 1: Define service attributes and collect customer data After domain experts are consulted and the service attributes defined, customer data are collected using a two-part questionnaire. The first part is Kano’s functional/dysfunctional questionnaire. The customers describe their feelings concerning whether each attribute is functional or dysfunctional by selecting from the provided descriptive options: ‘I like it this way,’ ‘It must be this way,’ ‘I am neutral,’ ‘I can live with it this way,’ or ‘I dislike it this way.’ For the second part, respondents rate their feelings about DS, PS, and AS levels for each attribute on a Likert scale, with values ranging from 1 (extremely dissatisfied) to 5 (extremely satisfied). 3.2. Stage 2: Define Kano’s quality categories for each attribute Using the data collected in the first questionnaire in Stage 1, each service attribute is categorized into one of five elements based on Kano’s evaluation table: (a) one-dimensional quality elements, (b) must-be quality elements, (c) attractive quality elements, (d) indifferent quality elements, or (e) reverse quality elements. Because the reverse quality element refers to a high degree of achievement resulting in dissatisfaction, and vice versa, a low degree of achievement resulting in satisfaction, it does not represent legitimate customer need in real-world practice (Chen, 2012). Therefore, this study focuses on a discussion of the first four quality elements: one-dimensional, must-be, attractive, and indifferent elements. 3.3. Stage 3: Evaluate attribute importance and performance based on the ZOT concept Expanding on the ZOT concept as proposed by Parasuraman et al. (1991), Campos and Nóbrega (2009) analyze the relationship between the importance of attributes and the ZOT of each. They determine that, when attributes have a higher importance, the customer has a narrower ZOT. Therefore, a measure of importance called ZAI is developed in the present study by considering the ZOT concept. To measure the ZAI of attributes, the following formula is proposed: n n ZAIj = DSij /n (DSij − ZOTij ) /n i=1 i=1 (1) n n = DSij /n ASij /n = DSj × ASj , j = 1, . . . , k, i=1 i=1 L.-F. CHEN ET AL. Downloaded by [University of Florida] at 02:35 25 October 2017 8 Figure 4. Kano-ZIPA research framework. where i is the ith customer, n is the total number of customers, j is the jth attribute, k is the total number of service attributes, ZAIj is the importance level of the jth attribute according to the ZOT concept, DSij is the DS rated by the ith customer regarding the jth attribute, ZOTij is the ZOT of the ith customer regarding the jth attribute, ASij is the AS rated by the ith customer regarding the jth attribute, DSj is the average of all customers’ DS regarding the jth attribute, and ASj is the average of all customers’ AS regarding the jth attribute. ZAIj is then used to measure importance, and its value ranges between 1 and 5. A greater value of ZAIj indicates that the attribute has greater importance. THE SERVICE INDUSTRIES JOURNAL 9 To determine how favorably the company has performed regarding these items, the ZAP is defined to measure performance according to the ZOT concept as follows: ZAPj = n n n n PS /n / (DS − ZOT )/n = PS /n / AS /n ij ij ij ij ij i=1 i=1 i=1 i=1 Downloaded by [University of Florida] at 02:35 25 October 2017 = PSj /ASj , (2) where ZAPj is the performance level of the jth attribute according to the ZOT concept, PSij is the PS rated by the ith customer regarding the jth attribute, and PSj is the average of all customers’ PS regarding the jth attribute. The formula is used to calculate ZAPj , which indicates how suitable an item is. The range of ZAPj is usually 0 , ZAPj ≤ 5. Therefore, a value closer to 5 indicates more favorable performance. 3.4. Stage 4: Discriminate between high and low ZAI and ZAP groups The average value of all ZAIj , denoted by AVEZAI, is used to classify high and low ZAI groups as follows: k ZAI /k. (3) AVEZAI = j j=1 An attribute is considered to have high importance if ZAIj is greater thanAVEZAI; otherwise, it is considered of low importance. Notably, when ZAPj . 1, then PSj . ASj , which indicates that the attribute is performing at or more favorably than the level that customers can accept. However, when ZAPj , 1, then PSj , ASj , which indicates that the attribute is performing below the acceptable level and the conditions should be improved. Therefore, 1 is used as a threshold to classify a high or low performance. An attribute with ZAPj . 1 indicates a tolerable performance and an attribute with ZAPj , 1 indicates an intolerable performance. 3.5. Stage 5: Establish the ZIPA matrix Similar to the traditional IPA matrix, the ZIPA matrix comprises two dimensions and four quadrants based on the measured ZAI and ZAP of items (Figure 5). The ZIPA matrix is completed by mapping attributes into the four quadrants according to the data collected in Stage 1. 3.6. Stage 6: Construct the Kano-ZIPA matrix and plan quality improvement strategies By combining the concepts of ZIPA and Kano’s model, a 4 × 4 Kano-ZIPA matrix can be constructed (Table 2), and the results obtained in Stages 2 and 5 used to assign each attribute to a space in the grid. According to the locations of the attributes in the Kano-IPA matrix, fundamental improvement guidelines can be suggested (Table 3). These are detailed in the following paragraphs. 10 L.