A Mechanism-Based Process Model of Postadoption Usage Dynamics: Evidence from Feature-Level Usage in Mobile Banking

Information systems (IS) usage is a widely recognized indicator of system effectiveness. However, prior research typically treated it as a static construct, operationalized it using aggregate measures, and overlooked its richness in feature-level usage and temporal evolution. Such limitations are particularly salient in mobile banking, where features vary substantially in the effort required, risks, and consequences, and where usage patterns change as user expectations are confirmed, habits are formed, and needs evolve over time. To address these gaps, we frame IS usage along two dimensions to reflect its richness: usage intensity, which reflects feature usage frequency and depth, and usage length, which represents the temporal span of continued engagement. We further argue that feature usage depth exerts a stronger influence on downstream outcomes than usage frequency. Based on expectation confirmation theory, habit formation theory, and the perspective of evolving user needs, we develop and test a mechanism-based process model of postadoption feature usage dynamics. Our model explains how and why postadoption usage evolves through shifts in mechanism dominance as users transition from evaluation-driven entry to habit-driven stabilization and need-driven disruption, resulting in different feature-level usage intensities and the emergence of distinct usage trajectories. By drawing on a multimethod design, in Study 1, we analyze 18 months of real-world mobile banking usage data to validate five theoretically derived longitudinal trajectories, namely, minimalist, decliner, centrist, maximizer, and expander, whose behavioral differences are linked with distinct downstream consequences. In Study 2, we examine the stage-contingent mechanisms underlying the trajectories by using survey data. Meanwhile, in Study 3, we combine a two-wave survey with a qualitative component to further validate the feature-level distinctions and the temporal process logic from users’ perspectives. The three studies offer a rich and dynamic understanding of postadoption usage and actionable implications for mobile banking design.

History: Jason Thatcher, Senior Editor; Huigang Liang, Associate Editor.

Funding: This work was supported by the City University of Hong Kong [Digital Innovation Lab of Information Systems Department]. This work was also supported by the Start-up Grant for New Faculty, City University of Hong Kong [Project No. 7200773].

Supplemental Material: The online appendix is available at https://doi.org/10.1287/isre.2024.0868.