Navigating the Holidays Returns Rush with AI

The winter holiday season is in full swing, and it’s no secret that “the most wonderful time of the year” is the most high-stakes time for retailers. This period is an opportunity to bring in record-breaking sales as consumers shop for gifts for loved ones, seasonal merchandise, and food for holiday get-togethers. Sales during the November and December months typically account for about 19% of total retail sales, and that number can be much higher depending on the sector. And this year, the National Retail Federation anticipates consumers will spend 2.5 to 3.5% more than they did in 2023.

While all this sales growth is an exciting prospect, retailers need to prepare for what inevitably follows a strong holiday season: the returns rush.

After Christmas and into January, retailers are faced with a staggering volume of returns as consumers return or exchange items they no longer want. A CRBE report predicts that 16% or $160 billion of all holiday sales will be returned this year.

As holiday returns pour in, retailers not only face increased operational burdens but also high costs. Each returned item needs to be shipped, inspected, restocked, and possibly marked down or liquidated—cutting deeply into profit margins.

To minimize the financial impact of returns, many retailers focus on stopping them from happening in the first place. These efforts include offering more detailed product information online, improving order fulfillment accuracy, and implementing paid returns policies. However, these preventative measures alone are not enough to keep inventory from coming back again. Retailers also need to rethink the way they manage returns behind the scenes.

Why Traditional Returns Management Falls Short

Most retailers still rely on outdated returns management processes, which are ill-equipped to meet omni-channel customer expectations to return items where and how they want.

The first issue with traditional systems is forecasting. Predicting future return patterns is crucial for effective inventory planning, yet most forecasting methods are inaccurate. Allocation and replenishment decisions fail to account for the influx of returned items, causing stockouts for some stores and overstock situations for others.

The second issue is lack of visibility for returns positioning. Ideally, each item would be routed where it will have the highest probability of reselling at the highest price before the season ends. However, typical inventory systems can’t help retailers determine optimal return paths. Instead, retailers end up using simple, static rules, such as:

• Send all online orders back to the closet fulfillment center.
• Ship all returned items to one central warehouse.
• Keep items in the store where they’re returned, even if that location does not carry the item.

This simplistic approach results in issues like warehouse bottlenecks and high transportation and handling costs. It also delays the reintroduction of products to the sales floor—if they make it back at all—which creates unnecessary stockouts, lost sales, and dissatisfied customers.

When returned products do make it back to the sales floor, they are rarely reallocated to an optimal location for resale. One-size-fits-all rules lack the sophistication to consider factors like seasonality, regional variations in demand, and individual store performance. Products end up sitting around unsold, eventually getting heavily marked down or liquidated at the end of the season.

All of these issues with traditional returns management ultimately hurt a retailer’s bottom line. The good news? AI and data analytics are powering modern strategies to optimize the entire returns process.