Find a Professional

Find a Professional

Case Study: Using Sales Analytics to Reform Product Returns and Store Operations


Services and Operations Management: Using Sales Analytics to Reform Product Returns and Store Operations

An interview with Michael Ketzenberg

The Company

A national jewelry retailer with 1,100 stores in the United States and Canada.

The Challenge

The company was experiencing problems with product returns. Although it did not have an industry benchmark, at a 20 percent companywide return rate, varying from a low of 7 percent to a high of 35 percent for each store, the retailer felt improved profitability could be gained via insights from the product return and associated sales transaction data.

Key questions from management included:

  • What is the effective return rate?
  • What should the optimal rate be?
  • What does the distribution look like in terms of time to return?
  • Is the return policy being adhered to?
  • What are the top 10 SKUs for returns?
  • What customer service patterns correlate to high- and low-return performance stores?

The Solution

The client provided four years of transaction data—about 60 million transactions in total. Additional data was provided in the form of store operations data, number of employees, amount and type of customer service training, and employee years of work experience. The transactional data was also supplemented with detail regarding which employee sold which product.

The Results

Among immediate results, 4.5 percent of returns were found to be made on an expired policy. Fixing the process for verification of policy would recover $12 million in revenue. A second benefit was the identification of the suppliers most prone to returns, resulting in changes to product supply. Insight was also gained on specific customers, identifying the most profitable customers via combination of aggregate sales and their profitability, factoring in likelihood to return. Finally, the store experience was changed, via the observations of common practices at high- and low-performing stores and the advisement of experiments to replicate the associated best practices. Follow-up observations validating the experiments confirmed the best opportunities to change the return rate.