Best practices in Rolling Forecasts

Everyone wants a crystal ball to be able to peer into the future. For businesses, that desire becomes a necessity because having a vision of the future allows for better and more strategic decision-making in the present.

A rolling forecast simply means that each quarter or month, a company projects four to six quarters or twelve to eighteen months ahead.

This allows executives and key decision makers to see both a financial and operational vision of the future. It also helps them assess next steps in their execution of their plan, understand critical pivot points in the plan and better judge the impact the economy may have on their plan.

I have seen rolling forecasts replace annual planning cycles with a continual planning process that results in more regular business reviews that look to the future. These reviews enable managers to understand problems, challenges and trends sooner and improve their proactive approach to those problems, challenges and trends.

You need a system and Excel is not a system (it’s a personal productivity tool). In the next nine best practices there will be an outline of functionality and practices that will outstrip the capability of Excel spreadsheets. The challenge I have seen arises when companies start rolling forecasts in Excel and then can’t keep them up to date because it isn’t just a single forecast. Once a baseline forecast is created, additional analysis/versions will be required to:

• Understand the impact of major and minor alterations to the plan
• Run the plan against different drivers
• Copy the plan as a baseline for other plans
• Perform variance and sensitivity analysis

To get the most out of rolling forecasts you need a system that can provide the necessary functionality to accommodate your budget, forecasting, planning, strategic management and measurement needs. Excel will be too labor intensive, prone to error, and too difficult to provide the reporting.

Understand your objectives of creating a rolling forecast. This will drive all of the other best practices related to rolling forecasts. While the overall objective of planning is to a) create a financial view of the vision for the future and to know the decisions you need to make and b) understand the financial impact of those decisions before you need to make them, your objectives and focus for the rolling forecast will dictate the areas of the plan that need more detail.

As an example:

• If cash is tight you will want to focus on the timing of events around cash i.e. payables and receivables policies, cash borrowings, line of credit.
• If your strategic plan calls for acquisitions you may consider building the plan with a high-level summary input of drivers that help plan for acquisitions.
• If you want to better manage finished goods inventory you may focus more on customer/product forecasts and inventory management.

The challenge is, there isn’t one overriding holistic model format that allows you to account for all of these in a first pass rolling forecast implementation.

Identify the rolling forecast duration. The challenge with rolling forecast duration is it sounds easy until you put pen to paper and begin to implement it. There are a number of questions:

• Will you re-forecast every month or every quarter? (many companies re-forecast monthly)
• Will you add a new month to every forecast or just add a new quarter at quarter end? (companies either add new periods quarterly or start planning for a two-year time frame)
• How long should the rolling forecast be – 12 months, 15 months, 18 months – or should you start by forecasting out two years and at the end of the first year tack on another year?
• How will the mechanics work of rolling in actual?

As a starting point, make sure your forecasting time frames are consistent with your business cycle and business needs. If sales 15 months from now are dependent on capital expenditures today, it is important to create rolling forecasts out past 15 months. It is also important to have a minimum of quarterly bias to the rolling forecast.

That is, don’t tack on an additional month to a rolling forecast at the end of every month, wait until quarter end and then add a new quarter, otherwise each month you are making preparers think about a new forecast period. I saw the companies that try the approach of monthly re-forecast where each month, one month of actual falls off and another month adds to the back end and there was considerable pushback by the forecast preparers.

The more accuracy is applied to the current year and close in months, it gives management on “how what we change this year will affect next year”.

The second scenario is adding an additional quarter at the end of each quarter and the plan may have 15 – 18 months rolling forecast data.  This has the benefit of potentially better precision because the later months aren’t as far out so more attention can be given to them and as quarters are added on it requires the planners to apply more rigor and accuracy to the plan.

Identify the rolling forecast comparison periods. While this sounds relatively straightforward, it gets tricky when you consider the need for comparison columns in reporting. It is tricky because you need to compare to multiple periods/combinations of actual and forecast. With annual plans, we compare the combination current year actual for closed periods and forecast for remaining periods to last year actual. We think in terms of annual sales and annualized expenses. With rolling forecasts we need to also roll in the actual and shift the comparison periods. We must be able to not only provide annual comparisons, (i.e. How will this year compare to last year?, Where will we end up for this year?) but we also need to provide how this 12/15/18 month rolling plan compares to the 12/15/18 actual results. This requires access to roll the forecast periods but also roll the comparison periods, for not just actual, but also for prior year actual.

