How to Fix Sales Forecasting with Revenue Performance Management

The eWEEK Knowledge Center recently published an article by contributor Joe Ruck on sales forecasting. While I agree with his diagnosis that forecasting is rife with challenges, I disagree with his prescription. Mr. Ruck's article treats forecasting as a means to answer the question "What will top-line revenue be?"

Instead, the true value of forecasting lies in creating an actionable revenue forecast that provides the entire organization with detailed information to drive execution. You may even call this comprehensive process that goes beyond sales forecasting "revenue performance management."

It is estimated that companies miss the equivalent of 10 percent of total annual sales in lost opportunity revenue that could have been captured as a result of better insight on sales activities and target markets. Here's an example: Company XYZ has a top-level forecast that predicts a price decay for a particular product line in the Asia Pacific region. The executives feel fortunate to have seen this shift coming and lower prices to match the forecasted competition. The business swallows the lost revenue, but survives because it doesn’t completely lose the market.

Suppose, in reality, that the price decay only existed for a specific product in Korea, rather than for the full product line in the entire region. The company lost a great deal of revenue by slashing their prices across the board. How can you prevent this from happening to you?

If it's broken, fix it

As Mr. Ruck described in his article, customer relationship management (CRM) systems are inadequate for creating a detailed, bottoms-up, forward-looking plan. CRM is great at providing a top-line gut check on revenue for new opportunities, but is grossly insufficient for:

1. Forecasting run-rate business

2. Product or account-based forecasting

3. Showing detailed changes on unit and price forecast

Instead, companies sometimes turn to statistical modeling or top-down predictions to fill the void. Using algorithms to churn out predictions has its uses, but no previous time period serves as an accurate model for the unpredictability of today's markets. On the other hand, predicting revenue with educated guesses or by multiplying probability and overall pipeline is goal-setting, not actionable revenue forecasting. For example, what does a plant produce based on 3X pipeline coverage of revenue?