Best of Three Worlds: Adaptive Experimentation for Digital Marketing in Practice

TL;DR

This paper explores adaptive experimental design (AED) in digital marketing, highlighting challenges in non-stationary industrial environments and proposing a framework for counterfactual inference tested in practice.

Contribution

It provides insights into AED challenges in real-world settings and introduces a new framework tailored for non-stationary industrial applications.

Findings

Identified key challenges of naive AED in non-stationary environments

Developed a practical AED framework for counterfactual inference

Tested the framework successfully in a commercial setting

Abstract

Adaptive experimental design (AED) methods are increasingly being used in industry as a tool to boost testing throughput or reduce experimentation cost relative to traditional A/B/N testing methods. However, the behavior and guarantees of such methods are not well-understood beyond idealized stationary settings. This paper shares lessons learned regarding the challenges of naively using AED systems in industrial settings where non-stationarity is prevalent, while also providing perspectives on the proper objectives and system specifications in such settings. We developed an AED framework for counterfactual inference based on these experiences, and tested it in a commercial environment.