Understand/analyze the dynamics of revenue and expense in your business and their related drivers. Rolling forecasts won’t work if you create them from a bottom-up forecast every quarter – they need to be driver based. This provides flexibility in the planning process and agility when you re-plan or create alternate plans. It also helps managers focus on what is really important. Over time, some of the drivers will be replaced if they are found as an inconsistent predictor of results and the stronger drivers will evolve to KPIs and used for goal setting. In addition, over time forecasts will also start to use consistent models and model drivers.

As an example, sales units should drive production volumes. Price increases could drive (or be driven by) material and labor cost increases.

To help understand what drives your business you can ask these questions:

• How does this line item affect our bottom line?
• Will $1 (pick an appropriate unit of measure – thousands, millions, billions) in sales affect this account?
• How does this number change our bottom line?

To be successful doing rolling forecasts it is important to use drivers to eliminate detail while creating a realistic expectation about the future.

Plan capital and strategic projects separately from the rolling forecast.

Capital and strategic projects should be layered separately into the plan.

These projects have two unique characteristics that require:

• That they are time independent of a typical rolling forecast. Projects can last months to years and treating them with the same duration as a forecast doesn’t provide the complete story around the projects.
• That they typically have a lot of dependencies which means project expenditures can be moved out or in a plan for a variety of reasons.

Examples:

- Sales units drive production volumes

- Price increases could drive (or be driven by) material and labor cost increases

Projects should be treated as separate entity/profit centers and have the ability to be the move in and out of rolling forecasts. In addition, they should have the ability to move time frames and adjust for changes in project scope.

Start with a small select group of key department/operations managers, plan on increasing the scope over time and plan on continual improvement over time.

While conventional wisdom is to involve as many participants as possible, companies have found it easier to start with a small group that cuts across the major departments and gradually involve more executives, lines of business managers and operational managers as processes/methods are refined. Let the small group get some practice with the process. It will help flush out the process and the more practice they have at preparing forecasts, the better they will become and the more mentoring they can provide as you add additional participants. Typically there is a risk (including career risk) if you roll something out that isn’t totally baked or based on theory, which could result in confusion, especially if you are doing it with a large group within the company.

As you expand the scope of the plan, and the more people share in the plan, the closer to reality the plan will be and the better the execution.

In addition, you should plan for and build in a process of continual improvement in the process. Managers should learn from their forecast accuracy record by carrying out post-mortems on the forecast. The objective is not to punish the guilty but to better understand how to do better, what changed, what were we surprised about, and what we should do differently to turn out better forecasts. To improve forecast, it is important to understand the cause for variability and learn to reduce them through post-mortems.

Consider the rolling forecast as your baseline plan.

Once you have the completed rolling forecast for the period/month, use it as your baseline plan. From this plan Finance can massage drivers, adjust values, and analyze alternate scenarios, black swan events and potential significant events, to create a picture of alternate futures and be prepared to make decisions if situations arise.

This does require the flexibility in the tool you are using to “mash up” scenarios and compare scenarios from different sources.

Tie your rolling forecast to your strategic plan.

Rolling forecasts, if well-prepared, form the backbone of a new and much more useful information system that connects all the pieces of the organization and gives senior management a continuous picture of both the current position and the short-term outlook.

Analyse/understand how external conditions impact your performance.

Business does not operate in a vacuum. Plan on qualifying the key external drivers of your business and do not confuse rolling forecasts with targets you receive your bonus on. Regarding targets and profitability, is it better to hit your target or gain targeted market share? Leading organizations are placing forecasting at the center of the management process. It becomes the essential tool for business managers to support their decision making, not just another management chore that needs to be done and they are basing targets not on fixed revenue numbers but rather percentages based on external market indicators.

Adaptive organizations focus less on annual budget targets or long-term views/targets, generated from within the company based on incremental sales and income year over year. Instead, they focus more on rolling views and goals based on market conditions, like achieving market share or achieving cost efficiencies over the competition by benchmarking against competitors or similar public companies.

Be prepared.

As you consider implementing rolling forecasts, remember the purpose is to provide a vision for the future and support better decisions. If you try to motivate behavior based on this forecast, the results and plans get skewed and the usefulness of the solution is compromised. Forecasts should not be used by executive management as a tool for questioning or reassessing performance targets. This means forecasts and targets must be independent if you want to obtain both relevant action plans and reliable forecasts that allow risks and opportunities to be identified and corrective action taken in the best interest of the company (which may be in conflict with the target). If you do merge the two, the term “sandbagging” comes to mind.

Conclusion

Many businesses have yet to discover the full benefits of evolving their planning process to include complete and accurate rolling forecasts. With so many external factors that affect the bottom line for these businesses, including the volatility of the economy, creating rolling forecasts is a sound way to ensure strategic decisions are made. Businesses need an accurate picture of a future in order to chart a course towards it.