Many organizations are finding better ways to plan and control by eliminating budgets and replacing them with a series of more effective processes. This article examines one aspect of the movement—rolling forecasts.
Some organizations spend up to six months or more in the annual budgeting process. The process is based on assumptions that often make the numbers out-of-date shortly after the budget is approved. Yet many companies march to that plan, even though it is based on assumptions developed in the middle of the previous summer.
Start by asking where the finance organization is positioned in your company. Unfortunately, most finance organizations are positioned to take a historical view. The problem is that finance has limited strategic potential from this position. To become strategic, finance must look forward. That is a key reason why forecasting is a more effective strategic tool than budgeting.
The keys to a successful forecast are that it should be: Timely—in good time to take action.
To be timely, you must consider how fast your organization can take action. These actions include adjusting existing plans, taking additional actions or implementing major new initiatives. The need to be able to forecast—to see into the future—is driven by how quickly we can react to what lies ahead.
Organizations can address these needs by developing longer forecast horizons or by becoming more adaptive and agile.
The need for timeliness also applies to individual plan elements as well as the whole. The approach taken by Southwest Airlines illustrates that organizations benefit by focusing on the plan elements that are critical to success and that are more volatile. This focus allows management to concentrate on critical factors rather than updating all forecast elements with the same frequency.Actionable—information about things we can change.
For forecasts to have maximum value, they should be designed to focus management on the actions that should be taken. For instance, if you have a driver-based model that indicates expected volumes are different from those previously expected, the forecast model should indicate how management can respond. This likely would be to add or reduce headcounts. An actionable model includes likely step level actions such as adjusting overtime, adjusting the temporary workforce or adding additional shifts.Reliable—unbiased, with an acceptable level of variation.
To be confident in using a tool, managers need to have confidence that it is reliable. In terms of forecasting this translates into a process that is free from bias. If managers are worried about bias they will hedge the results. With everyone applying their own subjective judgment, it becomes impossible to assess forecast accuracy or to evaluate attempts to improve forecasting.
One common problem is organizations that constantly exceed their forecast. “For over 21 years we always beat our budgets and forecasts. The only thing we learned was that we were really good at sandbagging,” said Paul Hensley, CFO of HOLT CAT.
In 2009, HOLT CAT eliminated budgets in response to the financial downturn. In 2010, the company decided to permanently eliminate budgets. If an organization always exceeds its forecasts, it has a consistent pessimistic bias. Rather than being a positive attribute of good management, this really means the organization has no idea of its true capability.Aligned—pointing in the same direction.
In many organizations, confusion is the norm. The sales leadership has one direction in mind. The operation team does not believe in the sales objective, so it heads a different way. Finance doubts both and, therefore, plots for a different point. As a result, employees are confused and shift their efforts based on who they spoke to last.
An organization that is aligned focuses on a single path. It will often have scenarios to identify events that could cause deviations. But everyone will begin discussion from a single aligned forecast. In well-controlled situations, that path and the normal variation around it are tracked. Consistent trends and anything outside the expected range are investigated to see if the course should be changed, but all are involved in those discussions. Cost effective—right the first time; no rework or waste.
Developing an effective forecast process is a lot like developing an effective golf swing. Learning begins with a repeatable process. With each iteration, the process is evaluated and improved.
Just like a manufacturing process, forecasting begins with inputs (assumptions), moves through conversion (the application of techniques or judgments) and results in outputs (a forecast). And just like a manufacturing process, if the quality of the input is poor or the conversion process is poorly designed and controlled, the finished product will be substandard. Steve Player is program director, Beyond Budgeting Round Table, North America.
Read an expanded version of this article here.