Model-based Dashboards for Customer Analytics

TL;DR

This paper introduces a probabilistic Gaussian process-based framework for customer analytics dashboards that models and predicts individual spending behavior over time, capturing complex event impacts and outperforming traditional models.

Contribution

The paper presents the Gaussian Process Propensity Model (GPPM), a novel nonparametric approach that integrates calendar time, recency, and lifetime effects for customer spending analysis.

Findings

GPPM effectively captures spending dynamics influenced by unknown events.

GPPM outperforms hazard and buy-till-you-die models in fitting and forecasting.

The dashboard provides an intuitive visual representation of customer purchase behavior.

Abstract

Automating the customer analytics process is crucial for companies that manage distinct customer bases. In such data-rich and dynamic environments, visualization plays a key role in understanding events of interest. These ideas have led to the popularity of analytics dashboards, yet academic research has paid scant attention to these managerial needs. We develop a probabilistic, nonparametric framework for understanding and predicting individual-level spending using Gaussian process priors over latent functions that describe customer spending along calendar time, interpurchase time, and customer lifetime dimensions. These curves form a dashboard that provides a visual model-based representation of purchasing dynamics that is easily comprehensible. The model flexibly and automatically captures the form and duration of the impact of events that influence spend propensity, even when such events are unknown a-priori. We illustrate the use of our Gaussian Process Propensity Model (GPPM) on data from two popular mobile games. We show that the GPPM generalizes hazard and buy-till-you-die models by incorporating calendar time dynamics while simultaneously accounting for recency and lifetime effects. It therefore provides insights about spending propensity beyond those available from these models. Finally, we show that the GPPM outperforms these benchmarks both in fitting and forecasting real and simulated spend data.