Perpetual Inventory

In business and accounting, the notion of perpetual inventory refers to a system, or in practice a software, where the information on the inventory quantity and availability of SKUs (stock keeping units) is updated continuously based on the records associated with business operations, primarily orders and stock movements. The goal of perpetual inventory is to maintain an accurate estimation of the real stock levels with as few stock counts as possible.

De facto design for modern software

All modern inventory management software leverages the perpetual inventory approach which consists of inferring, among other quantities, the current stock levels based on the business operations that are also recorded within the system. The system is typically given the initial stock level, and all further stock levels are computed automatically.

This approach typically uses the notion of SKU that represents a physical location that is supposed to contain a certain number of units of the same product. The typical perpetual inventory viewpoint typically assumes that units within an SKU cannot be differentiated.

The perpetual inventory approach is also predominantly coupled with FIFO (first-in first-out) inventory valuation, which is also computed based on the same historical records associated with business operations as the ones used for the perpetual inventory calculation.

Periodic counting as a corrective mechanism

Perpetual inventory relies on the assumption that the business operations records accurately reflect what is really happening as far as real inventory is concerned. However, manual entry errors introduce discrepancies between the quantities that are really flowing in and out, and their actual electronic counterparts. In particular, due to the cumulative nature of perpetual inventory calculation, errors accumulate over time.

Hence, even if the business operations are recorded following a process that ensures a very low percentage of inaccurate entries, every inventory rotation is going to degrade the accuracy of the indicators maintained through perpetual inventory, most notably the stock-on-hand. This issue is typically referred to as a phantom inventory problem where the electronic stock levels diverge from the real stock levels.

The phantom inventory problem is typically addressed by periodic recounting of the stock levels. In practice, warehouses and other tightly controlled environments need only infrequent recounts. In contrast, environments where there is little control of inventory processes, such as stores, require much more frequent recounts.