Bridging the chatbot connection: the role of AI-driven chatbot affordances in e-commerce purchase intentions

Abstract

Purpose – This paper explores how AI-driven chatbots, using social support and affordance theories, influence consumer behavior and purchase intentions (PIs) in e-commerce, addressing challenges like limited understanding and security concerns. Design/methodology/approach – A quantitative study was conducted to collect data from 385 customers who had interacted with AI-driven chatbots on e-commerce platforms. Purposive sampling was employed to ensure the relevance of the respondents. Partial least squares-structural equation modeling (PLS-SEM) analysis was performed using SmartPLS 4 to test the hypothesized relationships and analyze the data. Findings – The results show a small direct effect of chatbot affordances on PI, with stronger indirect effects via customer satisfaction (CS) and trust. Notably, trust does not directly influence PI, suggesting a reliance on word-of-mouth. Customer engagement (CE) plays a minor mediating role, highlighting the importance of emotional and experiential factors. Originality/value – This pioneering study within the Sri Lankan context addresses the underexplored area of AI-driven chatbots and their influence on PI. Additionally, the study provides nuanced insights into the mediating mechanisms of CS, trust and engagement in the relationship between chatbot affordances and PI.