Retail theft at self-checkouts doubles, AI poses possible solution

Retail theft, particularly via self-checkout points, has been increasing drastically over the past year. According to a Retail Threat Curve Report recently released by Everseen, cart-based loss (unscanned items left in the shopping cart at the end of the payment process) now accounts for 30% of all incidents at self checkouts, doubling over the past year.

This trend has represented significant financial loss for retailers, with average annual loss from this source alone exceeding $102,000 for a typical grocery store with 12 self-checkout lanes.

These figures are based on data collated from more than 1 billion transactions across 2023. As defined by Everseen, cart-based loss is a combination of unintentional and intentional (malicious) acts where shoppers may scan most items but leave more expensive ones such as meat and alcohol unscanned. This finding has alerted retailers to the vulnerabilities of self-checkouts and the increasing necessity for enhanced defence systems at these points.

Indeed, the retail industry's intensifying focus on inventory loss due to shoplifting has largely neglected the issue posed by cart-based loss. The National Retail Federation estimates that retail shrinkage - the discrepancy between advertised inventory and actual stock on hand - represents an approximately $112 billion problem. While efforts to combat shoplifting have led many stores to put products behind glass or in cages, the threat of cart-based loss at the checkout lingers heavily.

Alan O'Herlihy, Everseen's founder and CEO, underlined this issue by saying, "Retailers are under more pressure than ever to stop the high rate of lost inventory. If they do not have the proper defence systems at self-checkout, like AI-powered computer vision at this edge point to counteract the new threats, their EBITDA will continue to significantly deteriorate."

O'Herlihy added that previously retailers had no specific account for losses occurring at checkout and suggested these latest numbers could aid in focusing loss prevention efforts.

The recent surge in cart-based loss is not only a concern in terms of its frequency. The report also indicates that the average number of unscanned items left in the cart has increased from 1.6 to 3.8 per incident, with the average value of those items rising from $11.10 to $22.90. Therefore, retailers face an escalating threat both in the quantity and value of goods being stolen via self-checkouts.

Everseen's report underscores the need for retailers to reassess their theft prevention strategies and adequately address the vulnerabilities at self-checkouts. As cart-based theft continues and seemingly evolves, employing robust defence mechanisms such as AI-powered computer vision at self-checkout points may soon not be a mere option for retailers, but a pressing necessity.