Reviewing forecast accuracy – putting the ‘A’ in FP&A 

Your forecast is the best estimate of where you’re business activities are going to take you.

It helps guide decision making and planning and when used properly can prove invaluable. So why does talk of forecasting sometimes cause a shudder to run through your finance team? 

Too often, forecasts are put in place based solely on numbers or deliverables set by the board. At the end of the period in question, the targets may or may not have been achieved but there is little reflection on the process itself as the numbers were inflicted in a top-down manner, overwriting any input from individual departments. Managers are frequently tasked with targets such as achieving the same net profit margin across all operating units, but in reality the return on sales fluctuates between operations and requires a more targeted approach. This broad brushtroke approach is frequently the result of a delinquent planning system, often based entirely on spreadsheets. 

When a bottom-up approach is used, with involvement across the organisation, the result tends to be a desire for more accuracy. Reviewing your forecasting processes is ultimately one of those tasks recognised as desirable but rarely completed. Particularly when the focus is always on completing the next forecast, there is little time to reflect on how the plan is actually put together.  

When reviewing your forecast there are two methodologies: the pseudo-scientific approach of merely explaining what has just happened, and the more genuinely scientific approach of carrying out the value-added activity of analysing the results and using them to help make proactive decisions about what steps to take next. The latter is fundamental for a culture of continuous improvement with a focus on becoming better planners. 

It is important to recognise that any forecast, unless completely reverse engineered, is never going to turn out to have been totally accurate. The challenge is to achieve a tolerable margin of error which will then endorse the assumptions on which you are basing your decisions. 

Just because you expect error in your forecast, doesn’t mean you should just accept it as a given and move on without any interrogation. The discrepancies between your forecast and results provide valuable data and you should take the opportunity to review them and gain from the insight they provide. Sensitivities may have changed around particular key drivers, warranting further attention.  

If your team is overstretched and lacking in time then this analysis quickly falls by the wayside as a ‘nice to have’ rather than an essential FP&A activity. Tools like Planning Analytics free up time for this value-added activity to take priority. Your team can actually put the ‘A’ into FP&A, start digging into your results and establish how the model itself can be improved for future planning.  

If certain key drivers need looking at, the sandboxing capability within Planning Analytics allows your analysts to run alternative scenarios and test assumptions while leaving the existing model untouched until they have a clearer picture of how best to update it.  

The ideal forecasting process should look a lot like this:  

At a time when everything has changed, freeing up your team’s time to allow them to pursue the analysis to reveal the fundamental what and why of the developments will help you power profitable decision making. To take the first steps on your journey to a more agile method of working, get in touch today.