A Machine Learning Approach to Examine Impacts of Customer Satisfaction on Financial Performance of Firms via Online Reviews

This study investigates the impact of customer satisfaction on a firm's financial performance, as reflected in online reviews. The study analyses 66,171 customer reviews of products from publicly listed firms in the healthcare industry on the Jingdong platform. Initial text analysis was conducted to identify key words, followed by the extraction of five topics from the review texts. Subsequently, six machine learning classifier models were developed and their performances evaluated. The random forest classifier achieved the best overall performance, with an area under the curve of 0.931. Customer satisfaction and brand image are identified as significant predictors of financial performance. Furthermore, non-financial indicators—such as quality and appearance, service attitude, effectiveness, and affordability—also contribute to financial performance. This study provides practical tools and valuable insights for managing online reviews, constructively responding to customer feedback, and enhancing customer satisfaction to promote sustainable financial growth.