Picture this: You’re at the grocery store scanning items at a self-checkout station. It’s fast and convenient—until an alert flashes. You’ve accidentally missed an item in your basket.
We’ve probably all been there.
While a common, one-time mistake can be easily corrected within just a moment or two, retailers are facing this issue on a magnified scale. It’s magnified across thousands of stores, hundreds of thousands of shoppers, and hundreds of thousands of items every day.
Retailers must address these checkout challenges and the growing problem of shrink, much of which occurs during the checkout process. The retail industry faces over $100 billion in annual losses, with self-checkout systems often pinpointed as a key transaction area and significant contributor to shrink.
But what if AI could turn this challenge into an opportunity, improving not just checkout, but the entire customer experience? For forward-thinking retailers, it’s doing just that and proving to be a game changer.
Understanding the Scope of Shrink at Checkout
Shrink at checkout doesn’t come in just one form. While some losses are due to deliberate theft, other losses stem from honest mistakes or “green shopping patterns,” where shoppers accidentally fail to scan items. AI can identify and address both types of loss in real time, keeping a vigilant eye on checkout patterns and mitigating these behaviors.
One VP of asset protection and safety from a supercenter retailer aptly points out: “Customers are not trained cashiers, so it’s not only malicious activity we need to prevent, but also honest mistakes.”
Here are two examples of how AI helps reduce honest mistakes and unintentional errors:
Produce recognition errors: Honest mistakes, like misidentifying produce (for example, selecting bananas instead of yellow squash), contribute to shrink. AI can catch these errors with a high degree of accuracy, even when items are bagged.
Basket-based loss: Items sometimes go unscanned when placed in a basket without being checked—such as soda or mineral water at the base of your cart. These oversights can result in significant loss.
Rather than punishing these actions, retailers can use AI systems to take a non-intrusive and continuous improvement approach. User-friendly alerts can let shoppers know that an unscanned item is in their basket or prompt a store associate to assist with the transaction.
Unfortunately, “red actors” are also out there, blatantly attempting to steal. These would-be thieves have a variety of techniques and patterns, which AI can be trained to identify.
Their “five-finger discount” techniques may have some cute sounding names—“Hula Hooping” for passing an item behind you without scanning it or “Banana Switching” for replacing a higher-priced item’s barcode with that of a cheaper one. However, the impact on retailers’ bottom lines is far from amusing.
AI systems can identify more than thirty common loss patterns, including:
Non-scan: The top cause of loss at checkout, whether at self-checkout or staffed lanes, where customers intentionally fail to scan an item. AI systems can identify those unscanned items and either prompt the customer to scan or alert a store associate to intervene.
Product switching: Shoppers scan a cheaper barcode, holding it underneath a more expensive item—a tactic that vision AI can detect and prevent.
Abandoned transactions: Shoppers leave the point of sale with items in hand before completing the transaction. AI can flag these incidents for intervention.
Many retailers are seeing the potential and necessity of technology in addressing these challenges: “If you don’t implement technology—ideally some form of self-learning AI at self-checkout—you’ll be taking a major hit to your profit margins,” said a VP of asset protection and security for a multi-department retail company.
AI Delivering Measurable ROI
Beyond these points in the store, AI solutions can also provide proactive insights, helping retailers identify trends, such as high-theft items and shrink trends by store location, offering operational changes to minimize shrink further.
Well implemented AI is delivering measurable ROI across retail environments. A recent study conducted by Forrester showed that, over a three-year period, grocery retailers could experience more than 300 percent ROI and achieve break-even in under six months.
Future-Proofing the Entire Retail Experience
The role of AI in retail is not just about reducing shrinkage. It’s about creating a smarter, more efficient shopping experience from the stockroom to the shelf and from the basket to checkout. AI’s impact goes well beyond loss prevention, improving key operations like workplace compliance, inventory management, queue management, and bulk-item checkout.
Retailers who invest in AI solutions aren’t just reducing losses—they’re driving profitability, improving customer experiences, and positioning themselves for future growth. As AI becomes an integral part of the shopping journey, it enables technology and the human touch to work together, driving profits, reducing shrink, and delivering a seamless customer experience.
Alex Siskos is the SVP of Strategy at Evereen, a leader in vision AI. He has more than 30 years of experience in the retail strategy and analytics. He previously served in leadership positions at Walgreens and led the entry of Blue Yonder into North America, which was later acquired by JDA. He also held business process strategy, market development, and analytics roles at Accenture, Nielsen, and Crossmark.