Adoption Satisfaction in Citizen Service Portals: A Predictive Study of Usability Testing Routines

Abstract

The digitalization of public administration has rapidly transformed how citizens interact with government entities, positioning citizen service portals as critical infrastructure for modern governance. Despite massive investments in these digital platforms, user adoption rates frequently fall short of expectations, largely due to underlying usability issues that traditional evaluation metrics fail to capture adequately. This paper explores a novel methodological approach to predicting user adoption satisfaction by applying process modeling techniques to usability testing routines. By conceptualizing user interactions within these portals as sequential business processes, we extract dynamic behavioral patterns from event logs generated during usability tests. These process models reveal latent navigation bottlenecks, loop behaviors, and procedural deviations that static usability surveys overlook. Leveraging a mixed-methods research design, empirical data was gathered from a diverse cohort of participants interacting with a simulated municipal service portal. The extracted process features were subsequently utilized to train predictive models capable of forecasting long-term adoption satisfaction. The findings demonstrate that process-derived metrics significantly enhance the predictive accuracy of satisfaction outcomes compared to traditional self-reported measures. This research contributes to the domains of human-computer interaction and e-government by providing a robust, data-driven framework for anticipating user acceptance, ultimately guiding the design of more inclusive and efficient digital public services.