Subscribe to the Supply Chain Planning Blog

Keep up with the latest trends, research, and insights about supply chain planning, demand forecasting and inventory optimization.

← BLOG

Your Demand Forecast Is Wrong. Here’s How to Fix It.

By Matthew Kippen • 28 Jul 2021

If you were still using a computer or a telephone from the 1960s today, you’d get some strange looks, to say the least.

But many supply chain practitioners don’t realize that the most common approach to supply chain planning–using a demand-driven forecast as the primary input to future planning–is just as passé.

According to Gartner, 90% of supply chains will still be demand-driven by 2026.

As detailed in the report, demand-driven forecasts have been proven to be “rigid and fragile,” unable to respond to uncertainty, with an almost single-minded focus on the accuracy of forecasts.

As I see it, this is a singularly frightening prospect for one simple reason:

Your demand forecast is wrong.

This is why Gartner recommends building forecasts driven not by demand, but by uncertainty.

In other words: Your forecast should plan for how wrong you could be, rather than aiming for an unattainable accuracy target.

 

Forecast Accuracy vs. Uncertainty

Uncertainty-driven supply chain decisions assume that you will not be accurate on every forecast you planned.

The trick is to assess every choice by focusing on key performance indicators (KPIs) that tell you how well your supply chain can tolerate uncertainty, then analyzing probabilities of success in the resulting supply chain decisions.

The end result is that your supply chain becomes more resilient to changes in the market, both known and unforeseen.

Here are some recommendations to help move from forecast-driven to uncertainty-driven planning:

  • Realize that your forecast is going to be wrong. And if your forecast is wrong, then anything you use that forecast to make decisions on will also be wrong.
  • Recognize and prioritize supply chain uncertainty and variability, and how this can be used to improve planning and decision making.
  • Switch from deterministic to probabilistic planning. This allows you to focus on ranges of outcomes, allowing your demand planning to be tested for how well it can respond to uncertainty.
  • Develop your KPIs around this resiliency to uncertainty (ex: time to survive, time to recover, probability of execution).
  • Establish your digital supply chain twin.
Your forecast should plan for how wrong you could be, rather than aiming for an unattainable accuracy target.

 

Stop interpreting demand-driven as forecast-driven by changing the focus of your planning onto uncertainty rather than exclusively on the accuracy of your plans,” says Gartner analyst Tim Payne.

The Shifting View on Forecast Accuracy

This is a recent study that Gartner conducted, asking companies how important forecast accuracy was in achieving supply chain goals:

An interesting view here is how things have slowly started to shift, with a stronger focus on optimization rather than having the “most accurate” forecast.

Many companies have seen the repercussions that stringent demand-driven forecasts have caused across their supply chains and are looking for alternatives that deliver positive results to their businesses while still focusing on real, tangible ways to improve the day-to-day across their supply chain.

Especially at ToolsGroup, we’ve seen how probabilistic forecasting has changed the game–helping companies be resilient during times of crisis, serve their customers more effectively, and ultimately provide the highest returns to their shareholders.

According to Tim Payne, “Probabilistic planning (where the probability for different outcomes is calculated), would allow companies to set up their plans to be resilient to uncertainty so that the plans are more likely to be achievable and, thereby, deliver on the goals.”

Watch video: What Is Probabilistic Forecasting?

Especially with how world-shifting events such as the coronavirus have shown damning flaws in demand-driven forecasting, knowing how to account for and use uncertainty to a demand planner’s advantage has never been more important.

After all, it’s not just your forecast that’s wrong. Everyone’s forecast is wrong.

 

GARTNER is a registered trademark and service mark of Gartner, Inc. and/or its affiliates in the U.S. and internationally and is used herein with permission. All rights reserved.

This graphic was published by Gartner, Inc. as part of a larger research document and should be evaluated in the context of the entire document.

Six Tips for Success Using Machine Learning for Demand Planning

Subscribe to the Supply Chain Planning Blog

Keep up with the latest trends, research, and insights about supply chain planning, demand forecasting and inventory optimization.

Supply Chain Brief