Causal inference and model explainability tools for retail

TL;DR

This paper reviews and applies causal inference and interpretability techniques to retail data, demonstrating how explainable models and confounder control improve sales insights and causal understanding.

Contribution

It provides a practical recipe for applying interpretability and causal inference methods specifically to retail and e-commerce datasets, highlighting their benefits.

Findings

Explainable models have lower SHAP value variance.

Including multiple confounders yields correct causal effect signs.

Applying these techniques enhances sales insights in retail.

Abstract

Most major retailers today have multiple divisions focused on various aspects, such as marketing, supply chain, online customer experience, store customer experience, employee productivity, and vendor fulfillment. They also regularly collect data corresponding to all these aspects as dashboards and weekly/monthly/quarterly reports. Although several machine learning and statistical techniques have been in place to analyze and predict key metrics, such models typically lack interpretability. Moreover, such techniques also do not allow the validation or discovery of causal links. In this paper, we aim to provide a recipe for applying model interpretability and causal inference for deriving sales insights. In this paper, we review the existing literature on causal inference and interpretability in the context of problems in e-commerce and retail, and apply them to a real-world dataset. We find that an inherently explainable model has a lower variance of SHAP values, and show that including multiple confounders through a double machine learning approach allows us to get the correct sign of causal effect.