Forecasting Demand Despite Shortages

Forecasting future demand despite historical shortages is a massive challenge for demand planners. This article presents three techniques to estimate unconstrained demand during shortages: fill in missing values based on historical figures, use your forecasting tool, or use historical demand drivers. Finally, we will see how forecasting demand using more granular time buckets might help with data quality.

Nicolas Vandeput
6 min readJan 17



Shortages cripple supply chains around the globe. They harm service levels, frustrate your clients, and benefit your competitors while pressuring your teams to find quick (and often expensive) solutions. Moreover, they pose a massive long-lasting challenge — -even after you could replenish your inventory! — -How can you forecast future demand for a product that was recently out of stock?

Figure 1 Sales pattern interrupted by frequent shortages (data: online retailer).

Demand Forecasting or Sales Forecasting?

Before discussing how we should forecast demand despite shortages, let’s begin by highlighting the difference between demand forecasting and sales forecasting.

The objective of demand forecasts (and demand planners in general) is to deliver valuable pieces of information to supply chain decision-makers so they can make the best possible decisions (based on expected market demand, costs, and supply).

Figure 2 Picture a demand planner as a sailor on a boat with a spyglass. They bring information about what’s on the horizon to their comrades on the ship — leaving them to decide what’s best to do.

A demand forecast is an unconstrained unbiased view of the future (expected) market demand. This forecast should not include any supply constraint (such as your company’s ability to deliver on that demand). These supply constraints will depend on the decisions taken by S&OP and supply planners (based on the information you gave them about future expected demand). We cannot expect demand planners to know these…



Nicolas Vandeput

Consultant, Trainer, Author. I reduce forecast error by 30% 📈 and inventory levels by 20% 📦. Contact me: