Data is everywhere nowadays, and thanks to the combination of cloud computing, data science and enterprising companies, data analysis drives an increasing part of our lives. Yet human judgment continues to dominate B2B sales forecasting.
For example, the two most common forecasting methods (the weighted pipeline and “forecast categories”) are 100% judgmental. And most sales managers and sales operations managers would argue that forecasting is “more art than science.”
It is true that, unlike marketing, sales doesn’t naturally generate the thousands of data points required by statistical forecasting methods, for most SMEs at least. Does that imply that sales reps and managers are condemned to guesswork when discussing potential revenue?
Forecasters have traditionally responded to the shortage of company-generated data by turning the problem on its head. If your company’s unit sales are too low, derive your sales forecast from market data:
This is called demand forecasting. It usually comes in more sophisticated flavors than the basic model above. Here are some of the most common refinements:
Describing these refinements makes the two main drawbacks of demand forecasting for SMEs obvious:
Yet until 10-15 years ago, that was all professional forecasters had to offer, which explains why so many companies are stuck to judgmental methods. Thankfully for B2B sales managers and sales operations managers, the state of the forecasting art has changed.
Behavioral economists have shown that errors in human judgments are not random, but follow patterns that can be studied and corrected. Simple cognitive tricks will often improve forecast accuracy significantly.
Behavioral economists have also demonstrated that, judgmental forecasts have intrinsic value and, combined with quantitative forecasts, improve overall forecast accuracy. In a financial portfolio, each asset comes with its own combination of risk and return, but at portfolio level individual risks cancel each other out (to some extent) and the risk/return sum is greater than its parts. The same diversification mechanism is at work with combinations of judgmental and quantitative forecasting methods.
Big data is non-conventional internal and external data sources. Thanks to our connected world, the data sets that are available for analysis and forecasting are now much larger than traditional company or market sales data. On the Salesforce AppExchange, you will find applications offering to refine your forecasts using:
Contrary to traditional statistics, this approach to data analysis doesn’t assume that data is following an underlying law. It takes data patterns as they are and explores their implications through intensive data crunching.
Algorithmic modeling is on the rise for three main reasons:
It is well adapted to environments where data is relatively scarce and unstructured, just like B2B sales, which is why we at SalesClic have chosen to leverage its potential.
Learn more sales tips in our free ebook at the button below.