-F. CHEN ET AL. Zone II Importance (ZAI) High Zone I High ZAI & Low ZAP High ZAI & High ZAP AVEZAI Zone III Low Zone IV Low ZAI & High ZAP Low ZAI & Low ZAP Intolerable 1 Tolerable Downloaded by [University of Florida] at 02:35 25 October 2017 Performance (ZAP) Figure 5. Proposed ZIPA matrix. Table 2. Kano-ZIPA portfolio matrix. ZIPA ZAI ZAP High Kano’s categories Zone Must-be factor One-dimensional factor Attractive factor Indifferent factor Tolerable (I) (1) Minor strength (5) Major strength (9) Major strength High Intolerable (II) (2) Major weakness (6) Major weakness (10) Potential strength Low Low Intolerable Tolerable (III) (IV) (3) Minor weakness (4) Surplus (7) Minor weakness (8) Minor strength (11) Minor weakness (12) Minor strength (13) Possible opportunity (14) Possible opportunity (15) Non-critical (16) Redundancy (a) Must-be quality factors (P1) Minor Strength: A must-be factor with high importance is also expected to be high in performance. However, the level of fulfillment should not be too high, because satisfaction cannot be improved with and already better-than-adequate performance. Thus, an attribute assigned to cell P1 is considered a minor strength for a company. Managers should focus on improving efficiency based on cost control concerns, and develop an appropriate plan (e.g. for simplifying the process of delivering the service). (P2) Major Weakness: A high-importance must-be factor indicates something that must be adequately fulfilled. Intolerably low performance is considered a major weakness because it can result in unavoidable dissatisfaction. An action plan to fulfill these items adequately should be implemented as a top priority. (P3) Minor Weakness: A low-importance must-be factor with intolerably low performance can induce some customer dissatisfaction; thus, it is considered a minor weakness. A conservative improvement plan can be adopted for efficient fulfillment. (P4) Surplus: A low-importance must-be factor with a high performance generates a surplus for firms. The amount invested in such items should be reduced until a satisfactory, but not excessive, level is reached. THE SERVICE INDUSTRIES JOURNAL 11 Table 3. Segmented quality improvement plan of the proposed Kano-ZIPA model. Downloaded by [University of Florida] at 02:35 25 October 2017 ZIPA Kano’s categories Zone Must-be factor One-dimensional factor ZAI ZAP Attractive factor Indifferent factor High Tolerable (I) (P1) Minor Strength A proper plan is developed to improve efficiency. (P5) Major Strength A reasonably aggressive strategy is provided to exploit customer satisfaction. (P9) Major Strength To distinguish a company from competitors, a leverage strategy should be planned. High Intolerable (II) (P6) Major Weakness To reduce dissatisfaction, an aggressive improvement plan should be implemented immediately. (P10) Potential Strength An active improvement strategy should be adopted when resources are available. Low Intolerable (III) (P2) Major Weakness To fulfill adequately, a corrective and preventive action plan should be implemented immediately. (P3) Minor Weakness Budget cuts can be implemented here. (P13) Possible Opportunity To attain potential customers’ satisfaction and transform the items into attractive factors, these factors can be investigated. (P14) Possible Opportunity To prevent dissatisfaction, an observation plan can be implemented. (P7) Minor Weakness Budget cuts can be implemented here. Low Tolerable (IV) (P4) Surplus The amount invested should be reduced until it reaches an adequate level. (P8) Minor Strength Item quality to date should be ensured, but attention should be reduced. (P11) Minor Weakness To attain potential customers’ satisfaction, managers should further investigate this factor. (P12) Minor Strength To distinguish a company from competitors, strategy to enhance its importance to customers can be implemented. (P15) Non-critical Managers should not focus on this item. (P16) Redundancy The amount invested on items should be reduced to eliminate waste. (b) One-dimensional quality factors (P5) Major Strength: A high-importance one-dimensional factor is expected to have high performance because it can generate high satisfaction. Hence, a high-importance one-dimensional factor with high performance is considered a major strength. To enhance customer satisfaction, a reasonably aggressive strategy to continue this performance is recommended. (P6) Major Weakness: A high-importance one-dimensional factor with intolerably low performance can lead to serious dissatisfaction. Implementing an improvement plan should be a top priority, as indicated in P2. (P7) Minor Weakness: A low-importance one-dimensional factor with intolerable performance is considered a minor weakness because it can result in some dissatisfaction, although not enough to be a major concern. A conservative improvement plan can be adopted to increase the level of satisfaction. 12 L.-F. CHEN ET AL. (P8) Minor Strength: A low-importance one-dimensional factor with high performance is a minor strength for firms. Managers can pay less attention to these factors, although they should ensure that they maintain a decent performance level. Downloaded by [University of Florida] at 02:35 25 October 2017 (c) Attractive quality factors (P9) Major Strength: A high-importance attractive factor with a high performance can generate more satisfaction than a one-dimensional factor can. Therefore, to distinguish a company from competitors, a leverage strategy should be planned. (P10) Potential Strength: A high-importance attractive factor can potentially become a strength, generating unexpectedly high satisfaction if fulfilled. Therefore, managers should pay more attention to improving the fulfillment of high-importance attractive factors with low performance. An active improvement strategy can be adopted when resources are available. (P11) Potential Weakness: A low-importance attractive factor with low performance induces little dissatisfaction; however, if ignored, companies may lose an opportunity to generate