Follow Your Heart or Listen to Users? The Case of Mobile App Design

Firms strive to improve their products over time to compete effectively in the market. Typically, firms enhance products to stay competitive by adding novel features, imitating competitors, or leveraging customer input through social media. In the case of mobile apps, user feedback in the form of reviews includes suggestions of novel features or features that are already present in competing apps. Leveraging the information contained in reviews and version release notes of iOS apps, we develop a deep learning–based natural language processing approach to identify four types of app features: developer-initiated novel, developer-initiated imitative, user-suggested novel, and user-suggested imitative. We evaluate the impact of these feature categories on app demand. We observe that developer-initiated novel and user-suggested imitative features significantly boost app demand. Conversely, user-suggested novel features can negatively impact demand, especially when they are implemented contextually distant from user suggestions, though contextually close implementations have a positive effect. Although we observe that the aggregate impact of developer-initiated imitative features is insignificant, features that are slightly modified do have a positive effect on demand. The primary contribution of our study is to investigate user reviews as a source of ideas for new features and to evaluate their performance impacts relative to those of developer-initiated features.