Heterogeneous Causal Learning for Optimizing Aggregated Functions in User Growth

TL;DR

This paper introduces a deep learning-based causal optimization method for user growth marketing that directly targets business metrics, outperforming traditional approaches by over 20% in cost-efficiency and impact.

Contribution

The paper presents a novel deep learning framework for direct causal effect optimization in marketing, capable of handling complex constraints and outperforming existing state-of-the-art methods.

Findings

Our method outperforms R-learner and Causal Forest by over 20%.

The approach is validated through real-world deployments worldwide.

It effectively optimizes treatment allocation for user growth campaigns.

Abstract

User growth is a major strategy for consumer internet companies. To optimize costly marketing campaigns and maximize user engagement, we propose a novel treatment effect optimization methodology to enhance user growth marketing. By leveraging deep learning, our algorithm learns from past experiments to optimize user selection and reward allocation, maximizing campaign impact while minimizing costs. Unlike traditional prediction methods, our model directly models uplifts in key business metrics. Further, our deep learning model can jointly optimize parameters for an aggregated loss function using softmax gating. Our approach surpasses traditional methods by directly targeting desired business metrics and demonstrates superior algorithmic flexibility in handling complex business constraints. Comprehensive evaluations, including comparisons with state-of-the-art techniques such as R-learner and Causal Forest, validate the effectiveness of our model. We experimentally demonstrate that our proposed constrained and direct optimization algorithms significantly outperform state-of-the-art methods by over $20\%$, proving their cost-efficiency and real-world impact. The versatile methods can be applied to various product scenarios, including optimal treatment allocation. Its effectiveness has also been validated through successful worldwide production deployments.