high satisfaction. Therefore, managers should further investigate this factor. (P12) Minor Strength: A low-importance attractive factor with high performance is a minor strength because of the low contribution to satisfaction. To distinguish a company from competitors, however, a strategy to enhance its importance to customers can be implemented. (d) Indifferent quality factors (P13) Possible Opportunity: A high-importance indifferent factor with high performance can potentially become attractive. Managers should investigate the item and exploit it if it appears to be transforming into an attractive factor. However, in the absence of signs that this item could become attractive in the long run, managers should consider reducing resources spent on it. (P14) Possible Opportunity: A high-importance indifferent factor can potentially become attractive in the future. Managers should monitor this item and improve its performance in the presence of signs that it is becoming attractive. (P15) Noncritical: A low-importance indifferent factor with low performance is noncritical because it can produce neither satisfaction nor dissatisfaction. Managers should not focus on this item. (P16) Redundancy: A low-importance indifferent factor with high performance can produce only negligible satisfaction. Investments in this factor are wasteful and should be reduced. 4. Case study 4.1. The case The case study examines a company active in the banking industry, with more than 100 branches in Taiwan as well as several overseas subsidiaries, foreign branches, and a foreign representative office. According to SINGFIN’s report of ‘wealth management in THE SERVICE INDUSTRIES JOURNAL 13 Downloaded by [University of Florida] at 02:35 25 October 2017 Taiwan’ (SINGFIN, 2017), Taiwan’s wealth management market is one of the most competitive in the world. Taiwan has approximately 130,000 high net worth individuals (those with assets of at least $1million USD) as of 2016, and the number is expected to increase stably over the upcoming years. In addition, banks in Taiwan have crowded the market, and wealth management services are not limited to the financial domain. Therefore, managers in the case study company should develop service strategies that enhance their competitiveness in the market. 4.1.1. Defining service attributes and collecting customer data The questionnaire to understand the customers is based on SERVQUAL (a framework adapted from Parasuraman et al., 1991) and its five dimensions: (a) tangibles, (b) reliability, (c) responsiveness, (d) assurance, and (e) empathy. To fit the realistic operations of the case company, a discussion with two managers is also conducted to modify the questions. The questionnaire consists of two parts. The first of which is used to evaluate each customer’s DS, AS, and PS level rating according to a Likert scale that ranged from 1 (extremely dissatisfied) to 5 (extremely satisfied). The second part is Kano’s functional/dysfunctional questionnaire. The respondents are asked to choose one of the following answers: (a) satisfied, (b) it should be that way, (c) I am indifferent, (d) I can live with it, or (e) dissatisfied under the situations if they feel a particular attribute were fulfilled and if they feel a particular attribute were not fulfilled. A pretest was administered at two branches of the company in Taipei, and the respondents were 13 sales coordinators who were the first Table 4. Service attributes. Items X1 X2 X3 X4 X5 X6 X7 X8 X9 X10 X11 X12 X13 X14 X15 X16 X17 X18 X19 X20 X21 X22 X23 X24 X25 Description The branches of the bank provide complete and timely financial information (e.g. currency rate and interest rate). The branches of the bank provide services in a safe and private environment (e.g. an exclusive office). When services are provided, information is managed promptly and smoothly with equipment. The appearance of the financial consultants is clean and conveys professionalism. The bank provides a market research report on areas and countries. The bank provides a banking application on mobile devices. The financial consultants can explain the information of services, responsibility, and financial situation. The financial consultants understand customers’ financial needs. The bank provides tax and real estate planning services, and professionals are available for consultation with individual customers. The bank provides insurance consulting services, including explanations for policy terms, applications for insurance claims consultations, collection services, progress concerning claims notices, and so on. The financial consultants are active in providing services and care to customers. The financial consultants understand customers’ needs and protect their privacy and rights. The bank provides VIP priority when VIP members are at the service counter. The bank informs customers of sudden events, new information, and new product information. The bank provides additional high-level services, such as the reservation of local or overseas hotels and restaurants. The bank provides pick-up services from international airports several times per year as well as high-speed train tickets and airport VIP room reservations. The financial consultants keep their promises. The brand and reputation of the bank are reliable. The financial consultants offer recommendations based on individual needs and do not focus on selling goods because of commissions. The financial consultants explain the risks of goods in details. The financial consultants check financial conditions periodically. The financial consultants provide information on goods as well as investment performance (through visits, phone calls, and messages). The financial consultants inform customers of the fee required for every service. The financial consultants respond to customer complaints or suggestions effectively and in a short period. The overall service quality which the customers perceive from the bank. Downloaded by [University of Florida] at 02:35 25 October 2017 14 L.