Loss Prevention & Fraud Detection in Retail

In an industry known for its tight margins, any loss in retail is significant. Retailers often need to sell many multiples of an item to breakeven on a single loss. “Shrink” is a $60B dollar issue for the retail industry globally and growing rapidly. Most retailers attribute 50-60% of losses to associate actions. Increasingly, Organized Retail Crime (ORC) and Cybercrime rank at the top of the list of Loss Prevention (LP) concerns, along with associate fraud and returns fraud. Retailers are increasingly turning to technology to augment traditional LP methods, since those methods really do not prevent loss, but simply record it. Over 50% of asset protection executives state that they plan to invest in LP technology. Being able to identify loss as it happens allows retailers to take action and appropriately respond to it. It allows them to prevent losses and improve inventory accuracy.

When we hear the term “shrink,” we tend to think of shoplifting. The term “shrink” refers to action that negatively impacts the inventory accuracy, and hence, the ability to fully realize potential revenue. Losses can occur through theft, breakage/spoilage, administrative errors, supplier fraud (shorting orders), returns fraud (artificially inflating inventory), and more. Shrinkage means the retailer has less physical inventory than their system reflects. The result is they cannot realize the expected revenue based on perceived inventory levels. There are other downstream implications to inaccurate inventory, like out-of-stocks, or “outs,” which can prevent retailers from fulfilling online orders or meeting in-store customer demand. While a single “out” does not sound significant, for a mission-based shopper where their primary item is out of stock, the retailer may lose the rest of the basket. “basket” describes the value of a single transaction, e.g. a customer going to buy baby formula and a handful of other items may abandon their basket if the formula is out-of-stock. Worst case, the retailer may lose the loyalty of the customer who may take their business elsewhere.

Addressing Areas of Shrink at the Point-of-Sale (POS)

LP Professionals are trained to look for certain behaviors or actions to identify potential loss scenarios, Similarly, by integrating Meraki MV Cameras, POS terminals, and partner Artificial Intelligence (AI) models several areas of shrink can be addressed:

No Customer Present Transactions

Most retailers have policies in place that prevent associates from serving themselves to avoid the appearance of inappropriate behaviors. As such, any transactions taking place without a customer present are considered suspicious – this includes returns with no-customer present, particularly for cash or stored-value-card, which are untraceable. Similarly, an associate activating a gift card with no-customer-present is problematic. By integrating with the POS, the retailer can capture video or images of these high-risk actions for later review, based on the number of individuals in the frame at the time of the transaction.


If the number of items seen on the counter or conveyor belt do not match the number of items on the receipt, this can be an indicator of under-scanning. Under-scanning can take several forms: customers can obscure the barcode with another item or their hand at a self check or an associate deliberately does the same for a friend or family member. The latter action is known as “sweethearting.” This deprives the retailer of both the revenue and the product that is not scanned.

Label Switching

A customer may cover the barcode of a product with a barcode from a less expensive brand, or a completely different product. While not practical for all items in a store, it is possible to examine images of items being scanned, identify high-value or high-shrink items, like protein power, baby formula, razor blades, etc., and validate that the item scanned matches the image captured.


Cisco has the ability to identify theft and fraud at the POS and Customer Service Desk, improving store revenue and inventory accuracy. It can do this in other areas of the store as well. These smart cameras are not single use case devices and can also help provide customer and operational analytics, detect out-of-stocks for “fast movers,” improve merchandising decisions, and provide a safe and secure environment. They also frequently pull through other sensors to provide complementary data points. Lowering your “shrink” rate is possible and Cisco can help you achieve your goals.

See how Cisco’s portfolio of retail solutions provide the capabilities

retailers need to combat mitigate loss and prevent fraud.


Leave a Reply