How to enhance subway passenger satisfaction in the context of mobile payments and smart systems: a multi-city analysis in China

Purpose As subway networks expand and cities grow in China, mobile payment and smart access are central to operations, yet passenger satisfaction with these phygital (physical–digital) encounters is underexplored. This study aims to examine how subway app interface quality shapes overall passenger satisfaction across diverse Chinese cities. Design/methodology/approach Grounded in the stimulus–organism–response (S-O-R) framework, we analyze 18,487 App Store reviews of subway mobile apps from 10 representative Chinese cities. We apply machine learning to unstructured user-generated content (UGC), using PERT for feature extraction, latent Dirichlet allocation for topic modeling, and bidirectional encoder representations from transformers with whole word masking extension for fine-grained sentiment analysis. Findings Digital reliability is a critical hygiene factor for satisfaction. Drivers vary by city: high-density metropolises prioritize payment efficiency to ease time pressure, whereas developing cities are constrained by onboarding frictions (e.g. registration and verification). We also observe a spillover effect whereby app instability worsens perceptions of the physical ride experience. Originality/value We extend E-S-QUAL to smart mobility and validate the integration of digital and physical service quality within S-O-R using large-scale UGC. The findings offer actionable guidance for city managers and transit operators to improve resilient and inclusive service delivery and passenger satisfaction in the digital era.