-F. CHEN ET AL. line to deal with customers to ensure the wording was clear. A total of 24 service attributes and overall service qualities were investigated in this study. Please see Table 4 for these attributes. In February and March 2014, convenience sampling was used to recruit customers in the two branch banks and that agreed to take part in the research. By checking carefully with the respondents face-to-face, a total of 243 effected questionnaires were collected. Cronbach’s α coefficient is .924 for the customer PS level, .972 for the DS level, .956 for the AS level, and .752 for Kano’s questionnaire. All of these values are above the benchmark value of .70 recommended by Nunnally (1978) and Churchill (1979), confirming that the survey results show solid internal consistency and reliability. Table 5 presents a summary of the demographics and consumption patterns of the respondents. Moreover, it shows that most of the respondents are women (53.5%), married (80.9%), and have college degrees (53.1%). Nearly half are over 51 years of age (43.8%). Additionally, 21.7% of the respondents have worked in the service industry, and 19.5% are retired. Most of the respondents had been using the company’s services for 2–10 years (67.6%); the majority had also been receiving VIP service for 2–5 years (54.0%) or 6–10 years (33.6%). The type of VIP status is dependent on the amount respondents have invested in the company. The highest level is VIPC, followed by VIPB, and then by VIPA; nearly half of the respondents here are VIPB members (48.7%), which reveals that the largest proportion of customers were provided with mid-level service. New members of the wealth management service (less than 1 year) account for 10.6% of the respondents. Furthermore, the primary way that they had learned about this service was through the bank itself (55.5%). Approximately half of the respondents (47.8%) visit this bank twice every month. 4.1.2. Analysis of Kano-ZIPA model After collecting the DS, PS, and AS data, the ZAI and ZAP for each attribute can be derived. Table 6 shows the statistical results of the ZIPA model. Notably, the average ZAI value, 3.89, is used to classify the attribute importance into high and low groups, while 1.00 is used to classify the attribute performance into tolerable and intolerable groups. Each attribute is then assigned to one of the four zones in the ZIPA matrix (see Figure 6). In total, 58.33%, 8.33%, 12.50%, and 20.83% of the attributes are located in zones I, II, III, and IV, respectively. First, 14 attributes including X1, X2, X3, X4, X7, X11, X12, X13, X17, X18, X20, X22, X23, and X24 with the high level of importance and tolerable performance are located in zone I. The managers should ‘keep up the good work’ on these attributes. Next, managers should try their best to ‘concentrate’ on attributes X19 and X21 which located in zone II because these attributes are very important to customers but their current performance is intolerable. Third, although the performance of attributes X6, X15, and X16 located in zone III is intolerable, managers can set ‘low priority’ to improve them due to their low level of importance to customers. Finally, attributes X5, X8, X9, X10, and X14 located in zone IV are ‘possible overkill.’ Although their performance is tolerable, allocate resource on these attributes may not bring value to customers due to their low-importance level. From the data collected through Kano’s functional/dysfunctional questionnaire, Kano’s categories are identified for each attribute. Combining these results and the position in the THE SERVICE INDUSTRIES JOURNAL 15 Table 5. Demographics and consumption pattern of the respondents. Items Demographics 1. Gender 2. Marital status 3. Age Downloaded by [University of Florida] at 02:35 25 October 2017 4. Educational status 5. Occupation Consumption pattern 1. How many years have you been provided service in this bank? 2. How many years have been provided with the wealth management service (VIP service) in this bank? 3. Which channels were you informed about the wealth management service (VIP service) in this bank in the beginning? (Multiple choices) 4. How often do you come to this bank every month for the wealth management service (VIP service)? 5. Which type of VIP service do you receipt? Categories Male Female Married Single Below 20 years old 21–30 years old 31–40 years old 41–50 years old Above 51 years old Junior high school and below Senior high school/senior high vocational school College Graduate school and above Prefer not to answer Student Housekeeper Military officer/ government official/educational personnel Financial industry Manufacturing industry Freelance Service industry Retirement Others Less than 1 year 2–5 years 6–10 years 11–15 years 16–20 years 21–25 years 26–30 years Above 31 years Less than 1 year 2–5 years 6–10 years 11–15 years From relatives/ friends From the company Official website TV Newspaper/ magazine Others Less than Once Twice 3 times 4 times 5 times VIPA VIPB VIPC Frequency Percentage 105 121 183 41 1 11 51 64 99 7 38 46.5 53.5 80.9 18.1 0.4 4.9 22.6 28.3 43.8 3.1 16.8 120 31 30 3 33 23 53.1 13.7 13.3 1.3 14.3 10.2 29 34 10 49 44 1 12.8 15.0 4.4 21.7 19.5 0.4 4 76 77 38 23 5 2 1 24 122 76 5 71 146 6 23 17 0 59 108 37 11 11 91 110 25 1.8 33.6 34.1 16.8 10.2 2.2 0.9 0.4 10.6 54.0 33.6 2.2 27.0 55.5 2.3 8.7 6.5 0.0 26.1 47.8 16.4 4.9 4.9 40.3 48.7 11.1 ZIPA matrix, the Kano-ZIPA model is then analyzed, as noted in Tables 7 and 8. Practical service strategies are recommended as follows: First, two crucial must-be factors with unacceptable performance (X19 and X21) are considered to be major weaknesses causing widespread customer dissatisfaction. Managers should consider these attributes top priority and adopt a corrective action plan to 16 L.