Artificial Intelligence (AI) has quietly snuck into daily life. You may fear its power due to popular dystopian Hollywood hits, but innately you know AI is more effective than the human brain. 

Just think about how many times a day you turn to Siri, Google, or Alexa to tell you something you don’t know, like the hour-by-hour forecast, the 27th president of the United States, or how many ounces are in a cup. Think about smart responses on Gmail, automatic tagging on Facebook, and Google predictive searches. So why not extend the use of AI to all businesses, optimizing every inefficient and costly process? 

Consider retail returns for a moment. Due to the rise in online shopping, retailers have experienced an explosive growing problem in costly and mismanaged returns. In fact, industry experts project returns will cost retailers $550B by the end of the year. And that doesn’t include restocking expenses or inventory losses.

Could AI Help Solve the Returns Problem?

Dealing with the increase in returns has led companies to scramble for methods to manage them. But these methods are often manual, and reliant on human resources and decision making. In short, returns are such a headache for retailers that destroying returned or distressed inventory is often more profitable than re-circulating products back into the sales cycle. 

It’s time for a shift. Smart retailers must start looking at this historically challenging process as an opportunity to incorporate AI into their business model. AI and deep learning can ease a complicated reverse logistics process, and optimize in-store and online operations. With AI, retailers can take human thinking out of the equation and utilize machine learning to make the most effective disposition decisions. 

So what’s AI? What’s deep learning? And how can these tools drive more profits from retail returns? Here we explore. 

What Is Artificial Intelligence?

AI is a broad term that refers to a machine’s ability to conduct human cognitive functions, such as learning and solving problems. Through augmented artificial intelligence, machines can learn from human input allowing people to make better decisions based on AI’s information. For retail returns, this means machines can continually evolve to make smarter decisions about how to recover the most value. 

What is Deep Learning?

Deep learning refers to a machine’s use of artificial neural networks (ANNs) to facilitate various layers of learning. Instead of task-based algorithms, deep learning utilizes algorithms inspired by the human brain learning to make decisions with big data. Deep learning is akin to humans' ability to learn from experience–except machines do it much better. Deep learning has revolutionized analytics, enabling practical applications of AI for humans and businesses. For retail returns, AI means better decision making early in the supply chain to drive higher profits. 

Artificial Intelligence & the Business of Returns 

Through AI-based software, smart retailers can grow their bottom line and discover new market opportunities. According to a 2018 survey of 400 retail executives, AI could save retailers as much as $340 billion a year by 2022. The report projects that 80% of the savings would come from AI’s ability to optimize traditionally human-based processes for supply chains and returns. 

According to a survey by Tech Republic, only about 24 percent of businesses currently use or plan to implement artificial intelligence in their model. That means retailers have an exciting opportunity to consider implementing AI into their returns management strategy. 

Disposition Engines Are Reshaping Returns Management

Many retailers have no idea about the existing technology that’s revolutionizing returns management. One example of returns management technology based on machine learning is goTRG’s Returns Automated Disposition Application –”RAD”. RAD is a decentralized application that can be used on a mobile or in-store device, or integrated into an ecommerce sales platform, to employ real-time disposition decisions based on curated data and algorithms.

How does RAD work?

Immediately after scanning the product UPC, “RAD” app enables an in-store or online disposition for that item. goTRG’s RAD App is used in thousands of stores across North America by some of the savviest, most forward-thinking retailers. RAD’s disposition engine makes real-time decisions based on several factors, including customer demand, dimensions and weight for transportation expenses, repair and remarketing costs, velocity of sales, and ever-changing pricing trends. Rather than relying on biased employee decision making, this automated process uses deep learning to quickly process, reroute, and track merchandise, boosting efficiency. 

The ability to make a data-informed disposition at the point of return allows for an item to be routed directly on a path to its next destination, instead of being held up in a consolidation or return center for weeks or months. For seasonal items, or trendy items that are steeply discounted once the next fad has emerged, this quick turnaround is critical for recovery potential. Disposition decisions can include restock, refurbish, liquidate, recycle, donate or, as a last resort, destroy. By making the disposition decision further upstream, retail brands can decrease the number of touches needed or wasted transportation miles, thus increasing the profitability of the item.

goTRG’s uses data from the largest privately curated catalog in the world, with over  100M unique UPC templates, making us the first reverse logistics company to use big data to drive disposition decisions. Today, we work with 6 out of the 10 top retailers in the world, employing machine intelligence to revolutionize returns management through smarter disposition decisions.  


Through embracing AI, humans and businesses alike can optimize inefficient daily tasks and processes. Retailers in particular can decrease operational costs, remove touch points, improve sustainability efforts, and create value with few additional expenses. When retailers have reliable data and recommendations through AI machine learning, they can make better choices for their business, and recoup enormous revenue on returns. This is increasingly vital as returns rise in the digital world. The time to shift is now.