You can’t manage retail shrinkage if you can’t measure it
Moving beyond broad assumptions of where inventory loss is likely coming from, retailers should undertake a data-driven approach to measure the areas where losses are actually occurring as doing so enables the deployment of targeted mitigation techniques with the greatest potential impact. Given that brands are distinct, with unique product offerings, geographic footprints, market dynamics, operational systems, data landscapes, etc., the root causes of shrink vary widely, often necessitating tailored solutions working in concert to effectively address the challenge.
Without specific insight into the numerous sources of shrink for their business, retailers are left to leverage a disconnected network of siloed, off-the-shelf shrink solutions that often have custom purposes. This patchwork of solutions is not clearly and deliberately linked to the underlying root causes, leading to diminishing effectiveness across an enterprise issue and limited transparency into a tool’s return on investment. Frequently employed tactics of today are often blunt instruments that may help reduce shrinkage, yet they may also result in unintended consequences. For example, select retailers are testing private label–only stores since it is a working supposition that national brands are more desirable for resale by organized retail crime groups. Shifting a high-shrink store to private label only may in fact reduce shrink in that location, but it also may drive paying customers to nearby competitors or the brand’s other local stores, converting a loss problem into a possible sales and customer loyalty problem. With a unified view from multiple data sources retailers already have at the ready, a retailer could not only pinpoint the national brands most frequently stolen, but also could collectively mine point-of-sale transaction data and employee schedules, along with unstructured data, such as camera footage to glean insights as to the time of day when intentional harm is most prevalent and the customer and employee traffic patterns around the most frequently stolen goods.
Putting more context around the problem through the lens of aggregated data allows for more precise solutions, such as moving, removing, replacing or locking specific products; identifying bad actors; increased coverage at specific times; or modified hours. With all of the new technologies available and predictive modeling tools to assist in decision-making, a holistic data-driven model that pulls information from across the organization to pinpoint individual origins of shrink is not just possible, its pragmatic.
Rethinking retail shrink by connecting the dots
The closer you get to diagnosing the distinct root cause issues at a granular level, the more precise and targeted your mitigation techniques can be, which translates into less disruption to the customer experience and less costly investments in extreme countermeasures. When EY teams work with clients on the complex problem of shrink, we first leverage structured and unstructured data that already exists across the organization and look at it in new ways.
Our recommended approach includes these four critical steps: