A review-driven customer preference measurement model for product improvement: sentiment-based importance-performance analysis

An increasing number of people use social media to share their consumption experiences. Publicly available online reviews have become a significant source of information, which manufacturers use to better understand customer needs and preferences. To facilitate product improvement, this study first considers the inconsistencies between the numerical product ratings and the textual product reviews to establish the inconsistent ordered choice model (IOCM) for measuring customer preferences with regard to product features. The IOCM model effectively reduces the negative impact of inconsistent reviews on the quality of the customer preference measurement model. On the basis of customer preferences obtained from the IOCM model, we then develop a sentiment-based importance-performance analysis (SIPA) model to analyze the categorization of product features for guiding product development. Compared with the original IPA model, the proposed SIPA model in this paper introduces sentiment-importance into the IPA model that makes the product improvement more adaptive to customer preferences. Finally, we empirically evaluate the effectiveness of our proposed IOCM model and illustrate the utility of our proposed SIPA model.