How to execute annual financial planning and forecasting in today’s environment

Organizations should review their current enterprise performance management (EPM) infrastructure and associated capabilities to make sure they have the building blocks in place to maximize the value that Finance can deliver through enhanced business partnership. Using a framework like the one below can help contextualize EPM maturity and target initiatives to move up the curve and deliver greater value.

Click image to enlarge

Leading-practice organizations often begin the planning process by setting strategic guidance, translating the guidance to targets, and disseminating those targets to business units. In practice, the critical challenge often comes during that latter stage, when more aspirational goals of leadership come face-to-face with more realistic expectations of performance held with the business.

To counter this, organizations can use updated forecast results to aid strategic reviews and target-setting activities. Provided that steps are taken to remove bias from recurring forecast processes, longer-horizon forecasts can provide valuable insight into expected performance to inform strategic planning and may also be used to seed annual plans and provide a baseline for target-setting activities.

Many organizations conduct planning by having business units submit a select number of targeted scenarios (for example, high, medium, and low) before identifying the most likely scenario and finalizing a single set of financial results. Rather than trying to arrive at a single number, organizations should consider identifying a range of possible outcomes. Leading-practice organizations will use observed volatility to bolster ranges with assigned probabilities, allowing leadership to plan for a wide variety of outcomes while considering the likelihood of occurrence.

Additionally, factoring in multiple modeling approaches across statistical, driver-based, optimization, and trigger-based contingency models can help an organization be prepared to understand and project impacts across multiple levels of performance outcomes. Machine-enabled modeling capabilities can evaluate thousands of possibilities in real time from a broad array of possible driver values and assumptions, removing the practical barriers that once limited finance to modeling only a select number of key scenarios. Rather than high, medium, and low, a more comprehensive, probabilistic distribution of expected outcomes allows leaders and decision-makers to assess the likelihood of any number of scenarios.

Leading-practice organizations use driver-based logic to tie financial outcomes more closely to underlying economic and organizational drivers. In times of increased volatility and divergence from historical precedent, driver-based plans have an inherent advantage over their simpler, trend-based counterparts. While trending may have been sufficient in the past, historical results may no longer be a reliable indicator of future performance in today’s volatile economy.

Organizations should look to expand driver logic, as appropriate, throughout the planning process and rely on input from those in the organization who are closest to each respective driver when establishing scenarios and plan ranges. (Note: This does not mean organizations should try to make every plan item driver-based. Materiality should always be considered; if a line item makes up less than 2% of the total plan, even a 200% actual-to-plan variance may not move the needle.)

An increased reliance on external economic drivers should also be considered during periods when activity is heavily influenced by market forces or economic shocks. For example, setting GDP as an external driver, and given an observed 10% decline in GDP, a company can expect the top line to fall within a specific range. As GDP shifts throughout the year, forecasts can be adjusted based on actuals to validate the predicted ranges. Now more than ever, plan and forecast results will be as dependent on external economic drivers as internal ones.

Additional approaches to consider can include exploring more logarithmic or exponential trending curves, as well as approaches that bias a moving or weighted average for recency. Modeling based on historic and comparable events can also prove valuable in these instances, even if the macroeconomic event itself may seem unprecedented. Machine-enabled modeling can also offer the capability to automate the process of discovery and ongoing driver evaluation with the ability to update, adjust, and continuously improve performance as new information becomes available.

Many organizations are reconsidering the areas and depths to which they plan and forecast financial results. They are also thinking about the timing of their plan and forecast cycles (for example, delaying plan kick-off to incorporate more recent actuals and updated forecast data). Given this context, there is an opportunity for organizations to evaluate what activities truly add value to streamline and refocus efforts traditionally spent on the iterative planning process to realize a more sustainable, agile approach.

Acceleration of digital capabilities can also play a significant role in easing the burden of target-setting, scenario analysis, iteration, and any required resetting of the baseline, particularly when technologies that can help ingest new data and drivers, and score their relevance automatically, are deployed as part of the planning and forecasting capabilities.