Customer Sentiment in Web-Based Service Interactions: Automated Analyses and New Insights

We adjust sentiment analysis techniques to automatically detect customer emotion in on-line service interactions of multiple business domains. Then we use the adjusted sentiment analysis tool to report insights into the dynamics of emotion in on-line service chats, using a large dataset of telecommunications customer service interactions. Our analyses show customer emotions start out negative and evolve into positive feelings, as the interaction unfolds. Also, we identify a close relationship between customer emotion dynamics during the service interaction and the concepts of service failure and recovery. This connection manifests itself in customer service quality evaluations after the interaction ends. Our study highlights the connection between customer emotion and service quality as service interactions unfold, and suggests the use of sentiment analysis tools for real-time monitoring and control of web-based service quality.