-F. CHEN ET AL. Downloaded by [University of Florida] at 02:35 25 October 2017 Table 6. Statistical results of the proposed ZIPA model. No. DS PS AS ZAI ZAI level ZAP ZAP level ZIPA zone Improvement guidelines X1 X2 X3 X4 X5 X6 X7 X8 X9 X10 X11 X12 X13 X14 X15 X16 X17 X18 X19 X20 X21 X22 X23 X24 Ave 4.32 4.43 4.26 4.48 3.78 3.21 4.33 3.63 3.94 4.13 4.64 4.42 4.33 3.99 3.73 3.94 4.64 4.53 4.51 4.39 4.24 4.25 4.35 4.56 4.21 3.96 4.2 3.85 4.33 3.53 2.8 3.82 3.29 3.42 3.61 4.1 3.87 4.13 3.44 3.09 3.22 4.07 4.22 3.72 3.81 3.6 3.69 3.86 3.89 3.73 3.56 3.76 3.68 3.84 3.32 2.95 3.65 3.18 3.39 3.49 3.98 3.73 3.66 3.41 3.2 3.29 3.9 3.98 3.8 3.79 3.62 3.6 3.82 3.85 3.60 3.92 4.08 3.96 4.15 3.54 3.08 3.98 3.40 3.65 3.80 4.30 4.06 3.98 3.69 3.45 3.60 4.25 4.25 4.14 4.08 3.92 3.91 4.08 4.19 3.89 High High High High Low Low High Low Low Low High High High Low Low Low High High High High High High High High 1.11 1.12 1.05 1.13 1.06 0.95 1.05 1.03 1.01 1.03 1.03 1.04 1.13 1.01 0.97 0.98 1.04 1.06 0.98 1.01 0.99 1.03 1.01 1.01 1.03 Tolerable Tolerable Tolerable Tolerable Tolerable Intolerable Tolerable Tolerable Tolerable Tolerable Tolerable Tolerable Tolerable Tolerable Intolerable Intolerable Tolerable Tolerable Intolerable Tolerable Intolerable Tolerable Tolerable Tolerable (I) (I) (I) (I) (IV) (III) (I) (IV) (IV) (IV) (I) (I) (I) (IV) (III) (III) (I) (I) (II) (I) (II) (I) (I) (I) Keep up the good work Keep up the good work Keep up the good work Keep up the good work Possible overkill Low priority Keep up the good work Possible overkill Possible overkill Possible overkill Keep up the good work Keep up the good work Keep up the good work Possible overkill Low priority Low priority Keep up the good work Keep up the good work Concentrate here Keep up the good work Concentrate here Keep up the good work Keep up the good work Keep up the good work Figure 6. Analysis of collected data using the ZIPA matrix. THE SERVICE INDUSTRIES JOURNAL 17 Downloaded by [University of Florida] at 02:35 25 October 2017 Table 7. Statistical results of the proposed Kano-ZIPA model. No. ZAI level ZAP level ZIPA Zone Kano quality category Kano-ZIPA position X1 X2 X3 X4 X5 X6 X7 X8 X9 X10 X11 X12 X13 X14 X15 X16 X17 X18 X19 X20 X21 X22 X23 X24 High High High High Low Low High Low Low Low High High High Low Low Low High High High High High High High High High High High High High Low High High High High High High High High Low Low High High Low High Low High High High (I) (I) (I) (I) (IV) (III) (I) (IV) (IV) (IV) (I) (I) (I) (IV) (III) (III) (I) (I) (II) (I) (II) (I) (I) (I) Must-be One-dimensional Indifferent Indifferent Indifferent Indifferent Must-be Indifferent Indifferent Must-be Must-be Must-be Attractive Indifferent Attractive Attractive Must-be Must-be Must-be Must-be Must-be Must-be Must-be Must-be P1 P5 P13 P13 P16 P15 P1 P16 P16 P4 P1 P1 P9 P16 P11 P11 P1 P1 P2 P1 P2 P1 P1 P1 Table 8. Segmented quality improvement plan of Kano-ZIPA. Kano’s categories ZIPA zone I II III IV One-dimensional factor Attractive factor (P1) Minor strength X1, X7, X11, X12, X17, X18, X20, X22, X23, and X24 (P2) Major weakness X19 and X21 (P3) Minor weakness (P5) Major strength X2 (P9) Major strength X13 (P6) Major weakness (P7) Minor weakness (P4) Surplus X10 (P8) Minor strength (P10) Potential strength (P11) Minor weakness X15 and X16 (P12) Minor strength Must-be factor Indifferent factor (P13) Possible opportunity X3 and X4 (P14) Possible opportunity (P15) Non-critical X6 (P16) Redundance X5, X8, X9, and X14 fulfill them immediately. Second, two minor weaknesses (X15 and X16) are examined. These two attractive factors might generate satisfaction if addressed adequately; however, in this case, they are currently ignored. The managers should observe them. Third, two attributes (X2 and X13) of both high importance and high performance are considered major strengths. X2 is a one-dimensional factor, requiring an aggressive strategy to generate customer satisfaction; this is contrasted by X13, which has the highest ZAP value but is an attractive factor. A leveraging strategy exploiting X13 could distinguish the firm from its competitors. Fourth, 10 highly critical must-be attributes with high performance (X1, X7, X11, X12, X17, X18, X20, X22, X23, and X24) are considered minor strengths, where the service level must be maintained to prevent dissatisfaction. However, a highperformance level for these must-be factors does not increase customer satisfaction, and the managers should develop a plan to improve efficiency. Furthermore, two attributes (X3 and X4) are considered potential opportunities; managers can formulate 18 L.