From clicks to confidence: examining post-adoption satisfaction and continuance use of mobile trading apps
Аннотация
Mobile trading apps are a key component of the global FinTech landscape, enabling retail investors to access financial markets. Despite their rapid growth, these platforms operate in high-stakes environments where system reliability, transaction transparency, fraud risk, and ethical considerations shape users’ post-adoption evaluations. Prior research has primarily examined pre-adoption intentions using survey-based approaches, providing limited insight into users’ actual post-use experiences. This study develops a theory-informed, data-driven model of post-adoption satisfaction and continuance intention using an initial dataset of 5,500 Arabic-language app store reviews from three mobile trading platforms; after spam filtering and preprocessing, 3,381 high-quality reviews were retained for the final analysis. A sequential mixed-methods approach combined topic modeling and expert-guided construct development in the exploratory phase with Elastic Net regression and Random Forest in the confirmatory phase. The results indicate that Platform System Quality, Transaction Quality, Service Quality, Investment Utility, and Sharia Compliance are positively associated with linguistically expressed user satisfaction, whereas Fraud Concerns are negatively associated with satisfaction. User satisfaction is positively associated with continuance-oriented expressions, although the relatively modest explanatory power of the continuance model suggests that satisfaction alone does not fully explain continuance-related language in user reviews. Accordingly, the reported findings should be interpreted as associations among linguistically inferred constructs rather than independently validated causal relationships. The study contributes to FinTech and information systems research by contextualizing post-adoption theory for high-stakes mobile trading environments and by demonstrating a scalable, transparent, and theory-informed framework that integrates user-generated content analytics with machine learning.