How to eat an elephant: The recipe for starting with machine learning in demand forecasting

Are you ready to steer your business based on more precise forecasts but don’t know where to start? Creating a value-adding machine learning model for demand forecasting can be challenging. In this blog, we talk about the hurdles companies might face when implementing driver-based forecasting and how to overcome them.

how to eat an elephant: how to get started with machine learning in demand forecastingNow, let’s start with some facts: From a benchmark that we conducted in 2021 we know that while most companies are eager to start applying machine learning in the area of demand forecasting, only 10% are actually doing it. We further inquired what is stopping them: It turns out that tooling, knowledge, and having a solid business case are the biggest obstacles to moving forward.


Machine learning in demand forecasting: slice the elephant

As a first step you should ask yourself: “What will machine learning bring to my business?” Preparing a proper business case is crucial.

Luckily, that is something where EyeOn can be of help. We can build a proof-of-concept in a very short amount of time. Our data scientists build and apply machine learning models in demand forecasting in a very pragmatic and robust way. The proof-of-concept offers you:

  • Quick insight into the value and relevance of your demand drivers on the forecastability of your business
  • The maximum forecast performance that can be reached by using driver-based forecasting
  • The quantified improvement potential of driver-based forecasting versus your current forecasting method (in efficiency and effectiveness), including a detailed benchmark of the driver-based forecast performance versus your current forecast performance

What you need for a proof-of-concept

Proof-of-concepts that we have executed in the past, show that a limited data set that is related to your most important demand drivers can already bring much value. Next to the most obvious data needed, such as historic demand, product, and customer master data, it is critical to have your demand driver data available in a structured way for the same historic period. In case you do not have this data structured yet, our advice would be to start as soon as possible. Finally, in order to provide you with a detailed benchmark, we would need at least one year of your forecast snapshots in order to give you a complete view of the potential of driver-based forecasting for your business.


Bon appétit

The proof-of-concept shows great potential and you would like to grasp the benefits immediately? Our Planning Services team is ready to build and operate your driver-based forecasting process from the start. In the meantime we can work on getting your tooling and knowledge to the required level, so we can transfer the operation back to your internal organization.


Optimize your forecast performance

Do you want to improve your forecast performance? At EyeOnwe’ve developed the Fast Forecast Scan: a quick tool that provides you with rapid insights into the demand characteristics and forecastability of your business. As a first step, we perform a thorough deep dive into your data and provide actionable data quality insights. With improved data quality, the Fast Forecast Scan provides you, within a few days, with data-backed insights on the highest possible forecast accuracy that can be reached and identifies the main opportunities for improvement. Learn how the Fast Scan can help you reveal your forecast optimization potential here.

Or contact us directly to see how driver-based forecasting can be of added value to your organization!

Search for