Ensembling Experiments to Optimize Interventions Along Customer Journey: A Reinforcement Learning Approach - MSI

Working Paper

Ensembling Experiments to Optimize Interventions Along Customer Journey: A Reinforcement Learning Approach

Yicheng Song

University of Minnesota

Tianshu Sun

University of Southern California

Jan 7, 2022

Download

Develops a Bayesian reinforcement learning model to integrate data from multiple historical experiments to optimize interventions across the entire customer decision journey.

Topics

marketing analytics marketing experiments Omnichannels & Customer Journey

research recap

Optimizing Interventions Along the Customer Journey

Random controlled experiments for A/B testing can help companies improve their marketing or customer service, but optimizing interventions individually may not capture their interactions across the entire decision journey. A Bayesian reinforcement learning model can integrate multiple historical experiments to optimize interventions more holistically, improving both current impact and future learning.

By using MSI.org you agree to our use of cookies as identifiers and for other features of the site as described in our Privacy Policy.