Supply chain or demand forecasting can be a pain even in the most stable industries. And forecasting for the highly seasonal demand profile can be hard, but not impossible.

In the previous post, we talked about forecasting demand and gave tips to help find a more efficient way of planning for demand volatility.

We already mentioned the benefits of an accurate picture of demand, and its vital for enterprise‐level planning for financials, merchandising, assortments and so on. Your forecasting and planning must be accurate to determine what demand is right now and where it will likely be in the near future.

Another important aspect of demand planning, as well as, strategy and operations planning is that from a demand perspective we are building a plan of what sales and marketing will deliver on. So, we need to understand the potential issues and relay the subsequent information to other parties.

The reason why we need more accurate forecasts is, so we gain a better understanding of the demand for our products, as well as improved demand planning and forecasting will provide better inputs to strategy and operations planning processes as we move forward.

The problem begins when one starts thinking about forecasting accuracy of the volatile demand environment. However, you may be able to create a very accurate forecast from a volatile demand distribution, but it is not as easy to do so, as volatile demand environments aren’t always predictable.

In order to create more accurate forecasts, one needs to understand the variability, volatility rates and their relevance to forecasting.

Variability

Variability is a difference between what we expect and what actually happens. It is the statistical distribution of outcomes one can expect from a process. This can be based on your sales, demand or any of the other processes.

In recent years, this statistical distribution has widened, which means that managers now have to anticipate and plan for a widening array of demand and supply scenarios.

Volatility

Demand volatility is an expression of how much the demand changes over time and the predictability of the demand. This can have a significant impact on your business, whether it is any political or seasonal volatility or just changes in customer behavior.

Although volatility is not necessarily bad – changes in demand can also mean growth causes volatility, so do seasonal and cyclical swings. There could be other reasons – promotions, bonus packs, price actions, or any other management options.

Forecasting Accuracy

Forecasting accuracy is an expression of how well one can predict the actual demand – volatile or not. We need the accurate forecast to keep you ready to handle demand volatility to the maximum possible level, so if you are faced with an unfavorable market environment you can easily adapt and change your strategy and operations.

There are a couple of actions to help with volatile demand forecasting:

  1. Collect further market intelligence

Wherever possible gather as much information about your market. We suggest collaborating with other strategist and ops to increase the accuracy of your demand.

Additionally, look at some different ways of market data gathering, we wrote a blog post about threatcasting, which is a great technique to use for further your market research.

  1. Introduce more forecast modelling techniques

Whether using training test sets or cross-validation, it is up to you. The crucial part of a process here is that you can validate your data in some other way or minimise the variance in the forecast accuracy.

Additionally, we suggest introducing Monte Carlo analysis into this process. So, this might not a be a traditional way of increasing forecasting accuracy, however, it has one of the best rates of statistical probability in terms of forecasting accuracy, and it might be easier to do this with a piece of software, see what we can do for you with Monte Carlo integration.

  1. Introduce corporate management to reduce demand volatility

Corporate management can try to stay away from some less valuable business practices and hence cause a reduction in demand volatility. Or they can promote purchasing throughout the year. But can we temper seasonality and growth?  Do we want to?

Words of wisdom

An accurate forecast does not reduce demand volatility, but it will add to your ability to address this volatility.

At the end of the day, supply chains and sales teams will never operate optimally with or without a forecast. Just because your forecast doesn’t match customer demand fulfillment and execution with military precision doesn’t mean it was a waste of time or that you were wrong. As long as your data produced an acceptable plan that had timely execution and enough leverage for the time being – you have succeeded! Now try again with more precisions, it will get better!

And don’t forget if you liked this article share it and subscribe to our newsletter!

Subscribe to our Newsletter to learn more about financial planning, forecasting, and risk assessment!