Can Chatbot Anthropomorphism and Empathy Mitigate the Impact of Customer Anger on Satisfaction?

When customers initiate inquiries with negative emotions following a service failure, whether chatbot service agents can alleviate the undesirable outcomes resulting from negative emotions poses significant challenges for researchers and practitioners. Drawing upon computers are social actors (CASA) framework, this study examines how chatbot anthropomorphism and empathy features function as emotion-focused service recovery to mitigate the adverse effect of customer anger on satisfaction. The model is validated through a three-way factorial between-subjects experiment. The results demonstrate that anger negatively affects satisfaction. Chatbots with empathy features can mitigate this negative impact, whereas chatbots with anthropomorphism features cannot. The findings provide theoretical and practical implications for designing chatbot service agents to achieve emotion-based service recovery.