The recent US presidential election puts one more nail in the coffin of managing by gut feel. Weeks prior to Election Day, the data indicated a victory for the sitting President.
Business leaders still on the fence about whether to invest seriously in analytics technology need only look at the U.S. presidential election last month to see the need to jump quickly and deploy state-of-the-art analytics tools. Although I am not an American citizen, like many outside the U.S. borders, I watched the election with some fascination, not because I favored one candidate or another, but because it seemed to be a contest between political operatives who believe in data and those who preferred “gut feelings.”
High-profile supporters of Gov. Mitt Romney insisted well into election night that the Republican candidate would not just defeat President Barack Obama, but crush him decisively in both the Electoral College (a unique and baffling aspect to American presidential elections to those of us outside the country) and the popular vote.
They were particularly incensed that analytics expert Nate Silver, writing the FiveThirtyEightblog on the New York Times Web site, predicted through modeling poll data that the President would be reelected and with ease.
Some skeptics claimed the data used by Silver was full of bias, though he was completely open about the public information he used as well as how his model worked. So his critics created other Web sites and massaged the data so it reflected their desired outcome. Some Romney supporters spent hours on television decrying the data-driven conclusions of the FiveThirtyEight blogger by claiming the large and enthusiastic crowds that their candidate was attracting on the campaign trail proved he was surging with momentum. One writer in the Wall Street Journal claimed she knew Obama would lose because when she was traveling in Florida, an important state in the Electoral College, all she saw were “Romney” signs in people’s lawns.
They were all wrong and the data geeks got it right. The reason Romney’s supporters were so wrong is that they were too invested in their personal belief in their candidate to objectively look at the data, which for weeks prior to Election Day indicated a victory for the sitting President. They were evaluating their candidate’s likelihood of success based on their interpretation of non-scientific information that supported their preference.
A CEO should pay careful attention to this recent election. Are you surrounded by other executives who devalue data in decision making? Does your head of sales insist that the annual golf junket for clients is worth every penny when the data is inconclusive or contradictory? Is your chief marketing officer more influenced by a creative designer’s whim for an advertisement than by a business analyst’s data showing the ad’s adverse affect on viewers? Does your operations chief insist on persisting with established manufacturing processes “because we’ve always done it this way” instead of paying attention to data that indicates more efficient processes? Are board members considering an acquisition swayed by eloquent and persuasive colleagues at the table or by due-diligence data on the table?
This American election should put one more nail in the coffin of managing by instinct, gut feel or, as some prefer, experience. Certainly, experience is vital. It’s what helps you ask the right questions of the data. But when you ask a question and the answer you get is not what you want, don’t blame the data. Believe it. It’s the difference between winning and losing.
Photo Source: Jim Bourg/Reuters