After many years of managing diverse supply chains for different companies and industries, as well as experience as a business consultant, I am convinced of the need to have and use an accurate sales forecast. Regardless of the supply chain infrastructure set up behind our operation, and this could be a multimillion-dollar system, hardware and resource investment, if we fail to get the right mix of inventory transiting through our supply chain, we might not get the expected impact on our key metrics and bottom line, and the expected ROI would not be met on time.
Why is it important to look forward and distinguish our operations between a short-term, one- to three-month operation versus a business plan based on a four- to 12-month forecast range depending on product lead times or product development roadmaps?
Peter Drucker, considered the greatest business management philosopher of the 20th century said: “Long-range planning does not deal with future decisions but with the future of present decisions.”
Once, a manager told me, referring to the old Mexican saying, “If a child is drowning in a well, my priority will always be to get him out right away.” I then thought to myself that if you had foreseen the danger and closed this well months ago when you found out it was open, there would now be no one to save from drowning.
As a consultant, I have found many organizations of more than 100 people devoting most of their efforts and resources to what is currently happening, meaning they are short-sighted in focusing on how they are going to close the month, maybe the quarter. They are being reactive; rarely is there a group of people or at least one single person dedicated to foreseeing what is going to happen in four to six months, or a year from now, creating a strategic vision. And with the current situation on global supply chains, not looking at mid- to long-term scenarios makes it an impossible mission, or the difference between being reactive instead of being strategic. Planning on mid- to long-term forecasts, and introducing an S&OP process could help organizations to foresee and prevent many supply chain hassles and risks.
First we create forecasts, then we should measure their accuracy, and as we start understanding what are the relevant and most important variables that we need to consider for generating these forecasts, then we should start having a positive impact on our inventory mix. However, what does this mean? You know you have improved your inventory mix when you start reducing overstock and out of stock, which should be related to an improved forecast.
An accurate forecast eventually will help me get a healthy product mix in inventory and prevent having higher working capital invested in inventory, increasing the availability of products and their rotation. In other words, increasing sales and reducing the cash cycle.
This will also have an impact on the reduction of the use of obsolescence funds that will impact the bottom line. If these forecasts are shared collaboratively with suppliers and customers, it will facilitate an improved fill rate for suppliers, improve our credibility and the fidelity of our customers, and avoid expediting and, therefore, higher logistics costs.
Demand planning will help you better manage the product life cycle by predicting when the current product will reach the end of its life and when exactly the new product introduction will take place. This will ensure that you will always have enough product already in place across different channels to launch, to make a smooth product transition and maximize the marketing and promotional funds invested when introducing a new product.
With adequate business planning, availability can be ensured by defining the right coverage of your cycle inventory (the one that rotates) and planning your security stocks (the more variance on sales and lead times, the more security stock you will need).
However, you do not have to do it all yourself. Using AI and machine learning platforms to forecast sales and demand will help you automate processes and focus on other key aspects, such as improving relationships and closeness with customers and suppliers.
AI and machine learning models can analyze large amounts of historical data. This data can be used to identify patterns and trends in demand, which can help predict future demand more accurately, which no human or conventional statistical models can do properly.
With accurate forecasting based on data and artificial intelligence, increased sales can be achieved by increasing the availability of products, increasing service levels and helping to build customer loyalty, while reducing out-of-stock and excess inventory.
As a consultant, when I suggest to some managers that they should explore an AI solution that can help them solve these many challenges through generating a more accurate forecast and they say this is not their priority right now, I wonder what their priority might be. Inevitably, I think about the child and the well. Lol.