Data analytics offers a lifeline for companies struggling with returns

In news that will surprise no one in the industry, product returns remain one of the most daunting challenges in today’s modern retail supply chain. Retail companies continue to struggle to establish an efficient, cost-effective reverse logistics network, with some even adopting a “keep it” customer policy.

However, some companies are beginning to achieve a level of success with returns by adjusting how they approach the issue. They are shifting some of the focus away from managing returns to instead exploring how they can best minimize the underlying issues causing returns in the first place. And data analytics, which has always been an invaluable tool in enhancing supply chain performance, is strengthening the effort. 

A consistent supply chain challenge

Current industry statistics point to the simple fact that returns will continue to be a major supply chain challenge that needs to be solved. And the longer it goes unaddressed, the more impact it will have on a company’s bottom line.

According to the National Retail Federation (NRF), returns in 2024 reached $890 billion, or 17% of total retail sales. This represents a 15% increase over 2023 and is more than double the return rate seen in 2019. Holiday returns were up even more—surpassing the 2023 holiday season by more than 20%. The costs associated with processing returns can exceed the recoverable value. This includes expenses from shipping, inspection, handling, as well as potentially repackaging, restocking, reselling and disposal. Estimates for the cost to handle a return can range from 20% to 65% of the items’ original value. 

This is why a number of companies have understandably adopted a “keep it” or returnless policy, especially for low-priced items where the cost of processing the return outweighs the item’s value. Most don’t advertise it as an official policy, but rather inform the customer, on a case-by-case basis, when a return is initiated. This type of policy can deliver cost savings by eliminating the handling of the return. It can also increase customer satisfaction and loyalty by giving the customer a “free item” and eliminating the hassle of a return.

However, these benefits are not without their downside. This can include customers feeling burdened for now having to dispose of an item they do not want or need. It also provides an opportunity for abuse, as some customers may try to exploit the system to receive free refunds and keep the product, which could grow as word-of-mouth spreads. And there is, of course, the lost profit from the item not being restocked or resold.  

This is why this approach should only be viewed as a short-term solution at best, employed as a company moves to establish and/or strengthen its reverse logistics infrastructure. It should be carefully implemented and monitored. 

Using data to get to the root of the problem

A more strategic, long-term solution for the returns issue is to instead go straight to the root of the issue—the reasons items are being returned—and fix it. While this would have been a very daunting task a few years ago, the proliferation of data in the supply chain and the advancement of data analytics have made this approach more viable and manageable. 

The ascendancy of advanced analytics in logistics and supply chain organizations signifies a notable shift in the paradigm of data utilization within the industry. From the early days of accumulating vast datasets and undertaking basic analysis, the industry now employs complex analytical methods to unearth insights and enable data-driven decisions along and across segments of supply chains. Companies are collecting and analyzing data on a wide range of factors across the entire supply chain, including sustainability, efficiency, productivity and product movement. 

Rather than merely having access to these large volumes of data (data quantity), logistics professionals can now intelligently use and analyze this data for strategic advantage (data quality), which is essential to answer the increasingly complex and specific questions asked by businesses today. Examples of this can be found in forecasting. For instance, machine learning can be used to analyze historical order patterns to improve peak demand management. The result is more accurate forecasts by workstream that enable companies to better prepare for demand fluctuations and efficiently manage extreme peaks without compromising service levels.  

This same approach and mindset can be used to help minimize the reverse logistics challenge. For those companies willing to make the investment and commitment, the opportunity exists to use much of the available supply chain data to fix the root cause of returns. By leveraging data insights at each stage of the product lifecycle, including product quality and packaging, businesses can identify root causes of returns and implement targeted improvements.

Collecting and analyzing data at key points

So, what does this approach look like? It should include the entire retail supply chain, from supplier locations (e.g., manufacturing plants, supplier distribution centers, ports) through distribution (e.g., retail distribution centers, return centers, retail stores).

Data should be gathered on four key areas: product quality (which includes accuracy), packaging, waste and supply chain performance. Each of these areas contributes to how well a product meets customer expectations, arrives in good condition and performs as intended.  

1. Product quality 

Aside from product accuracy, product defects or inconsistent quality are among the top reasons for returns. First, you must deliver the correct product. Assuming that happens, customers who receive a product that doesn’t function properly or that’s damaged are likely to initiate a return. Analyzing this data helps in the detection of manufacturing defects or quality issues, as well as pinpointing recurring issues. It allows for real-time quality control and root cause analysis. Data points to examine can include transportation of the product, movement of the product within facilities, the import quality of the product, and overages, shortages and damages (OS&D). 

