April 2007 - Issue #14
Forecasts are always wrong so why bother?
By definition, a forecast is always wrong, so why bother? The main reason is because your customers are key to your business, and your forecasts/ demand plans represent your customer demand. And, if you provide excellent service to your customers in the most efficient and consistent manner, you will likely be one of the rare businesses that can demonstrate sustainable and profitable growth.
So, since forecasting / demand planning is a critical success factor, what are the best practices that can be applied to any industry/ company? I've seen many examples of forecasting process, tools etc., and yet the example that stands out and achieved the best results is based on a pragmatic approach that could be applied to any industry or company. There were three critical components to the successful example.
First, people are #1. Although this sounds like a cliché, it has proven to be the most critical aspect to achieving results. In my experience, I've seen several systems, processes, statistical formulas only a Stanford graduate could fully understand, etc; however, regardless of the process or tool, the person who achieved results was an undervalued (by higher management) forecasting guru. Believing in simplicity, I used observation to determine the "magic formula" for how she achieved outstanding results AND was respected by the customers and sales team (which is no small feat!). And, I discovered that her secret to success was communication (with customers, Sales etc). I know it sounds too easy to be true, but it "worked" consistently.
Second, process is key. My forecasting guru's secret to success was developing a repeatable process - and continually looking for opportunities for improvement. In essence, the base process looked like the following: develop a system-generated forecast (based upon history, adjusted with reasonable underlying assumptions), review the key exceptions (for example, those forecasts with high forecast error), pick up the phone to discuss with the appropriate parties (communication!), adjust the forecast as makes sense, monitor progress/ metrics. It doesn't sound like it requires a rocket scientist to master; however, I've seen only a rare, few examples where this process was consistently applied. It isn't necessarily exciting or new, and there are no "magic tools". Instead, this process requires persistence, follow-up and communication skills.
Third, tools are important. With that said, I'd put my forecasting guru up against anyone so long as she had a piece of paper (or excel spreadsheet) and a telephone. The key is to talk with your customers and sales team. Provide data, ask about trends, question assumptions, and push back. There is nothing more important. With that said, once you have the right person in the right position with the right process and the right tools, additional forecasting tools will allow them to achieve the next level of results. There are several tools in the market that can improve your forecasting and automate pieces of the process so that your forecasting/ demand planning resource can focus on the keys to success. In my experience, I've found Prescient's software to be one of the best, since it provides the greater amount of critical functionality for the best value.
There is no downside in prioritizing additional time to forecasting - at the minimum; you'll have a better understanding of your customer's needs.
Back to Top
Lisa's Tips: Cycle counting
The main goal of cycle counting is to cycle through your inventory to ensure inventory accuracy on a consistent basis. (In contrast, a physical inventory is a point-in-time process, typically yearly, to ensure inventory accuracy). The keys are as follows:
- People: It is important to make sure the right resources have the right responsibilities. A counter requires a detail-orientation and should take counting seriously (it is amazing how much time and energy is saved with a cycle counter that understands the value of counting right the first time), whereas a reconciler requires an analytic ability with enough of a big picture understanding to make sense of the results and research.
- Cycle count frequency - cycle counting must be performed on a consistent enough basis to provide assurance that the inventory is continually reviewed (and can be relied upon). Best practice processes typically count weekly - at minimum, monthly.
- Cycle count calendar - the key is to make sure that you've cycled through your warehouse at least once a year and have covered 100% of your warehouse. Depending on your level of accuracy and impact on customer service, a more frequent basis might be required.
- Cut-off processes: you might as well throw out all your work if everything moves while its being counted. It is critical to identify a process that works for your business where you freeze the inventory for a period of time and/or coordinate closely on all moving parts and cut-off's.
- Variance analysis: there is no reason to count if the majority of the emphasis isn't on understanding your variance root causes and developing solutions. Don't get bogged down in resolving every dollar - focus on the significant ones, and you'll achieve the "80/20". This is the key to an effective cycle counting program.
Back to Top
Recommended Reading
"John W. Gardner on Leadership" by John Gardner - although it isn't written in an ideal style for my reading preferences, it is an insightful work on leadership principles. Gardner had an expansive career in academia, the U.S. Marine Corps, in government positions and serving on commissions of several Presidents, and as a director of many corporations. With this type of diverse background, he is in a unique position to expand thinking in terms of leadership.
Back to Top
|
