What is a Rolling Forecast?

The first question we almost always get asked by our clients is, “What exactly is a rolling forecast?”

Most people and companies have some form of forecasting in place already to update their full year projections. “Isn’t a rolling forecast just a matter of adding a month or two or three at the end of the budget and lopping off the same number of months from the beginning?” Technically, yes, in its most primordial form, that is all it is. However, since the only tool available to most companies that would enable this is the budget, simply shifting the company’s timelines would result in organizations developing, updating and maintaining detailed budgets on a continuous basis throughout the year. Most would agree that this is not sustainable (if not downright frightening!) and the result is that it will fall by the wayside. The problem is that without taking into consideration all of the other valid and useful purposes the budget serves, this method is not likely to be sustainable. What is missed is the organization’s opportunity to become more dynamic and nimble, one that uses its forecasting processes as a true competitive advantage to achieve superior results.

We’ve seen several definitions, but we believe a rolling forecast is:

  • A process that continually adapts to changing business conditions
  • Designed to enable business decisions to be made by focusing on key drivers and levers
  • Comprised of a fixed number of forecast periods determined by how far into the future a firm can reasonably see
  • A rapid process that is completed in a day or less

We believe that a rolling forecast is NOT:

  • The budget plus x months
  • A mathematical exercise
  • Completed at the chart of accounts level
  • Focused on the calendar or fiscal year

A rolling forecast typically has a fixed number of forecast periods. The number of periods is dependent on several factors including the industry, business cycle, product life cycle, etc. Regardless of the length of periods, it will always follow the pattern below.

A Rolling Forecast

A=Actual data F=Forecast

Contrast this view with what typical forecasts look like.

A Typical Forecast

A=Actual data F=Forecast

A typical forecast that only updates the budget is focused on achieving the goals for that fiscal year. The total number of periods evaluated is constant but the number of forecast periods is reduced by one period each time. The focus is typically on executing the plans that were made in the budget, regardless if the underlying assumptions are still valid. Instead of looking beyond the year, it stops suddenly in December as if business will not be conducted in January. In short, it follows an arbitrary cycle that does not align with the natural course of business. Thus, what could have been a valuable exercise becomes nothing more than a variance explanation and results in missed opportunities.

The second question we get from our clients is, “If we adopt a rolling forecast, should we kill the budget?” The short answer is no; the long answer will come in our next entry where we demonstrate how a rolling forecast and the budget work together to enable the true execution of corporate strategy.

2 thoughts on “What is a Rolling Forecast?

    • Thank you for the comment on our blog. The term “8+4 Forecasting” is another way to describe forecasting for the same period covered by the budget. It denotes that there are 8 months of actual data and 4 months of forecasted data. Organizations who use this method typically have a new forecast every month for the same time horizon and name them 8+4, 9+3 (9 months of actuals, 3 months of forecast) 10+2 (10 months of actual, 2 months of forecast), etc. This method holds the overall time period constant and simply adds one more month of actual data and one less month of forecast data every month. This is also known as “Forecasting to the wall” as the end point is typically the end of the fiscal year and the forecasts do not predict periods beyond the year end. Best practice is not to follow this process but instead use a rolling forecast as described in the blog post.

Leave a Reply

Your email address will not be published. Required fields are marked *


You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>