Demonstrates how a firm can incorporate visual features extracted from product images into a prediction model of return rates prior to a product’s launch. Based on a large data set from a European apparel manufacturer and retailer, which includes over 1.8 million online and offline transactions over four years involving near 10,000 unique fashion products; uses a machine learning approach that combines several image feature extraction tools to quantify information from images, and a gradient boosted regression tree prediction model.