Corporate forecasting can be improved by avoiding the pitfalls and applying some practical lessons developed over the years by practicioners.
1. Concerning revenue, don't focus on what you want or think you need, focus on what is realistically within reach. Forecasts, especially when not done at the usually more conservative Finance/Controlling department, often exhibit the famous earnings hockey stick. This is espcially true if the department building them has very little day to day involvement in the numbers and key KPIs. Turnover for a particular product or service is usually comparably easy to model well, when taking into account past growth rates and variables that affect them (e.g. number of stores selling the product). Forming bottom up KPIs for every turnover category, then aggregating those streams, is certainly more robust than a wild guess.
2. On the operating expense, don't starve your company's sales, marketing, and admin. It's easy to forecast short term cost since salaries and rents don't generally change much overnight. But if entire departments are restructured, we often forget to plan for contingencies - new systems very seldom work as designed from the first day. Employees who know they are let go will not be motivated to work much any more. Companies often need additional capacity to cover for activity increases, even though managers are often not willing to acknowledge this. Also, forcing e.g. travel cost savings onto your sales staff without planning for that to have secondary effects on turnover is a frequent mistake.
3. Political influence by any of your stakeholders interested in a certain budget should be managed well. On one hand, management usually requires ambitious turnover goals and asks for those year after year. Sales will then realize that with goalposts that are obviously hard to achieve, their chance at pocketing their bonus is dwindling. Hopefully, these two interests cancel each other out. But don't count on it! Try to be diplomatic and steer the business plan towards a financially realistic and sustainable path.
4. Watch all KPI on which bonuses depend. You are biased when forecasting those. Adjust for your bias.
5. Spend the right amount of time and thought on KPIs and ratios that are actually consequential and those that are not. It is hard to figure out which are the inconsequential KPIs, figures that are actually not drivers of any useful result. It is psychologically hard to let them go after that much analysis and after having used them for so long. But once you have figured out the variables that are comparably important in driving outcomes, that is where most of your thought should go. Budgets are often put together on tight deadlines, so saving on time by dropping some of the complexity is sound strategy.
6. Market comps are underused. Most of the data we base our forecasts on are internal company data. Remember, you're not alone in your market. Your competitors will also have plan data available on their own financial reports and reasons on how to meet targets. Using available data from competitors is a major opportunity often missed.
7. Forecasting is often only the privilege of the few, of managers who may or may not have been trained in doing what they do, and who may or may not have sufficient time to formulate a reasoned forecast. Intel and Hewlett-Packard, the Department of Defense and many others are now starting to use forecasting markets to harness the wisdom of the crowd inside their companies, not only for financials but also e.g. for demand, investments, and predictive maintenance. They simply realized that finding, scoring and rewarding good forecasters within their ranks works.
8. Ask yourself whether your forecasters are sufficiently qualified and trained. Do you know what a base rate is? Or how to optimize calibration? Is awareness training concerning own biases regularly refreshed? Are your forecasters actively open minded and encouraged to pursue approaches that may be new? Are they numerate enough?
9. Use advanced analytics – but don't lose human judgment. Today, many companies sell fancy analytics and data visualization packages to make all kinds of corporate predictions. They are good, but only as good as the humans in front of the computer. In the hands of good analysts, they can be a powerful tool to explore previously unknown patterns, though you should be careful about not getting lost on minutiae (see 5.).
10. Think one step further, often also called thinking outside the box – hiring and vetting good forecasters who have this ability into management will optimize your forecasts. Experienced forecasters will be more adept at identifying indirect effects, unidentified revenue drivers, or synergies. Hire one of our consultants for your board!