Smart Green Nudging: Reducing Product Returns Through Digital Footprints and Causal Machine Learning.

This article empirically examines the use of personalized green nudging—informing consumers about the environmental impact of product returns—to reduce returns in e-commerce, based on a large-scale randomized field experiment (n = 117,304) with a leading European fashion retailer. The study finds that a dual green nudge, combining an informational prompt in the shopping cart and a postpurchase reminder, decreases product returns by 2.6% without harming sales, primarily by reducing "bracketing" behavior (ordering multiple variants to return some). Using causal machine learning (CML), the authors identify heterogeneous customer responses, revealing that personalized ("smart") nudging can approximately double the effectiveness of the intervention by targeting only those likely to reduce returns. The findings highlight the potential of combining subtle behavioral interventions with advanced machine learning to achieve both economic benefits for retailers and environmental gains, while also noting managerial challenges in implementing personalized strategies.