A Data-Driven Design Framework for Customer Service Chatbot

User experience in customer service is critical. It is because customer service is what a customer first requests for a service. The service fails to satisfactory response will cause a crucial damage. Albeit business includes a chatbot for better responsiveness, customization is still necessary to fulfill the satisfaction from customer service. For customization, a designer performs qualitative research such as surveys, self-reports, interviews, and user observation to pull out key characteristics and to build personas based on the characteristics. However, a small sample size and cognitive limitation of a researcher demand more data to model persona better. Therefore, in this study, we introduce a data-driven framework for designing customer service chatbot that utilizes the past customer behavior data from clickstreams and a customer service chatbot. We apply this framework to a cartoon streaming service, Laftel. In result, we generate three types of customer service chatbots for three personas such as explorer, soft user, and hard user. In the future, we will validate our result by conducting a field experiment.