Skip to main navigation Skip to search Skip to main content

Dynamically agile operation of mobile apps: a hidden Markov model

Xinhui Liu, Qiqi Jiang, Kaiwen Bao, Lele Kang*

*Corresponding author for this work

Research output: Journal Publications and ReviewsRGC 21 - Publication in refereed journalpeer-review

Abstract

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. © 2026 Elsevier B.V.
Original languageEnglish
Article number104320
JournalInformation and Management
Volume63
Issue number3
Online published12 Feb 2026
DOIs
Publication statusPublished - Apr 2026

Funding

This work was supported by the National Natural Science Foundation of China (NSFC 72072087).

Research Keywords

  • Agile operation
  • Dynamic capabilities
  • Hidden Markov model
  • Mobile apps
  • Poisson regression model

Fingerprint

Dive into the research topics of 'Dynamically agile operation of mobile apps: a hidden Markov model'. Together they form a unique fingerprint.

Cite this