“We start our planning with the forecast.”
This is what I’ve heard in the last three (3) organizations I’ve engaged with.
These three (3) organizations often started their Sales & Operations Planning (S&OP) meetings with a comparison of forecast versus actual sales data. In most cases, the actual sales data didn’t come out close to the forecast.
In one organization, when actual sales didn’t match the forecast, the vice-president chairing the meeting would scold the sales managers. She would insist that sales managers get their forecasts right the next time. And in every next meeting, the forecasts would just be as far off from actual, and the vice-president would just rant angrily again.
Supply chain planners frequently insist on accurate demand forecasts so that they can plan production and purchasing schedules that would match actual sales. But unless the organization has only one (1) customer and only one (1) product item to sell, forecasts never would come close to actual demand. Planners would end up changing production plans and vendors would complain about the changes they need to make in their supply schedules.
Just like why we need forecasts to get an idea of what tomorrow’s weather will be, we need demand forecasts to get an idea of how much of an item will be sold in the foreseeable future. Forecasts help us anticipate outcomes and plan appropriately.
But the real value of demand forecasting is that it allows our enterprises to be more in touch with our customers. The process of forecasting begins with gathering data about what our customers intend to buy over time. And more than that, what our customers feedback would give an indication of how they view our products and services.
Academics teach forecasting via scientific formulae and statistical tools. These tools rely on historical data and extrapolate possible outcomes through mathematical equations. With advances in data science and artificial intelligence that can monitor and predict individual behaviours, organizations are developing the capability to predict buying patterns more accurately.
Forecasting, however, is still just as much an art as it is a science. There is still a margin of uncertainty with every forecast that doesn’t make it easy for supply chain planners to come out with schedules that match sales by 100%.
If our enterprises can’t rely on forecasts to accurately predict demand outcomes, then why do forecasting?
As mentioned above, forecasting starts with getting in touch with customers and it is in this activity that forecasting provides the highest value to an organization.
By talking to our customers about how much they plan to buy, we become familiar with our customers’ points of view about the business. We become familiar how our customers see our products and we in turn see their outlooks. We get to know if they are optimistic or pessimistic about the business.
In other words, forecasting isn’t about just sales numbers on a spreadsheet that supposedly tells how much would be sold in so many months. Forecasting is about assessing how interested our customers are in continuing to do business with us.
When we seek a weather forecast, most of us don’t really ask how many inches of rain there will be tomorrow; we ask whether it’s going to be sunny or rainy. In some respects, demand forecasting is the same; as much as we value how much volume we expect to sell, we really want to know if customers like our products and will buy more.
When we understand our customers’ wants and needs, how they view our products, and their plans in terms of not only how much they will buy but also how much more or less they will buy, then we would have realized the value of forecasting.
Ellery's Essays 