-F. CHEN ET AL. improvement plans to enhance these attributes, and transform them into attractive factors. Finally, four indifferent attributes of low importance (X5, X8, X9, and X14) with high performance are classified as redundant items, and one must-be attribute of low importance (X10) is classified as a surplus item. Attribute X6, which has the lowest ZAP, is noncritical because it is an indifferent factor of low importance and the low performance of this item neither affects customer satisfaction nor dissatisfaction. To optimize resource use, the firm should consider reducing investments in these attributes. However, X10 should be preserved at an adequate level. Downloaded by [University of Florida] at 02:35 25 October 2017 4.2. Comparison and discussion To compare the proposed Kano-ZIPA approach with traditional IPA, the importance of each attribute is derived using standardized regression coefficients, by conducting multivariate linear regressions of the AP ratings on overall customer satisfaction. The derived importance is adopted for comparison because it is the most commonly used indirect importance measure in IPA studies. The adjusted R 2 is .567, and the significance value (P) is .000 for the regression model. Table 9 presents an outline of the results of traditional IPA and the Kano-ZIPA model. Traditional IPA and the Kano-ZIPA model offer opposing suggestions for 41.67% (10/24) of the attributes (X5, X9, X12, X13, X14, X16, X17, X20, X22, and X23). Traditional IPA suggests that X5, X9, and X14 be addressed using a ‘concentrate here’ strategy, whereas the Kano-ZIPA model offers recommendations for ‘redundancy.’ Because these are indifferent factors, they should not receive additional effort and should even be deemphasized. Problematically, the strategies proposed by traditional IPA can result in wasted resources because of the incompleteness of the analysis. For example, traditional IPA assigns X12, X13, X17, X20, and X23 to the ‘possible overkill’ group, yet according to the Kano-ZIPA model, the attributes are a strength. Thus, a position in the ‘possible overkill’ group may result in a loss of strength if these factors are eliminated. Additionally, X16 is deemed ‘low priority’ according to traditional IPA, but a ‘minor weakness’ requiring immediate improvement according to the model proposed herein. Traditional IPA also places X22 in the ‘concentrate here’ quadrant; however, X22 is a must-be factor according to the Kano-ZIPA model, and thus we recommend that managers pay sufficient attention to this factor but focus on cost control. Although similar strategies are recommended to address some attributes, the KanoZIPA model offers more strategy details than does traditional IPA. Traditional IPA recommended addressing X15, X19, and X21 by using a ‘concentrate here’ strategy, whereas the Kano-ZIPA model defined them as ‘weaknesses.’ Unlike the Kano-ZIPA model, traditional IPA reminded managers to ‘concentrate here’ without providing a precise recommendation to direct the improvements. Traditional IPA provides directions to ‘keep up the good work’ regarding X1, X7, X11, X18, and X24, which is consistent with the Kano-ZIPA model categorizing these factors as a ‘minor strength.’ This recommendation means that the work should be maintained; however, these factors are also ‘mustbe’ factors, and traditional IPA does not discuss the cost control concern that Kano-ZIPA raises. Moreover, traditional IPA considers X3 and X4 ‘overkill,’ and hence, unlike the Kano-ZIPA model, fails to identify potential opportunities. THE SERVICE INDUSTRIES JOURNAL 19 Table 9. Comparison of the proposed Kano-ZIPA model and traditional IPA. Downloaded by [University of Florida] at 02:35 25 October 2017 Traditional IPA Kano-ZIPA No. AP score Standardized regression coefficients AP level AI level X1 3.96 .076 High High X2 4.20 .086 High High X3 3.85 .113 High X4 4.33 .014 X5 X6 X7 3.53 2.80 3.82 X8 X9 X10 X11 Kano’s quality categories Zone Decision M (I) O (I) Low Keep up the good work Keep up the good work Possible overkill I (I) High Low Possible overkill I (I) .048 −.056 .154** Low Low High High Low High I I M (IV) (III) (I) 3.29 3.42 3.61 4.10 −.262*** .294*** −.238*** .088 Low Low Low High Low High Low High I I M M (IV) (IV) (IV) (I) X12 3.87 −.154** High Low Concentrate here Low priority Keep up the good work Low priority Concentrate here Low priority Keep up the good work Possible overkill M (I) X13 4.13 .018 High Low Possible overkill A (I) X14 X15 3.44 3.09 .130* .076 Low Low High High Concentrate here Concentrate here I A (IV) (III) X16 3.22 −.084 Low Low Low priority A (III) X17 4.07 −.181** High Low Possible overkill M (I) X18 4.22 .265*** High High M (I) X19 3.72 .096 Low High Keep up the good work Concentrate here M (II) X20 3.81 −.104 High Low Possible overkill M (I) X21 3.60 .101 Low High Concentrate here M (II) X22 3.69 .119 Low High Concentrate here M (I) X23 3.86 −.012 High Low Possible overkill M (I) X24 3.89 High High Keep up the good work M (I) (P1) Minor strength (P5) Major strength (P13) Possible opportunity (P13) Possible opportunity (P16) Redundancy (P15) Non-critical (P1) Minor strength (P16) Redundancy (P16) Redundancy (P4) Surplus (P1) Minor strength (P1) Minor strength (P9) Major strength (P16) Redundancy (P11) Minor weakness (P11) Minor weakness (P1) Minor strength (P1) Minor strength (P2) Major weakness (P1) Minor strength (P2) Major weakness (P1) Minor strength (P1) Minor strength (P1) Minor strength .388*** Decision Ave 3.73 .041 Notes: 1. *P < .1; **P < .05; ***P < .01. 2. M: must-be, O: one-dimensional, A: attractive, and I: indifferent. The comparison shows that the proposed Kano-ZIPA outperforms traditional IPA for the following reasons. First, Kano-ZIPA provides a clear standard for the range of the horizontal and vertical axes, the measurement scale, and the placement of the crosshairs. Second, Kano-ZIPA adds additional concepts, such as the gap between the PS and AS, and solves the problem of misleading strategy suggestions that are triggered by slight changes in an attribute’s position. Third, Kano-ZIPA overcomes the shortcomings of traditional IPA and enhances its advantages with Kano’s quality categories. Finally, firms can easily conduct further investigations to improve service quality by interviewing more respondents or performing statistical analyses. 20 L.