Customer orientation is crucial for business success, especially in innovation and product development processes. Considering its limited resources, an organisation must recognise which product attributes generate value for customers. A key to successful product development is the identification of customer needs and expectations. One of the most popular approaches to discover the characteristics of customer needs is Kano's two-dimensional model. It considers customer satisfaction as a multi-factorial construct implying that certain quality attributes are more important in ensuring customer satisfaction than others. Recently, various regression approaches have been developed to simplify the process of data collection and allow a deeper quantitative understanding of the asymmetric and nonlinear relationships between attribute performance and customer satisfaction. However, the usefulness of these approaches is currently under debate due to the lack of validity testing. This study assesses different regression approaches theoretically and empirically. Moreover, a regression approach based on the elastic net regression and the squared multiple correlation analysis is proposed. This newly developed regression approach outperforms the other regression approaches in terms of classification performance. This paper provides guidance to scholars and managers in selecting the appropriate approach to analyse customer quality requirements and satisfaction.

以客户为导向对于业务成功至关重要，尤其是在创新和产品开发过程中。考虑到其有限的资源，组织必须认识到哪些产品属性为客户创造了价值。成功开发产品的关键是识别客户的需求和期望。发现客户需求特征的最流行方法之一是卡诺的二维模型。它将客户满意度视为一个多因素结构，这意味着某些质量属性在确保客户满意度方面比其他属性更重要。最近，已经开发了各种回归方法来简化数据收集过程，并允许对属性性能和客户满意度之间的不对称和非线性关系进行更深入的定量理解。然而，由于缺乏有效性测试，这些方法的有用性目前正在争论中。本研究从理论上和经验上评估了不同的回归方法。此外，提出了一种基于弹性网回归和平方多元相关分析的回归方法。这种新开发的回归方法在分类性能方面优于其他回归方法。本文为学者和管理人员选择合适的方法来分析客户质量要求和满意度提供了指导。然而，由于缺乏有效性测试，这些方法的有用性目前正在争论中。本研究从理论上和经验上评估了不同的回归方法。此外，提出了一种基于弹性网回归和平方多元相关分析的回归方法。这种新开发的回归方法在分类性能方面优于其他回归方法。本文为学者和管理人员选择合适的方法来分析客户质量要求和满意度提供了指导。然而，由于缺乏有效性测试，这些方法的有用性目前正在争论中。本研究从理论上和经验上评估了不同的回归方法。此外，提出了一种基于弹性网回归和平方多元相关分析的回归方法。这种新开发的回归方法在分类性能方面优于其他回归方法。本文为学者和管理人员选择合适的方法来分析客户质量要求和满意度提供了指导。提出了一种基于弹性网回归和平方多元相关分析的回归方法。这种新开发的回归方法在分类性能方面优于其他回归方法。本文为学者和管理人员选择合适的方法来分析客户质量要求和满意度提供了指导。提出了一种基于弹性网回归和平方多元相关分析的回归方法。这种新开发的回归方法在分类性能方面优于其他回归方法。本文为学者和管理人员选择合适的方法来分析客户质量要求和满意度提供了指导。