Forecasts are predictions about the future, and thus are destined to be wrong. The real challenge is to make them as accurate as possible. That means having a forecasting process, and knowledgeable people who are in touch with customers and the marketplace, to feed the best, most up-to-date information into that forecasting process. The final ingredient is to have sources of market information that “inform” the forecasters' conclusions.
Statistical estimates are good, but educated, informed, statistical forecasts are even better.
There are several things that can be forecast with reasonable accuracy:
- How fast can suppliers and producers respond—and with how much inventory—to an unexpected surge in demand?
- How quickly and gracefully can they cut back if the forecast was too optimistic?
- How long does it take to staff up and train a service organization for a surge in demand?
A good sourcing/production/operations organization should know the answers to those questions, and thus be able to forecast “response-ability” quite accurately.
Ultimately, there are only three proven solutions to providing “good service” (timely delivery for in-stock positions, with controlled inventory levels) in the face of volatile demand, when even the best of forecasts are wrong. Here they are. Use them individually or in combination, and give yourself and your company the best chance of success.
1. More lead-time to adjust sourcing and production.
2. More inventory because inventory can buffer a mistake, especially an oversold forecast.
3. More, flexible, productive capacity, with low (or no) changeover/set-up times. This is the best of all; hard to attain, tough to maintain, but the best, most responsive solution of all. This is the capacity to ramp up products/services as demand increases, or to cut it back when demand falls short.
Obviously this last option is the best, if it is affordable, if it is possible, and if the transport time is not too great. However, sometimes the toughest job is slowing down an overheated supply chain when the forecasts fall short. Once goods are "launched" from faraway places like Asia, they are like ICBMs; they arrive here a few weeks later, whether you want the "payload" or not, and then you drown in unwanted inventory that ties up money and warehouse space.
The bottom line: stop searching for a magic solutions—there aren’t any. But these next three steps will help you combine the above advice to compensate for faulty forecasts.
- First, put in place the three ingredients: A good forecasting process/system; knowledgeable people who collaborate with customers; and robust market research to help confirm or deny what seem to be the best forecasts.
- Second, it is imperative to forecast your ability to respond to volatility in supply, or goods, or people, and be ready to serve or satisfy customer demand.
- Third, and perhaps most critical of all, start thinking about what combination of the three operational solutions—inventory, lead time or flexible capacity—will best fit the needs of your customers, and help you outperform your competitors. These are what compensate for forecast errors when the best ones are still not good enough
And always remember, through it all—“The purpose of a business is to create and keep a customer!”1—and to keep the customer happy. That means doing all you possibly can to achieve that goal.
John L. Mariotti is President and CEO of The Enterprise Group. He was President of Huffy Bicycles, Group President of Rubbermaid Office Products Group, and now serves as a Director on several corporate boards. He has written eight business books. His electronic newsletter THE ENTERPRISE is published weekly. His website is Mariotti.net.
1 I’ve seen this quote attributed to Peter Drucker but I first read it in Theodore Levitt’s classic book, The Marketing Imagination.