A Comparison of Return Rate Calculation Methods: Evidence from 16 Retailers

The product return rate (RR) is an important metric for retailers; even small RR changes can significantly impact retailers' profit. Companies and researchers typically favor and employ one of three methods to calculate the RR: based on the number of returned items, these items' revenue, or their profit contribution. Interviews with 24 managers and industry experts reveal that two methods, item-based and revenue-based, are often used. However, little is known about how much the interpretation of RRs depends on the calculation method. In this article, the authors rely on extensive datasets to investigate these potential differences empirically. Analyzing more than 8 million transactions at sixteen different retailers, the authors find that RRs calculated via the three methods differ on average by 24.3%. The size of these differences makes cross-sectional comparisons of RRs calculated via different methods difficult. In contrast, the authors find that the developments of the RRs over time are similar, which allows for a meaningful time-series comparison of RR developments, regardless of the method. Finally, this research shows that all three calculation methods result in RRs that are equally insightful (leading) indicators of relevant retailer performance metrics. (c) 2021 New York University. Published by Elsevier Inc. All rights reserved.

Automated summary

This article investigates the product return rate (RR) calculated via three different methods and finds an average difference of 24.3% between them. Despite this, the trends of RR over time are found to be similar, allowing for meaningful time-series comparisons. Ultimately, the research demonstrates that all three calculation methods result in equally insightful leading indicators of relevant retailer performance metrics.