Sales forecasts are a FREAKING BIG DEAL inside most companies. Every executive from the CEO down needs to know how the company is going to perform in order to set investor expectations and plan day-to-day operations. And since sales forecasts are so important, it's good to know that most companies have a HIGHLY SCIENTIFIC method for creating them. It generally looks something like this:
- STEP #1: The request goes out for an accurate forecast. Sales manager through the company call together their sales reps and ask for an accurate prediction of what they'll sell.
- Step #2: The sales reps make a conservative guess. The sales reps guess-timate what might sell, and then clips off 10%, so that if a deal falls through, the rep can still make the number.
- Step #3: The sales managers adjust the guesswork. Because they know that the reps are padding, the sales manager adjust the guesswork. reflecting what they hope will really happen.
- Step #4: The sales VP re-adjusts the forecast. Because the sales VP knows that the sales managers are adjustingthe numbers, he bumps the number down a bit, just in case.
- Step #5: The marketing VP does his own forecast. Because the marketing VP doesn't trust the sales VP, he makes his own forecast, based on historical sales figures.
- Step #6: The head of manufacturing does his own forecast. The CMO knows the process is BS, and so so he makes his own forecast of what to build, based upon what he prays will be needed.
- Step #7: The CEO makes up a brand new forecast. The CEO tells the investors that the company will make big numbers, then tells the sales VP to change his forecasts to match.
Fortunately, there is a better way, but it requires the following three fundamental changes in the way your sales team does business:
- CHANGE #1: Track customer milestones not sales activity. The source of most errors in forecasting is a sales process that tracks the activities that sales reps do, rather than the actions that customers take. This "vendor-centric" sale process encourages reps and managers alike to think of the forecast as something that reflects the outcome of sales activities, rather than predicting the behavior of the customer base. For example, when predicting how many sales are going to close within the next month, it's not useful to know that your reps made 57 in-person sales calls. What's useful is knowing how many customers made a "verbal acceptance" and the average percentage of such customers that actually sign a contract.
- CHANGE #2: Decouple forecasting from the sales quota system. As long as sales managers try to use forecasting as a motivation tool, it will remain inaccurate. There's nothing wrong with having sales quotas, but the time to talk about those quotas and compensation isn't when you're trying to come up with a meaningful projections for investors and the rest of the company. The easiest way to do this is to revisit quotas on a monthly basis so that they're not discussed in the same context as the forecast. This takes some mental discipline on the part of the sales manager, who must resist the temptation to conflate the two activities. But they need to be separate in order to keep the game-playing and politics out of the forecast.
- CHANGE #3: Reward, rather than penalize, forecasting accuracy. In many companies, top management beats up any sales manager who walks in with a forecast that's not what top management would prefer it to be. While there's no question that top management should consider replacing a sales manager and sales team that can't consistently deliver what's needed to make the company successful, that's not a decision that should be made during a forecasting exercise. The forecasting process needs to reflect reality, not wishful thinking, and shooting the messenger is a bad way to get a picture of reality. So top management has to take a "just the facts" attitude when dealing with forecasts and consider issues about sales strategy and personnel deployment separately.