2. Product packaging

Improper or inadequate packaging can lead to damaged products during transit, prompting returns and harming brand perception. Analyzing this data can identify which packaging types or materials lead to higher damage rates, and pinpoint where damage most frequently occurs in the delivery chain. It can also inform efforts to optimize packaging design for better protection and implement cost-effective, damage-resistant solutions. Data points to examine that go beyond just the consumer unit packaging that arrives at a customer location can include dunnage use, packaging compliance, pallet use and fit, and stretch wrap application as products move through the supply chain prior to final delivery. 
 

3. Supply chain performance

Delays, mis-shipments, and improper handling during transportation can contribute to customer dissatisfaction and returns. Analyzing this data provides visibility into shipping times, accuracy and handoffs, as well as supports better logistics planning and vendor management. It can also help flag inconsistencies in delivery performance by region or courier, and allows for root cause analysis of late or incorrect deliveries. Data points to examine can include procurement performance, inventory management, order fulfillment, logistics and transportation.  

4. Waste

Excess inventory, improper storage or the expiration of perishable goods can lead to degraded product quality, resulting in customers initiating returns. Analyzing this data can uncover issues with product shelf life and storage conditions, identifying inefficiencies or loss due to spoilage or obsolescence. It enables more accurate demand forecasting, reducing overproduction and supporting first-in, first-out (FIFO) and just-in-time (JIT) strategies. Data points to examine include shelving and storage, consumer feedback and order accuracy. 

A process should be established to examine the data in real-time, so serious issues and patterns that lead to preventable returns can be addressed as they are uncovered. Throughout the year and on a regular, ongoing basis, a report should be compiled that gathers all the data points and factors in historical customer data and competitive industry benchmarks on returns. This report should also include a list of recommendations on return processes, procedures and policies that, when enacted, will help reduce the number of returns initiated.   

Once this analysis of returns is completed, it can be integrated with additional data analytics efforts to further strengthen reverse logistics performance. This can include predictive analytics that help forecast return volumes more accurately by analyzing patterns in past returns, seasonality, customer behavior, and product categories. It can also include integrating reverse logistics data with inventory systems to enable faster inventory reconciliation and deployment of returned goods. 

Ensuring your data is usable and provides value

Finally, it is worth pointing out that data needs to be accurate and accessible to be useful in any decision-making, whether it be for returns or any other part of your supply chain.  Be mindful of the common expression, “garbage in, garbage out,” which conveys the idea that incorrect input will almost always lead to a failed output. When incomplete or poor data is shared that provides an inaccurate snapshot of the current state of the operation, it can result in ineffective or inappropriate resources and strategies being employed based on the false picture presented. 

Here are 10 tips for ensuring your data analytics strategy will help position not just your returns process, but your entire supply chain, for success.  

1. Adopt a mindset that connects the dots between data analytics and supply chain operations. Recognize that data provides the building blocks and visibility necessary for supply chain optimization. 
2. Eliminate data silos that impede enhancement efforts by integrating systems and advancing digitalization to enable synergies.
3. Evaluate your foundation to ensure you have the process optimization and labor management functionality within your supply chain systems that provide the data needed to orchestrate people, technology and equipment. 
4. Get your data in shape by identifying gaps or missing data sets and making sure the data is standardized, accessible, digestible, actionable and trustworthy. 
5. Gather data from your entire supply chain, as this will provide a more complete view of how your supply chain is performing and where improvements can be made. There is valuable information everywhere, but some areas are overlooked.
6.  Take a 360-degree view that includes data gathered from every part of your supply chain (e.g., suppliers, distributors, vendors, retailers, etc.) to strengthen the insight gained.
7. Let the data tell the story. Finding correlation is easy and not the mission. Use data to identify the causes of returns and then create appropriate solutions. Don’t use the data to create solutions in search of a problem to solve. 
8. Develop use cases that identify areas where datasets can be used to implement enhancements that deliver operational value and resolve issues draining productivity and hindering efficiency. 
9. Establish best practices based on successful applications that can further optimize your data analytics strategy and guide new implementations.
10. Have a plan to expand by creating a framework and process for continually adding new data sets that provide further opportunities to create synergies throughout the supply chain, inside and outside the warehouse.

About the author

Thomas Borders, vice president, operations, DHL Supply Chain North America. He has led diverse teams across a variety of industries and functional areas and is passionate about delivering the best end-to-end post-purchase solutions for our customers.

SC
MR