Restock Alerts Versus Waitlists : How to Manage Transient Excess Demand

We analyze two different mechanisms for allocating a scarce product over a finite horizon when demand exceeds supply: timed restock notifications and waitlisting. The first mechanism gives all customers an equal chance to acquire the product by visiting the retailer at a pre-announced date and time. The second mechanism allocates the product to customers on a first-come, first-served basis. We characterize the total lost sales under each mechanism and compare their relative performance with respect to this metric. There are no performance differences in the steady state; on the transient path to steady state, however, we find significant differences between the two mechanisms. Whereas the restock mechanism is highly sensitive to search cost yet relatively insensitive to how quickly customer utility decays over time, the opposite holds for the waitlist mechanism: it is highly sensitive to the rate of decay but relatively insensitive to search cost. For a retailer grappling with product shortages, the best mechanism to use depends on three factors: how inconvenient or costly the mechanism is for the consumer, how long the customer is willing to wait for the product before her valuation drops to zero (i.e., rate of decay), and how large the demand is relative to the supply. If consumers' inconvenience costs are high then the retailer may prefer a waitlist; if the rate of decay is high, then restock alerts are likely to perform better. Provided that search cost is low, the restock mechanism is almost always preferable when demand far outweighs supply