A supermarketâ€™s loss prevention solution pays for itself in four months.
Grocery retailers, like all retailers, are susceptible to cashier-related shrink. By some estimates, 50% to 75% of all lost sales occur at the front of the store, due to theft, errors, and fraud. Niemann Foods, which operates more than 59 supermarkets and c-stores in the Midwest under the County Market, Cub Foods, and Save-A-Lot names, is no exception. The retailer, however, is enlisting technology to combat employee-related shrink. In mid-2006, Niemann Foods updated its March Networks LP Data Mining solution to expand its arsenal of shrink-fighting tools.
Niemann Foods originally installed LP Data Mining (known as ShrinkTrax prior to March Networks' acquisition of Trax Retail Solutions in July 2006) in 2001 to address customer- and cashier-related fraud. The solution produces exception reports used to address retailers' loss prevention, operations, and risk management needs. Its primary feature allows retailers to identify cashier transaction patterns — such as above-average numbers of voids, cancellations, and other unusual transactions — that might indicate theft, fraud, or errors. The resulting data is then analyzed to determine whether the flagged transactions necessitate further examination.
Keith Beckett, loss prevention manager for Niemann Foods, used LP Data Mining after its installation to review store transactions going back almost a year. He quickly identified patterns indicating potential problems at specific cashier stations. Within 16 weeks of installation, Beckett discovered two cases of fraud — one relating to refunds, the other to coupon redemptions. "I found that two individuals were responsible for 99.9% of abnormal transactions," he notes. "We produced evidence that made our cases much stronger for the legal system. We recovered $6,300 in one case and $9,800 in the other."
In mid-2006, Niemann Foods upgraded to the latest version of LP Data Mining, providing Beckett with even more tools to identify fraud at the POS. "The new version gives us more refined information, more quickly," says Beckett. "For example, with the new version, we can set up customized queries or scenarios — such as too many coupons being redeemed by a customer at one time or meat department refunds greater than $20 — and if such a transaction occurs, it will identify the KPI (key performance indicator) and e-mail the information to the relevant store within 24 hours of the occurrence. This automatic, daily reporting ability improves our reaction time, so if there's a problem, we can correct it more quickly."
The information provided by the new system reduces the time Niemann Foods' LP staff searches for dubious patterns, because the solution systematically generates reports of questionable transactions, purchasing, or cashiering patterns. The reports are based on criteria defined by Beckett and his staff, with help from March Networks. "I'm still learning everything the system can do for us, as it's very feature-rich," says Beckett. "But the information we can get via LP Data Mining has helped us reduce shrink so much that we figure the solution paid for itself in four months."
Integrate Exception Reporting, Video Surveillance
Niemann Foods wants to eventually integrate the LP Data Mining solution with its stores' video surveillance systems. Doing so would allow LP staff to locate the video for specific exceptions in a matter of seconds. Presently, to access video of a suspect transaction, Beckett notes the transaction time and cashier station from the LP Data Mining report, switches to the video surveillance system, and searches the video archive for the relevant footage. "When we acquired the Trax system, we didn't have digital recorders in all stores or a dedicated camera at each checkout lane," says Beckett. "We're getting closer to integrating the two systems, but for now, we operate them in parallel."
Promote Employee Excellence
The new version of LP Data Mining enables Niemann Foods to proactively reduce cashier-related shrink through training and employee recognition programs. The data mining metrics are used along with scorecards provided by the retailer's secret shopper program to recognize star performers. "The front end of the store is vulnerable to fraud," says Beckett. "It is a principal point of customer contact, so friendliness, customer service, and accuracy are qualities we promote. The transaction- and cashier-specific data generated with LP Data Mining helps us improve all facets of our operations."
For More Information On March Networks
Go To www.marchnetworks.com