Uplift Modeling for Multiple Treatments with Cost Optimization
This addresses a neglected use case in marketing and interventions where different treatments have different costs, but it is incremental as it builds on existing uplift modeling methods.
The paper tackles the problem of uplift modeling for multiple treatments with varying costs, extending standard models to support this scenario, and evaluates performance on synthetic and real data with a production implementation.
Uplift modeling is an emerging machine learning approach for estimating the treatment effect at an individual or subgroup level. It can be used for optimizing the performance of interventions such as marketing campaigns and product designs. Uplift modeling can be used to estimate which users are likely to benefit from a treatment and then prioritize delivering or promoting the preferred experience to those users. An important but so far neglected use case for uplift modeling is an experiment with multiple treatment groups that have different costs, such as for example when different communication channels and promotion types are tested simultaneously. In this paper, we extend standard uplift models to support multiple treatment groups with different costs. We evaluate the performance of the proposed models using both synthetic and real data. We also describe a production implementation of the approach.