“I think this is the perfect career move,” Nicolas Vandeput, founder of SupChains, says of incorporating data science skills as a Demand Planner.

Like so many industries, the way we’ve done things in the past in demand planning may not be the way we do things going forward. The line between data scientists and demand planners is blurring.

There will be a time in the future when answers become a commodity, and questions are the premium. Adopting a scientific curiosity and understanding the right questions to ask will become a valuable skill.

The effect of data science on demand planning and supply chain planning can’t be underestimated. Demand Planners can no longer rely on how we’ve always done things.

Does that mean, however, that data scientists can be effective demand planners?

We spoke with author of Data Science for Supply Chain Forecast and founder of SupChains, Nicolas Vandeput, who laid out what data scientists bring to demand planning and how demand planners can leverage data science to evolve in this rapidly changing field.

“I think this is the perfect career move,” Vandeput says of incorporating data science into your demand planning repertoire, on an episode of IBF’s On Demand podcast.

We know data science is not demand planning. There are significant differences in these fields. But there can be valuable synergy between them, as well.

Vandeput unpacked exactly what Data Scientists and Demand Planners can learn from each other, how a data scientist can apply their expertise to this field, and how a Demand Planner can become the supply chain Data Scientist in your company.

Adopting A Probability Mindset In Demand Planning

Demand Planners live in a world of ambiguity. We’re never “right”; we just hope to be close in our forecasts. This lack of precision can be a challenge for Data Scientists, who specialize in answers.

Is that ambiguity a problem that data scientists need to overcome?

Vandeput doesn’t think so. “In data science, you try to be as accurate as you can, but you totally accept being 99% [accurate],” he says. So, data scientists accept some ambiguity, too — but they typically have a clearer understanding of just how much there is.

He says this is because of the “science” part of data science. A scientific mindset is all about experimentation, observation, and curiosity. Data Scientists, then, must be able to test new ideas, accept failures, and move on to the next idea.

The real difference that a data scientist brings to the table is a probability mindset. While Demand Planners are comfortable with ambiguity, data scientists can accept that ambiguity, but also consider the probability of accurate forecasts with a given model and adjust their models to achieve higher probability of accuracy.

Adopting a probability mindset, rather than simply accepting traditional levels of ambiguity, could help demand planners achieve more accurate forecasts.

The Project vs. Process Workflow

We might see Data Scientists and Demand Planners as complementary but distinct roles that require different skill sets. In that case, a breakdown of roles would look something like this:

  • Data Scientist: Work on a project basis. Focus on developing a forecasting model based on data you receive, and hand it off to the demand planner.
  • Demand Planner: Work in an ongoing process. Apply the model to manage assumptions and stakeholder needs. 

As a Demand Planner, you manage a process, perhaps weekly or monthly, with ongoing adaptation to new “inputs” or pieces of information. 

Data Scientists, on the other hand, have a project to work on. They develop a model and hand it off to a Demand Planner once it’s working. They may have to adjust the model or develop new models, but that won’t require such an ongoing process.

“I do really think that if you have a deep expertise in your market or your business as a Demand Planner, you really know your data,” Vandeput points out. “You know what client is important, seasonality, or which promotion is important.”

That knowledge and skill establishes who could be a good data scientist for a specific program, Vandeput says. But to become a broader asset in your company, you must consider some of these shifts toward a data science mindset.

Applying a Data Science Background to Demand Planning

We’re seeing a lot of people coming out of college with expertise in data science who weren’t necessarily thinking about demand planning as a career. But the jobs are finding them. As companies become more aware of the benefits of machine learning and AI, they’re more interested in putting people with a background in data science in demand planning roles.

People with an academic background in data science can move into a demand planning role thanks to their advanced analysis skills. But to successfully make the transition, it’s imperative to broaden that skill set to understand the language of sales and supply chain management.

Here are Vandeput’s tips to make the transition:

Talk to people. A data scientist can develop a good model, but that model can’t see everything. Talk to clients or production facilities, for example, to tap into that human intelligence about the supply chain.

“As Data Scientists, you’ll be able to bring a really good model,” he says. “But from there as a human, you need to bring some kind of extra layer of intelligence.”

Stay curious. Keep that scientific mindset, that curiosity, and incorporate collaboration. Incorporate new inputs — what you learn from conversations with people in your supply chain — to develop stronger models.

“As data scientists,” Vandeput says, “it’s really clear that if you ask different people with different mindsets, the input, you’re going to end up with a better number.”

What Demand Planners Can Learn From Data Scientists

Data Scientists aren’t the only ones who need to adapt to better serve the demand planning process, though. As data science becomes an increasingly important part of the supply chain, Demand Planners can look for opportunities to start challenging the way we’ve always done things.

That scientific mindset, the curiosity, the ability to admit when you’re wrong and to look objectively at your forecasts and assumptions are skills we need as Demand Planners. 

If you’re in a Demand Planner role, consider how you can expand your skills to become what Vandeput calls the “supply chain data scientist” at your company. You might think about the following:

  • Add coding skills: You don’t have to become an expert at R or Python, but a basic understanding of coding can help you utilize the resources at your disposal, such as packages of code you can copy and paste to develop new models.
  • Incorporate external data: If you’re only looking at internal sales, you’re not seeing the full picture. Incorporate additional inputs to create more accurate forecasts and avoid repeating mistakes.

Combine these empirical skills with your ability to communicate, collaborate, and orchestrate those ongoing processes puts you in a position to meet the changing needs of demand planning in the future.

This is based on an episode of IBF’s On Demand podcast, a leading show for demand planners and business forecasters about the latest trends and future of demand planning, forecasting, predictive analytics, and S&OP.