A longitudinal big data approach to theorizing consumers' continuance intention to use loyalty apps

Abstract

In an increasingly competitive fintech landscape, the ability to sustain user satisfaction and continuance intention beyond initial adoption is emerging as a resolute frontier. However, most prior studies have modeled continuance intention through survey-driven approaches and rarely explore organically expressed feedback in loyalty-centric fintech settings. This paper explores the nuances of post-adoption behaviors within fintech loyalty applications by analyzing the multilingual reviews posted by users on the app store as a way to mine their collective voice, or User Generated Content (UGC). Employing a sequential mixed-methods approach, we integrated confirmatory regression modeling of continuance behavior with topic modeling, sentiment analysis, hierarchical clustering, and Latent Dirichlet Allocation (LDA). Our analysis on 5669 app reviews revealed a dynamically-configured constellation of service quality, information quality, system quality, perceived usefulness, and evolving perceptions of risk. These elements enable user satisfaction and increase intention to continue using the service. It is worth mentioning that perceived risk silently acts as a dispositional control factor that can significantly erode loyalty away from positive user experiences. Merging classical theoretical approaches with unstructured user feedback enhances our understanding of loyalty within fintech in practice. This research provides a comprehensive framework to establish trust, satisfaction, and continuance use. These insights will enable developers, managers, and policy makers to forge resilient connections with users in a trust-scarce, value-starved, experience-saturated environment where lasting competitive advantage resides. We recommend that fintech loyalty platforms embed real-time feedback systems, visible security cues, and performance-aligned incentives to translate satisfaction into sustained user engagement.