-F. CHEN ET AL. Downloaded by [University of Florida] at 02:35 25 October 2017 5. Conclusion To maintain their position in the market and use limited resources efficiently, firms must understand their customers’ needs. Moreover, they should examine their performance to identify service attributes and classify them appropriately. In this study, the advantages of IPA and the ZOT are combined to provide a simple and efficient tool for firms. In addition, the proposed Kano-ZIPA model, which integrates the ZIPA matrix with Kano’s categories, enables firms to devise suitable action plans. The case study of a wealth management department in the banking industry establishes the effectiveness of the proposed methodology: Based on the results, the department should conduct more in-depth investigations, such as statistical analysis on different customer segmentation to identify customer needs, and plan substantially stronger strategies accordingly. Furthermore, the proposed IPA matrix is demonstrably easy to implement after collecting data from the two-part questionnaire. This study proposes an effective approach, and implements it with objectivity and accuracy, despite some limitations in personnel, time, and external factors. For example, the questionnaires were administered in an active company in the banking industry. The respondents were all VIP customers in the wealth management department, and the services they receive require a certain degree of privacy. Therefore, the data were collected by several well-trained company staff, which facilitated the protection of the customers’ personal information; however, the company might also have distorted the real situation. We thus recommend a follow-up survey to be conducted by external staff to reduce this conflict of interest. In addition, the questionnaires were filled out and returned in one session. Factors such as limited time and participants’ emotions or distractions may have influenced the results. Because the analysis is based on the collected data and suggestions provided accordingly, a future survey with the full cooperation of the respondents is recommended. Other suggestions for future study are as follows. First, this study analyzes customer expectations for each service attribute by combining the strong points of IPA, ZOT, and Kano’s model; however, the results only reveal customers’ points of view, without considering the resources (money and others) spent on the service attributes, and how these might be affected through the suggested quality improvement strategies. Given that these strategies may bring about substantial change, future research is recommended to consider more viewpoints, such as those of human resources and financial departments. Second, to avoid a tediously long questionnaire that would inconvenience respondents, a consideration of service attributes was omitted in this survey. More service attributes can be examined in the future. Finally, the proposed approach should be tested in additional industries to confirm its validity and compare the differences that would affect the generalization. Disclosure statement No potential conflict of interest was reported by the authors. Funding This work was partially supported by Grants from Ministry of Science and Technology, Taiwan [MOST 105-2410-H-030 -033 -MY2]. THE SERVICE INDUSTRIES JOURNAL 21 Downloaded by [University of Florida] at 02:35 25 October 2017 References Albayrak, T. (2015). Importance Performance Competitor Analysis (IPCA): A study of hospitality companies. International Journal of Hospitality Management, 48, 135–142. Anderson, C. R., & Zeithaml, C. P. (1984). Stage of the product life cycle, business strategy and business performance. Academy of Management Journal, 27(1), 5–24. Arbore, A., & Busacca, B. (2011). Rejuvenating importance-performance analysis. Journal of Service Management, 22(3), 409–429. Azzopardi, E., & Nash, R. (2013). A critical evaluation of importance performance analysis. Tourism Management, 35, 222–233. Bacon, D. R. (2003). A comparison of approaches to importance/performance analysis. International Journal of Market Research, 45(1), 55–71. Berger, C., Blauth, R., Boger, D., Burchill, G., Bolster, C., DuMouchel, W., et al. (1993). Kano’s methods for understanding customer-defined quality. Center Quality of Management Journal, 2(4), 3–36. Berry, L. L., & Parasuraman, A. (1991). Marketing services: Competing through quality. New York, NY: The Free Press. Campos, D. F., & Nóbrega, K. C. (2009). Importance and the zone of tolerance of customer expectations of fast food services. Journal of Operations and Supply Chain Management, 2(2), 56–71. Chang, H. C., Liang, G. S., Chu, C. W., & Chou, C. H. (2012). Prioritizing service attributes for improvement using fuzzy zone of tolerance. International Journal of Innovative Computing Information and Control, 8(1A), 75–89. Chen, K.-Y. (2014). Improving importance-performance analysis: The role of the zone of tolerance and competitor performance. The case of Taiwan’s hot spring hotels. Tourism Management, 40, 260–272. Chen, L.-F. (2012). A novel approach to regression analysis for the classification of quality attributes in the Kano model: An empirical test in the food and beverage industry. Omega-International Journal of Management Science, 40(5), 651–659. Chen, L.-F. (2014). A novel framework for customer-driven service strategies: A case study of a restaurant chain. Tourism Management, 41, 119–128. Chen, L.-F. (2015). Exploring asymmetric effects of attribute performance on customer satisfaction using association rule method. International Journal of Hospitality Management, 47, 54–64. Chen, S.-H., Wen, P.-C., & Yang, C.-K. (2014). Business concepts of systemic service innovations in eHealthcare. Technovation, 34(9), 513–524. Chu, C.-H., & Guo, Y.