According to a SiriusDecisions survey, b-to-b sales reps spend an average of 2.5 hours each week preparing forecasts. However, only 21 percent of companies achieve 90 percent or greater forecast accuracy on a 30-day forecast horizon.
The three pillars of forecast accuracy are tools, process and human behavior. We’ve spent billions solving the first two, but until the advent of predictive analytics, we’d barely scratched the surface of the third. Human behavior – especially a proclivity toward self-preservation and self-deception – causes us to act in ways (often subconsciously) that make forecast accuracy difficult to attain. Predictive analytics tools, if used correctly, could be the instruments we need to address the following counterproductive forecast behavioral patterns:
Self preservation. People deliberately misrepresent sales forecasts to protect themselves. I offer your own experience as proof: As a sales rep or sales leader, have you ever committed to a number you couldn’t support in order to avoid an unpleasant conversation during a forecast call? Why would you take the risk of posting an indefensible commit? It’s a calculated risk – pay now or maybe pay later. The hope is that, over time, the situation will improve and the number will somehow be achieved.
Social psychologist Dan Gilbert explains the reasoning behind the decision to misrepresent the sales forecast, referencing an equation developed by 18th-century mathematician Daniel Bernoulli: (odds of gain) x (value of gain) = expected value. When applied to sales forecasting, it looks like this: (manager’s reaction) x (employee standing) = self-esteem. Your manager’s reaction (odds of gain) is how they’ll respond to the forecast information you convey (with praise or with criticism). Value of gain is how you think your manager will view your performance after the forecast conversation (i.e., positively or negatively). Multiply those variables to calculate expected value, which is your self-esteem quantified.
For example, if you inflate your forecast to meet your manager’s expectations, your manager’s reaction might be scored a 10. Your employee standing might be scored a 7, resulting in a self-esteem score of 70. Alternatively, you could communicate a reality-based forecast that scores a -10 on the managerial reaction scale. Your employee standing might be scored a 5, giving you a self-esteem score of -50.
Faced with a choice between a self-esteem score of 70 or a score of -50, people naturally choose the first option, thus ensuring that forecasts will be inaccurate. Gilbert also points to a human propensity to choose immediate gratification (or pain avoidance) over the uncertain delayed gratification of a future event. We also tend to overestimate the value of immediate gratification because while we do have the ability to estimate future outcomes given present decisions, we’re innately bad at it.
Predictive analytics can help with this problem because they suggest a forecast based on reality, not hope, and will be the catalyst for an honest conversation, in the present, between employee and manager. As sales leaders, will you allow honesty? Can you handle the truth?
Self-deception. The second problem with forecasting is – as psychologist Cortney Warren has explained – we are “masters of self deception” and we “lie to ourselves because we don’t have enough psychological strength to admit the truth then deal with the consequences that will follow.”
So, not only do we fool ourselves into forecasting unrealistic deals and numbers, we also continue to lie to ourselves even when the truth is evident. Warren has posited that the best way to deal with this lying is to understand the techniques we use to pull the wool over our own eyes. The techniques, Warren’s definitions, and how they apply to sales forecasting include:
Predictive analytics is to sales forecasting honesty what wearable fitness trackers are to fitness honesty. Fitbit, the manufacturer of one popular tracker, boasts that its users take 43 percent more steps when using the device. Seeing the data makes users confront the truth: they either walked or they did not. There’s no self-deception.
Predictive analytics, in much the same way, demonstrates to the rep the steps they’ve taken in developing their sales pipeline and how it will manifest in the future. The deals are both present and mature (as judged by data science not intuition), or they are not. A coaching culture of free expression, honesty and action will allow predictive analytics to work.
With the opportunity to finally solve the forecast accuracy problem, the question is: can we handle the truth? Or maybe the better question is: how will we handle the truth? Sales organizations should keep the following in mind:
Accept that forecast accuracy requires more than tools and process. Understand and address the human element. Use predictive analytics as a guide to finding the best opportunities rather than as lie detectors to intimidate and coerce. Finally, choose a predictive analytics vendor that statistically proves the accuracy of their data model, both initially and over time.
For more on the psychology behind the human tendencies that lead to inaccurate sales forecasts, check out the TED Talks from Dan Gilbert and Cortney Warren on the subjects of self-preservation and self-deception.