Data Science Day @ Columbia University

Automating the customer analytics process is crucial for companies managing distinct customer bases. In dynamic spending environments, visualization plays a key role in understanding events of interest. These ideas have led to the real world popularity of analytics dashboards, a tool that has received scant attention from academics. In this work, we develop a nonparametric framework for understanding individual-level spending using Gaussian process priors over latent functions that describe customer spending along calendar, interpurchase, and customer lifetime dimensions. These curves form a dashboard that provides a visual representation of purchasing dynamics. The model flexibly and automatically captures the impact of events that influence spend propensity, even when these events are unknown a-priori. We illustrate the use of our model on spending data from two popular mobile games, and show that it generalizes common customer base analysis models, leading to better performance both in fitting and forecasting spending incidence.