Abstract
This study improves the understanding of market dynamics and performance of mobile apps based on the NBD-Dirichlet model, a widely established stochastic theory of consumer behaviour. Using a standalone app for panel data collection, this study analyses 350,000ca. app accesses for 237 apps in four popular categories, over 30 days, in Brazil. It focusses on the comparison of observed and expected market size and loyalty of each app (including shared loyalty or duplication of app users). The results reveal the typical traits of stochastic behaviour. For example, the repertoire of apps consumers routinely use is predictable, and most apps share users with other apps in line with market size. However, there are some striking differences. Above all, the levels of app loyalty often vary more than the NBD-Dirichlet model predicts, given the app size. Overall, the findings significantly advance scholarly research on mobile apps, improving the understanding of the competitive structure of popular app categories. As we discuss, this intelligence can help stakeholders setting strategies for market survival and growth.
Keywords
Get full access to this article
View all access options for this article.
