Dynamically Agile Operation of Mobile Apps: A Hidden Markov Model

Agile operations are increasingly important for managing product development within the dynamic digital product market. Recognizing the unique characteristics of mobile apps, this study investigates how app developers adopt and dynamically adjust agile operation states over time and examines how these states impact app performance. Drawing on 24 months of longitudinal data from 254 leading iOS mobile apps, we employ a hidden Markov model to identify and analyze distinct agile operation states. Our results reveal four unique agile operation states, highlighting that high update frequency and strong adoption of user feedback are unlikely to co-occur within a single state. We further find that transitions between these states significantly influence app market performance. Specifically, when apps exhibit very low overall update frequencies, strategically reducing user feedback adoption to increase update frequency moderately impacts performance positively. However, increasing update frequency significantly while disregarding user feedback affects performance negatively. Additionally, for hedonic apps, enhancing update frequency proves more beneficial for performance than extensively incorporating user feedback.