Artificial intelligence-enabled proactive service recovery in business-to-business markets: a dynamic capabilities perspective on agility and decision-making

Purpose This study aims to examines the extent to which artificial intelligence (AI)-enabled proactive service recovery influences service agility and AI-driven recovery decision-making in business-to-business (B2B) markets. The study moves from reactive to proactive recovery models and conceptualizes service recovery as a sequential capability-deployment process. Design/methodology/approach Drawing on the dynamic capability view and organizational information processing theory, this study models the links among sensing and seizing capabilities, customer engagement and service recovery performance. The measurement model was validated using exploratory and confirmatory factor analyses based on survey data collected from 408 managers of B2B firms. Furthermore, the hypotheses were tested using structural equation modeling. Findings AI-enabled proactive service recovery significantly enhances both service agility and AI-driven recovery decision-making, with a stronger effect observed for decision-making capability. Service agility increases customer engagement, and both customer engagement and AI-driven recovery decision-making improve service recovery performance. Service recovery performance represents the focal outcome of the proposed capability-deployment process. Originality/value This research advances the B2B service recovery literature by conceptualizing AI-enabled recovery as a sensing capability and presenting a parsimonious, process-based dynamic capability model. By distinguishing between analytical and operational seizing mechanisms, the study offers a more detailed explanation of how AI-enabled organizational capabilities translate into relational engagement and effective service recovery performance.