A year ago, when COVID-19 closed down business facilities, it became impossible for many companies to execute inventory counts following common practices—and for auditors to observe counts on-site. The work-around for many companies was to use technology to place the auditor virtually on-site to observe the count and evaluate whether it was done correctly, as required by the auditing standards.
This was accomplished by outfitting company personnel with smart glasses, head-mounted cameras, or even smartphones to relay live video to the audit firm’s team. During the height of the pandemic, more companies also began to use statistical sampling to reduce the need for employees to spend time on-site to count every item.
Both of these approaches have proved reliable and, more importantly, they can make inventory counting faster and cheaper. As a result, we believe that these may become common practices long after the health risks of on-site observation of physical inventory counting are mitigated. This is the focus of our new whitepaper, Transforming inventory counting .
In this paper, we share insights from our experience working with clients to implement capabilities to facilitate remote inventory count observation procedures. In one pilot we ran, counters used smart glasses to feed video to remote observers. A count that had previously taken three days was completed in five hours.
We also learned that the live video feed gave companies a way to supervise inventory counting centrally and ensure that the counting was done correctly and consistently. Central monitoring can also give companies insights into how to improve the process.
Video monitoring is not for every type of business or for every type of inventory. Important variables to consider are the number of inventory locations, reliable WiFi connectivity, and the nature of the inventory—work-in-progress inventories and those requiring expert judgment to determine quantities may not be suitable. Companies should also understand the risks (e.g., fraud) and work with their auditors to understand the costs and benefits of remote observation.
Transforming inventory counting also provides a guide for how and when to use statistical sampling. This technique involves determining a statistically valid number of items to count to estimate the entire inventory. KPMG recently helped a major retailer introduce statistical sampling, using a targeted sampling precision of plus or minus 3 percent, at a 95 percent confidence level. When the sampling method was used, the entire inventory count was completed in one day—compared with two weeks for the traditional wall-to-wall count.
As with remote inventory count observation, statistical sampling does not work in all cases and involves risks that must be clearly understood. Before deciding to use statistical sampling, management should perform a thorough risk assessment, including an upfront analysis about the sufficiency and appropriateness of the sampling methodology being considered and the potential for bias.
We encourage you to read the full report and learn more about how these methods are being used today and into the future.