Decoding customer experiences on meal delivery apps: A cross-platform text-mining analysis of online reviews through the lens of service psychology theories

The rapid growth of mobile-based meal delivery services has reshaped dining habits, yet our understanding of customer experience (CX) across competing platforms remains limited. Existing studies have often focused on single platforms or used only survey-based methods, overlooking how customer perceptions vary across multi-actor service ecosystems and how they align with established service psychology frameworks. To address this gap, we leverage qualitative text mining to examine 7500 Google Play Store customer reviews of Uber Eats, DoorDash, and Grubhub, collected through stratified random sampling. Using topic modeling, sentiment analysis, and cross-platform comparison, we uncover both shared and platform-specific drivers of CX. Our results show that platform-related issues dominate consumer discussions (similar to 60 %), followed by delivery riders (similar to 20 %) and restaurants (similar to 10 %), with recurrent concerns over service fees. Positive sentiments emphasize utilitarian gratifications such as convenience, app efficiency, and app usability, while negative sentiments focus on order issues, delivery delays, and customer support failures that are often attributed to platforms as supervisory entities. By integrating uses & gratifications theory, attribution theory, and Herzberg's two-factor theory, we help advance our theoretical understanding of utilitarian versus emotional benefits, responsibility attribution in multi-actor ecosystems, and the roles of hygiene versus motivator factors in shaping customer satisfaction. In practice, our findings provide actionable insights for managers to tailor platform-specific strategies, improve accountability mechanisms, balance utilitarian and hedonic features in digital design, and ensure the transparency of pricing models.