-J. (2015). Developing similarity based IPA under intuitionistic fuzzy sets to assess leisure bikeways. Tourism Management, 47, 47–57. Churchill, G. A. (1979). A paradigm for developing better measures of marketing constructs. Journal of Marketing Research, 16(1), 64–73. Cohen, J. F., Coleman, E., & Kangethe, M. J. (2016). An importance-performance analysis of hospital information system attributes: A nurses’ perspective. International Journal of Medical Informatics, 86, 82–90. Cui, C. C., Lewis, B. R., & Park, W. (2003). Service quality measurement in the banking sector in South Korea. International Journal of Bank Marketing, 21(4), 191–201. Garver, M. S. (2003). Best practices in identifying customer-driven improvement opportunities. Industrial Marketing Management, 32, 455–466. Gelbrich, K., Gäthke, J., & Grégoire, Y. (2015). How much compensation should a firm offer for a flawed service? An examination of the nonlinear effects of compensation on satisfaction. Journal of Service Research, 18(1), 107–123. Grönroos, C. (1984). A service quality model and its marketing implications. European Journal of Marketing, 18(4), 36–44. Hsiao, Y.-H., Chen, L.-F., Chang, C.-C., & Chiu, F.-H. (2016). Configurational path to customer satisfaction and stickiness for a restaurant chain using fuzzy set qualitative comparative analysis. Journal of Business Research, 69, 2939–2948. Downloaded by [University of Florida] at 02:35 25 October 2017 22 L.-F. CHEN ET AL. Johnston, R. (1995). The zone of tolerance: Exploring the relationship between service transaction and satisfaction with the overall service. International Journal of Service Industry Management, 6 (2), 46–61. Kano, N., Seraku, N., Takahashi, F., & Tsuji, S. (1984). Attractive quality and must-be quality. The Journal of the Japanese Society for Quality Control, 14(2), 39–48. Keyt, J. C., Yavas, U., & Riecken, G. (1994). Importance performance analysis: A case study in restaurant positioning. International Journal of Retail and Distribution Management, 22(5), 35–40. Lai, I. K. W., & Hitchcock, M. (2015). Importance-performance analysis in tourism: A framework for researchers. Tourism Management, 48, 242–267. Lewis, R. C., & Booms, B. H. (1983). The marketing aspects of quality in emerging perspectives on service marketing association. Chicago: American Marketing Association. Lewis, B. R., & Mitchell, V. W. (1990). Defining and measuring the quality of customer service. Marketing Intelligence & Planning, 8(6), 11–17. Lo, C.-C., Wang, C.-H., Chien, P.-Y., & Hung, C.-W. (2012). An empirical study of commercialization performance on nanoproducts. Technovation, 32(3–4), 168–178. Martilla, J. A., & James, J. C. (1977). Importance-performance analysis. Journal of Marketing, 41(1), 77– 79. Matzler, K., Sauerwein, E., & Heischmidt, K. A. (2003). Importance-performance analysis revisited: The role of the factor structure of customer satisfaction. The Service Industries Journal, 23(2), 112–129. Murali, S., Pugazhendhi, S., & Muralidharan, C. (2016). Integration of IPA and QFD to assess the service quality and to identify after sales service strategies to improve customer satisfaction – a case study. Production Planning & Control, 27(5), 394–407. Nadiri, H., Kandampully, J., & Hussain, K. (2009). Zone of tolerance for banks: A diagnostic model of service quality. Service Industries Journal, 29(11), 1547–1564. Nunnally, J. C. (1978). Psychometric theory. New York, NY: McGraw-Hill. O’neill, M. A., & Palmer, A. (2004). Importance-performance analysis: A useful tool for directing continuous quality improvement in higher education. Quality Assurance in Education, 12(1), 39–52. Parasuraman, A. (2004). Assessing and improving service performance for maximum impact: Insights from a two-decade-long research journey. Performance Measurement and Metrics, 5(2), 45–52. Parasuraman, A., Zeithaml, V. A., & Berry, L. L. (1985). A conceptual model of service quality and its implications for future research. Journal of Marketing, 49(4), 41–50. Parasuraman, A., Zeithaml, V. A., & Berry, L. L. (1991). Understanding customer expectations of service. Sloan Management Review, 32(3), 39–48. Phillips, L. W., Chang, D. R., & Buzzell, R. D. (1983). Product quality, cost position and business performance: A test of some key hypotheses. Journal of Marketing, 47(2), 26–43. Rau, H.-H., Wu, Y.-S., Chu, C.-M., Wang, F.-C., Hsu, M.-H., Chang, C.-W., … Chiu, H.-W. (2017). Importance-performance analysis of personal health records in Taiwan: A web-based survey. Journal of Medical Internet Research, 19(6), e209. Sampson, S. E., & Showalter, M. J. (1999). The performance-importance response function: Observations and implications. Service Industries Journal, 19(3), 1–25. SINGFIN. (2017). Wealth Management in Taiwan 2017. Retrieved September 30, 2017, from http:// www.singfin.com/wp-content/uploads/Taiwan-Wealth-Management-Report-2017-Brochure.pdf Teas, R. K., & DeCarlo, T. E. (2004). An examination and extension of the zone-of-tolerance model: A comparison to performance-based models of perceived quality. Journal of Service Research, 6(3), 272–286. Tontini, G., & Picolo, D. J. (2010). Improvement gap analysis. Managing Service Quality, 20(6), 566–584. Voss, G. B., Parasuraman, A., & Grewal, D. (1998). The roles of price, performance and expectations in determining satisfaction in service exchanges. Journal of Marketing, 62(4), 46–61. Walker, J., & Baker, J. A. (2000). An exploratory study of multi-expectation framework for services. Journal of Service Marketing, 14(5), 411–431. Zeithaml, V. A., Berry, L. L., & Parasuraman, A. (1993). The nature and determinants of customer expectations of service. Journal of the Academy of Marketing Science, 21(1), 1–12.