Chatbot or Human? Rethinking Complaint Handling in Customer Service

ABSTRACT AI chatbots face major challenges in managing emotionally charged customer complaints. While they offer scalability, speed, and unlimited availability, they are perceived to be impersonal, particularly in service recovery situations that require empathy. This research examines when and how consumers accept chatbots in complaint‐handling contexts and how specific design attributes influence that acceptance. Across four pre‐registered studies conducted with student, professional, and online (Prolific) consumer samples, we employ between‐subjects and discrete‐choice experiments (DCEs) to test preferences for chatbot versus human agents and for chatbot attributes such as response speed, autonomy, and escalation dynamics. Study 1 ( N = 200, university participants) suggests that consumers prefer human agents to resolve complaints, even after a chatbot has successfully resolved their issue. Studies 2 and 3 ( N = 75 and N = 300, professionals and online consumers) reveal that faster response time significantly increases chatbot acceptance. Study 4 ( N = 160, university students) explores chatbot‐to‐human handovers and illustrates that early, user‐initiated transfers are preferred over delayed or those triggered by frustration. Together, the findings show that chatbot‐mediated complaint handling alters how justice and coping mechanisms operate in service recovery. Specifically, procedural efficiency can coexist with relational disengagement, and coping‐related adaptation becomes visible through behavioral choices enacted during the interaction. By revealing these shifts, the study advances justice and coping theory beyond their human‐centric assumptions and offers guidance for designing fairness‐oriented hybrid service systems.