Return rates in online retail: background and opportunities

Step by Step Toward a Clean Data Set

Follow these steps to ensure that your returns data is complete, comparable, and usable for further analysis:

• Regular data collection: Track all relevant metrics, such as packages shipped, items sent, merchandise value, and, of course, the quantity of returned items. Only with complete data can you reliably calculate the rates.
• The Right Calculation Method: Use the three standard metrics—alpha (per package), beta (per item), and gamma (per value)—to gain different perspectives on your customers’ return behavior. This will help you determine whether the primary concern is logistical effort, product selection, or financial loss.
• Segmenting the data: Break down the information by category, customer group, price segment, or even reason for return. This helps you understand where returns occur most frequently and which factors contribute to or even cause them.
• Integration & Automation: Integrate your online store, inventory management, and logistics systems so that returns data is captured automatically and, whenever possible, in real time. This saves time and minimizes errors.
• Analysis & Reporting: Set up regular reports and dashboards to highlight developments and identify trends early on. This makes it easier for you to make data-driven decisions.
• Monitoring Your Strategies: Even after making some improvements, you should continue to keep an eye on your e-commerce return rate. This will help you identify which strategies are effective and where you still need to make adjustments.

Key Internal Systems for Analyzing the Return Rate in E-Commerce

Once you've laid the groundwork for comprehensive and reliable data collection, the question arises: Where does the necessary data come from? It typically comes from your company's central systems. These provide the key metrics you need for calculations, comparisons, and trend analyses.

• ERP (Enterprise Resource Planning) Systems: An ERP system, such as SAP Business One or Microsoft Dynamics 365 Business Central, consolidates all core business processes. It provides you with data on orders, inventory levels, goods received and shipped, and invoices. When it comes to the return rate, it’s especially important that you have a clear overview of all shipped and returned goods transactions.
• Merchandise Management Systems (MMS): A merchandise management system like plentymarkets or JTL-Wawi focuses on the flow of goods. It provides data on inventory movements, shipped items, and returns at the item level. This makes it very easy to determine the beta return rate, and you can analyze which products your customers return most frequently.
• Logistics and shipping management systems: These systems, such as Shipcloud or Sendcloud, manage the processing of package shipments and returns. They provide data on shipped and returned packages, as well as information on shipping times and reasons for returns. This makes them particularly relevant for the alpha return rate.
• CRM (Customer Relationship Management) Systems: A CRM system like HubSpot or Salesforce collects and manages customer data as well as interactions with customer service. It provides insights into which customer groups return items more frequently and which service factors play a role in this. This information is valuable for developing targeted measures to reduce the return rate.
• Financial and accounting software: Programs such as Lexware or DATEV track sales, credit memos, and refunds. They are essential for calculating the gamma return rate because they show the actual financial impact of returns on total sales.

Factors Affecting the Return Rate

A recent EHI survey of retailers shows just how much return rates vary depending on the product category. Textiles are returned particularly frequently: For 12.2% of the textile retailers surveyed, the return rate exceeds 50%. In the electronics retail sector, by contrast, 85.7% of retailers report a return rate of up to 10%. Among furniture retailers, this figure is as high as 93%. Low return rates also predominate in the toy, book, and media sectors.⁴

In addition to the product category , price, product type, and target audience also influence the return rate. Returns are often less common for low-priced items because the effort involved in returning them is higher relative to the item’s value. In the fashion sector, on the other hand, fit, size, material, and discrepancies between how the product is depicted and the actual item are particularly common reasons for returns.

Assessment of the Return Rate in E-Commerce

The return rate is not a metric that can be easily compared across different companies and industries. What matters much more is the extent to which returns affect the profitability of a given store.

Returns don’t just incur shipping costs. Inspection, restocking, reconditioning, refunds, and potential loss of value also eat into profit margins. According to an EHI retailer survey from 2025, processing costs for 53.3% of retailers amount to up to 10 euros per returned item. Another 13.9% estimate costs of up to 20 euros. At the same time, 27% of retailers are still unable to quantify their return costs precisely. ⁴

Whether a return rate is problematic therefore depends primarily on the contribution margin. The costs of merchandise, packaging, shipping, marketing, returns processing, and overhead must leave sufficient room for a profit. Comparing the beta return rates of individual products can help identify particularly unprofitable items and product groups.

Prevent Returns Before a Purchase Is Made

AI offers particularly great potential even before a purchase is made. The better customers can assess a product, the lower the risk that it will be returned due to false expectations or an incorrect size.

• Improved product presentation: AI-powered tools create and optimize more engaging product descriptions, images, and videos. The more realistic and detailed a product’s presentation is in the online store, the less likely customers are to be disappointed after delivery, and the lower the return rate will be.
• More Accurate Size Recommendations in Fashion E-Commerce: One of the most common reasons for returns is the wrong size. AI-powered tools use body measurements, purchase histories, and, in some cases, image recognition to suggest sizes tailored to each individual. This reduces mispurchases and lowers the return rate in the long term.
• Virtual Try-On and Augmented Reality: Whether it's clothing, furniture, or home decor, AR applications allow your customers to virtually try out products before purchasing them. Customers can place a sofa in their own living room or digitally try on a piece of clothing. This increases purchase satisfaction and reduces returns due to unrealistic expectations.

Provide more targeted advice to customers and help them make purchasing decisions

In addition, AI can provide personalized advice to customers and reduce uncertainty during the purchasing process. It also makes it possible to identify and target customer groups with an above-average return rate more effectively.

• Behavioral Design to Reduce Uncertainty in the Purchasing Process: Using data-driven insights, AI can identify behavioral patterns in the ordering process and adapt the store design accordingly. Clear purchase recommendations, targeted assistance, and optimized user guidance eliminate uncertainty and reduce returns resulting from poor decisions.
• AI Shopping Assistant: While brick-and-mortar stores have staff members available to provide in-person advice, the digital space offers the advantage of anAI Shopping Assistant. This AI-powered shopping assistant, built on LLM technology, provides customers with conversational guidance and helps them find the right product more quickly. It identifies shoppers’ needs and actively suggests suitable products—in natural language and around the clock. Thanks to this targeted and personalized advice, customers can make confident purchasing decisions, and you, as the store owner, can prevent returns.

An AI shopping assistant provides conversational guidance that allows users to actively ask questions and receive personalized assistance. (Source: Own illustration)

• Identify and target frequent returners: Artificial intelligence analyzes customers’ purchasing behavior and order histories. This makes it possible to identify customer groups with an above-average return rate. Based on this information, as a store owner, you can, for example, develop targeted advisory services, offer alternative payment methods when necessary, or provide incentives to reduce the likelihood of returns.

Efficiently Process Returns After They Are Received

Not all returns can be prevented. In such cases, AI can help speed up the subsequent processing and facilitate economically sound decisions.

• Smart Sorting and Automated Return Evaluation: With the help of AI, incoming returns can be efficiently inspected and classified. The system identifies the condition, resale value, and potential refurbishment costs of individual items. This helps speed up return processes, reduce costs, and optimize the reuse of merchandise.