Striking the right balance: Customising return policy leniency for managing customer online return proclivity and satisfaction

Abstract

E-commerce retailers (e-tailers) commonly adopt generous return policies which not only act as a guarantee to protect the customer's purchase but also help in maintaining their satisfaction. However, this strategy can backfire by encouraging impulsive purchasing behaviour and resulting in a surge of product returns. This creates what is termed the product return policy leniency dilemma. To address that, this paper aims to empirically discover the relationships between product return policy leniency dimensions (time, monetary, effort, scope, and exchange) and two output variables – customer return proclivity and satisfaction. We develop a hybrid method combining machine learning-based data extraction and logistic regression, using a large empirical dataset comprising return policies and reviews from Walmart. The results show that three leniency dimensions – monetary, effort and scope drive customer return proclivity and satisfaction. Time only drives the satisfaction but not proclivity while exchange is in reverse. Our findings imply that customers are amenable to reasonable restrictions in return policies regarding time, effort, and exchange. However, overly lenient return policy terms may fail to adequately address the return policy dilemma. Additionally, partial refund/restocking fees are acceptable for customers with return proclivity if they perceive the initial purchasing cost heavily. Allowing some hazardous/bulky products to be returned under condition may also be seen as a generous term from prospective returners. Overall, e-tailers should display flexibility by incorporating different levels of leniency across five dimensions to balance return satisfaction and intention. This study provides e-retailers a guidance to design an appropriate bespoke return policy.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 12 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.