Loss prevention is a KPI-driven function, and as such it’s easy to get excited when compliance appears to be improving. But it’s important to remember that compliance improvements may not be as positive as KPI monitoring makes them appear. LP professionals should think carefully about improvements they see on paper and decide whether it is suspicious.
Imagine an LP professional reads a report showing that even with social distancing measures in place, the transaction rate for his district has actually increased 10 percent from the previous week. This may seem like something to celebrate at face value (efficiency is a good thing, right?), but upon closer inspection, this improvement is suspicious. Did the cashiers find a legitimate way to increase productivity despite social distancing in their checkout lines, and while cleaning the code pad after every use as required? Or could this be because the cashiers are not enforcing social distancing and other compliance measures?
The best way to identify suspicious improvements like the cashier scenario lies in data analytics. A good solution like prescriptive analytics can sift through the data and alert LP to any unexpected changes to KPIs, as well as how best to respond. Following are some examples.
Delivery Errors
A retailer adopted a prescriptive analytics solution to help its supply chain team verify compliance. The team created a pattern (an algorithm that identifies data behaviors and insights) that flagged when any store’s weekly credit-to-shipment (vendor credit issued vs. total dollar value of shipments) ratio fell at least 25 percent below its five-week average. The logic in the pattern was that either the delivery vendor had magically become 25 percent better, or it was still making mistakes (quite common in supply chain) and the retailer’s people were not reporting the errors as they should. If the latter was the real reason, the retailer was missing out on vendor credit.
The results were staggering. The pattern revealed the retailer was losing nearly $15,000 per month to vendor errors that were never submitted for credit. The retailer investigated every opportunity, following the guidance from the prescriptive analytics solution and uncovered training gaps within the credit-submission process. Retraining was conducted and compliance (and vendor credit) increased.
Unsold Publications
Depending on the publisher, grocers are often able to send back any unsold books and magazines for credit, especially the inventory on consignment. This process is mostly handled by the publisher’s direct store delivery (DSD) associates. When the process is followed, the retailer recovers hundreds of thousands of dollars. If the process isn’t followed, the losses can be hidden, but nonetheless accumulate to a high amount.
Thus, one of my grocery customers deployed a prescriptive analytics pattern to flag its publication vendor’s DSD associates who had delivered publications over the past two weeks, but submitted a below-benchmark number of unsold ones for reimbursement over the same period, as compared to similar stores.
The pattern was an amazing success. Dozens of opportunities were identified on a weekly basis across the network, including one notable case at a mall in which the DSD associate was unaware he had to pick up the retailer’s unsold publications outside the main entrance, since the pickup was at a time the mall was closed. Due to the misunderstanding, upon finding the mall closed, he simply skipped the pickups. This case alone cost the retailer nearly $4,000 in missed credit opportunities weekly. When presented with evidence of the non-compliance, the publication vendor issued the retailer a large credit as compensation based on the recent history of employees’ mistakes.
Waste Claims
Prescriptive analytics helps with both planning and ensuring compliance around food expirations — including rotations, pulling, markdowns, and damage reporting. By analyzing each store’s data, it identifies when items need to be rotated, marked down to move quickly (basically changing the demand curve by adjusting the product’s price), or pulled from the shelf. Additionally, it can be programmed with a grocer’s standard operating procedures and alert a relevant employee exactly how to take action on near-expired items.
As an example, an international grocer adopted a prescriptive analytics solution to ensure compliance around waste policies. Shortly thereafter, the solution sent a store manager an opportunity related to her store. The store’s waste claims for expired items that week were nearly 80 percent less than the previous week’s claims, far below the benchmark for the like-cluster stores. This decrease in waste seemed positive on paper, but the store manager could only think of three explanations for this curious anomaly:
- The store employees had pulled the expired products as required but had not filed their claims properly.
- The store employees had not pulled the expired products off the shelves, and they were still sitting there for customers to buy.
- Demand had surged beyond expectations, consuming all products on the shelf.
Either scenario presented a problem, so the store manager made a department visit to investigate and understand what happened. She found that, as feared, no expired products had been pulled from the shelves. After asking the department manager to do so, she interviewed the claims specialist on duty. It turned out the store’s full-time claims manager was out on vacation that week, and her replacement had not been adequately trained on processing waste claims, only doing rotation.
The store manager ordered retraining on claims processing for the replacement specialist, and the store’s waste practices soon returned to normal. Thanks to the prescriptive analytics solution’s alert, the issue was detected, and the expired products were pulled before any customers could unknowingly consume them.
When compliance appears to be improving, it’s important to check the data thoroughly to ensure it’s legitimate. A prescriptive analytics solution is a retailer’s best bet for cutting through any misleading figures and identifying what’s really happening behind the façade.
