Paper Type
Short
Paper Number
PACIS2025-1484
Description
Humanized chatbots are widely used by online platforms to deliver cost-effective customer service. While prior research suggests that chatbot humanization often positively impacts user experience, these effects are often context-dependent. This study investigates the boundary conditions under which humanized chatbot designs can effectively improve customer service outcomes, with a focus on the contingent roles of service personalization and the nature of the service context (embarrassing vs. non-embarrassing). Two experimental studies were conducted using online tourism platforms as the empirical context. The results reveal that humanized chatbot design boosts consumer satisfaction and recommendation adoption when paired with high personalization. However, in less personalized or highly embarrassing contexts, the effects are weaker or even negative. These findings highlight that the success of humanized chatbot designs depends on contextual factors, underscoring the importance of selective implementation in customer service design.
Recommended Citation
Figueroa Cabrera, Juan David; Zhang, Jun; Zhang, Jingzhi; and Wang, Chuang, "Boundary Conditions for the Impact of Humanized Chatbots on Service Outcomes: Evidence from Two Experimental Studies" (2025). PACIS 2025 Proceedings. 12.
https://aisel.aisnet.org/pacis2025/hci/hci/12
Boundary Conditions for the Impact of Humanized Chatbots on Service Outcomes: Evidence from Two Experimental Studies
Humanized chatbots are widely used by online platforms to deliver cost-effective customer service. While prior research suggests that chatbot humanization often positively impacts user experience, these effects are often context-dependent. This study investigates the boundary conditions under which humanized chatbot designs can effectively improve customer service outcomes, with a focus on the contingent roles of service personalization and the nature of the service context (embarrassing vs. non-embarrassing). Two experimental studies were conducted using online tourism platforms as the empirical context. The results reveal that humanized chatbot design boosts consumer satisfaction and recommendation adoption when paired with high personalization. However, in less personalized or highly embarrassing contexts, the effects are weaker or even negative. These findings highlight that the success of humanized chatbot designs depends on contextual factors, underscoring the importance of selective implementation in customer service design.
Comments
HCI