مدل های تشخیصی سازمانی مشتری گرا مبتنی بر داده کاوی در پایگاه داده های شکایت مشتری

The purpose of this paper is to develop a customer-oriented organisational diagnostic model, ‘PARA’ model, based on data mining of customer-complaint databases. The proposed ‘PARA’ model, which is designed to diagnose and correct service failures, takes its name from the initial letters of the four analytical stages of the model: (i) ‘primary diagnosis’; (ii) ‘advanced diagnosis’; (iii) ‘review’; and (iv) ‘action’. In the primary-diagnosis stage, the customer-complaint database is comprehensively analysed to identify themes and categories of complaints. In the advanced-diagnosis stage, a data-mining technique is employed to investigate the relationship between the categories of customer complaints and the deficiencies of the service system. In the review stage, the identified weaknesses of the service system are reviewed and awareness of these weaknesses is enhanced among the organisation’s employees. In the action stage, a strategy of action plans for improvement is developed. An empirical case study is conducted to demonstrate the practical efficacy of the ‘PARA’ model. The paper concludes by summarising the advantages of the proposed model and the implications for future research

Customer satisfaction is recognised as one of the most important key performance indicators of success. However, it is always difficult to eliminate all causes of customer dissatisfaction and complaints. Customer satisfaction is influenced by such a variety of factors—including the attributes of a product or service, the individual needs of customers, and the service quality provided by front-line personnel—that even a fully ‘customer-focused’ service program cannot eliminate all product or service failures. Most organisations are aware that service failures must be handled appropriately to avoid harm to the organisation’s goodwill and profits (Hart, Heskett, & Sasser, 1990). Service recovery has thus become an increasingly important issue to prevent the loss of customers ( Kelley and Davis, 1994, McColl-Kennedy and Sparks, 2003, McColl-Kennedy et al., 2003, Sparks and McColl-Kennedy, 1998 and Varela-Neira et al., 2008). The term ‘service recovery’ refers to remedial actions that are taken to re-establish the satisfaction of customers when product or service failure has occurred ( Chaston, 1993 and Zemke and Bell, 1990). Many studies have demonstrated that effective service recovery can transform negative evaluations into positive impressions, thus maintaining good relationships with customers ( Hoffman et al., 1995, Karatepe, 2006, Kelley et al., 1993, Maxham Iii, 2001, Smith et al., 1999, Sparks and McColl-Kennedy, 1998 and Spreng et al., 1995). Appropriate service recovery has also been shown to enhance the trust of customers and increase their willingness to re-purchase ( Hung and Wong, 2007, Maxham Iii, 2001, Spreng et al., 1995, Tax and Brown, 1998, Tax and Brown, 2000 and Yu and Dean, 2001). Conversely, ineffective service recovery is one of the main causes of switching behaviour ( Keaveney, 1995). Most studies of service recovery (Barlow and Moller, 1996, Boshoff, 1997, Boshoff and Leong, 1998, Johnston and Fern, 1999, Keaveney, 1995, Tax and Brown, 1998, Tax and Brown, 2000 and Wirtz and Mattila, 2004) have focused on the effectiveness of specific remedial actions—such as exchanges of goods, apologies, or offers of compensation. Relatively few have studied the question of how to improve the service system by learning from the experiences of previous service failures and avoiding repetitions. It is the contention of the present study that the prevailing focus on remedial actions and compensation for a service failure is essentially a passive and reactive approach to the problem of service failure, whereas efforts to improve the existing service system represent a creative and proactive strategy. To improve a service system and minimise service failures, it is necessary to collect and analyse customer-complaint data periodically and comprehensively. Although some studies have indicated that improvement actions should be based on customer complaints (Bosch and Enríquez, 2005, Gustafsson et al., 1999, Tax and Brown, 2000 and Tax et al., 1998), these authors did not propose a comprehensive model to identify the sources of service failures. Chen et al., 2005, Chen et al., 2006 and Chen et al., 2004, who have investigated the importance of service-system design and management, also emphasised the need to make better use of customer-complaint databases to diagnose failures in service systems. Against this background, the present study seeks to establish a customer-oriented organisational diagnostic model of service failure based on customers’ complaints. The proposed ‘PARA’ model takes its name from the initial letters of the four stages of the model: (i) ‘primary diagnosis’; (ii) ‘advanced diagnosis’; (iii) ‘review’; and (iv) ‘action’. The diagnostic model provides a systematic analysis of service failures based on the customer complaint database. Data-mining techniques are then utilised to establish correlations between the identified categories of customer complaints. The model then develops a strategy of improvement actions for the service system. The model provides constructive customer-focused recommendations for improvements in service delivery through scientific analyses of service failures. The remainder of this paper is organised as follows. The next section reviews the relevant literature on organisational diagnosis. The development of the proposed ‘PARA’ model is then presented. The practical efficacy of the proposed model is then demonstrated in an empirical case study of public-sector services in Taiwan. The paper concludes with a summary of the main advantages of the model and the implications for future research.

The ‘PARA’ (primary diagnosis, advanced diagnosis, review, action) model presented in this paper seeks to have the voice of customer (VOC) taken into account in the diagnosis of failures within a service system. Based on the data mining analysis of a customer complaint database, the model enables a comprehensive diagnosis of service failures and develops appropriate improvement actions. The empirical case study reported in this paper has demonstrated the practical efficacy of the four stages of the ‘PARA’ model in diagnosing service failure and developing an improvement strategy in the context of regional public-sector services in Taiwan. The study has demonstrated three advantages of the ‘PARA’ model compared with extant organisational diagnostic models: • The core concept of the ‘PARA’ model is customer focus, which differs from the managerial and consultant perspective adopted by most other diagnostic models. The ‘PARA’ model thus addresses the gaps that exist between a service system and the real needs of customers. • The proposed model offers a comprehensive data mining of customer complaint databases, rather than dealing with customer complaints on a case-by-case basis. • The ‘PARA’ model integrates the views of customers, managers, and consultants in developing a consolidated improvement strategy of actions to eliminate the causes of service failure. Several suggestions for further research arise from the present study. First, in the ‘primary diagnosis’ stage of the model, future research might choose to use different approaches to ascertain the VOC, rather than the database of customer complaints used in the present study. Secondly, future research might choose to use other techniques of artificial intelligence—such as artificial nerve network (ANN), TRIZ theory, or text-mining techniques—to analyse the VOC in the advanced diagnosis stage of the ‘PARA’ model. Finally, the ‘PARA’ model might be applied in other service settings, such as service systems in the private sector or non-profit organisations.