Bridging technology adoption and consumer behavior with predictive machine learning: An integrated UTAUT-EKB analytical framework

Abstract

The rapid development of electronic commerce requires a unified approach that integrates technology adoption and consumer decision-making processes. However, these domains have traditionally been examined separately, resulting in fragmented explanations of user behavior. Theoretically, the paper links the motivational predictors of the Unified Theory of Acceptance and Use of Technology (UTAUT) to stage-based progression in the Engel-Kollat-Blackwell (EKB) model of consumer behavior, and provides a cross-disciplinary account of technology-enabled consumer behavior. Using a moderate-sized dataset from the American Express digital offer program, machine-learning predictive analytics are employed to assess how UTAUT constructs shape transitions across EKB decision stages. The results show that performance expectancy and behavioral intention are the strongest predictors of adoption, with an XGBoost classifier achieving robust performance (AUC = .86). Funnel analysis shows a significant drop-off between evaluation and purchase, highlighting the role of perceived value and transaction simplicity. Practically, the results provide practical information to digital commerce platforms to allow them to optimize their targeting and reduce drop-off rates at each decision stage in the sequence.