Video Recognition Reduces Shrink
There are two ways to make money: make more and spend less. In the retail environment, reducing shrink equates to spending less. Making more is proving tough in our current economy, so retailers are looking for more robust shrink reduction solutions. Once upon a time the old security camera mounted on the ceiling may have been adequate deterrent, but over time the efficacy of this archaic practice has dwindled. New technology may make existing security cameras relevant again.
If not the just the security camera, than what?
A new solution from StopLift Checkout Vision Systems, Scan-It-All™ is a video recognition technology that mathematically analyzes the pixels of digitized video, scrutinizing how a cashier handles each item to determine whether or not it was properly scanned. The system is capable of understanding the full set of fraudulent behaviors. One such fraudulent behavior is “sweethearting” (the practice of pretending to scan merchandise; not charging the customer who is often a friend, family member or accomplice of one kind or another) by
StopLift reports detecting and confirming 130,000 U.S. incidents in 2010 alone, this includes both manned and self-checkouts.
Scan-avoidance is also responsible for a percentage of shrink. Shrink-avoidance occurs when the cashier fails to ring up merchandise. This may occur by mistake or by items left in the shopping cart; Middle of the Basket (MOB) and Bottom of the Basket (BOB). StopLift is also capable of detecting this form of shrink.
StopLift is conveniently based in the Cloud which means that any Flash-enabled tablet allows the retailer to see StopLift video of captured incidents of “sweethearting” and other scan avoidance and save the video indefinitely, instead of being relegated to the DVR and its storage limitations.
As soon as a “sweethearting” incident occurs, the software, which is constantly monitoring 100% of the security video, flags the transaction as suspicious. It quickly reports the incident to the retailer, identifying the cashier and the date and time of the theft.
StopLift’s “Scan-It-All” video recognition technology works with existing off-the-shelf overhead cameras. The retailer doesn’t need to purchase or install any new cameras or change anything about the point of sale.
StopLift’s patented computer vision technology visually determines what occurs during each and every transaction to immediately identify fraud at the checkout. Dishonest associates are identified on the basis of video evidence— the first time they steal, rather than months or even years later. This means significantly reduced inventory shrinkage, deterring future theft, and boosting profitability.
Malay Kundu, CEO of StopLift, the technology’s creator, explained that the technology eliminates costly, time-consuming human review of video, drastically reduces and deters fraud at the checkout, and significantly improves profitability. Rather than take a one-size-fits-all approach, StopLift develops targeted applications to address the specific needs of retailers from different sectors including general merchandise, grocery, and specialty retail.
“Malay Kundu is truly a visionary in addressing sweethearting,” said Mark Gaudette, Director of Loss Prevention at Big Y, headquartered in Springfield, MA. “This type of theft has been costing the retail industry in the U.S. $10 billion a year and, before StopLift technology became available, there’s been no way to detect it.
“We are excited to be able to control more of our shrink and loss through the use of this emerging technology,” he said. “In addition, StopLift has enabled us to improve our cashier work force overall through better training as well as better systems to detect and control employee theft.”
Retailers have tried to track retail theft through data mining, but, as Kundu notes: “How do you do data mining when there’s no data?”
Supermarkets, with their lower profit margins, are particularly vulnerable to sweethearting, which has accounted for an almost 35% profit loss industry-wide in the United States alone.
Tom Perkins, Director of Loss Prevention at Hannaford Supermarkets, based in Scarborough, Maine, which has implemented a pilot program using StopLift, said: “Having the Stoplift technology in place has enabled us to discover losses that we were previously unaware of. This is one of the most significant leaps in loss prevention technology I’ve seen in more than a decade. Despite several new kinds of loss prevention technology available to us, none of them has addressed sweethearting.”
StopLift Checkout Vision Systems grew out of Kundu’s Harvard Business School research study “Project StopLift” on Retail Loss Prevention. With technological research insights Kundu developed while at MIT, Project StopLift concluded that video recognition could be used to automate and, thus, make possible the comprehensive examination of surveillance video. Prior to founding StopLift, Kundu developed facial recognition systems for identifying terrorists in airports.