What's the one thing holding back your company from embracing analytics and becoming more data-driven? Hint: It has nothing to do with the large, water-loving mammal with which it shares a name.
Data is the voice of the customer, and you've got to bring that right into the heart of the organization. In the end, that voice has to take primacy over what the CEO thinks.
former CEO, Tesco
Speaking at the recent SAS Premier Business Leadership Series, Andrew McAfee, author and principal research scientist at MIT, said the "highest-paid person's opinions," or HiPPOs, often stand like an immovable behemoth and block the path to effective data-driven decision making.
He described a situation common to many businesses that have embraced analytics. In those companies, data teams often conduct analyses in an effort to find an optimal strategy or course of action. But then the resident HiPPO, which may be the organization's president or a member of the C-suite, considers the data alongside other factors, such as their own gut instinct or knowledge. This approach minimizes the impact of data and prevents organizations from letting data direct the decision-making process from start to finish.
Many executives who may fall into the HiPPO category feel there is value in this process, according to McAfee. After all, they typically boast many years of experience and great depth of expertise. But, McAfee said, evidence actually shows that letting so-called experts make decisions from their gut delivers much worse results than embracing data-driven decision making. The success of Nate Silver, formerly of The New York Times, in using statistical analyses to predict the state-by-state outcomes of the 2012 elections, was one example McAfee cited.
McAfee pointed to a study conducted at the University of Minnesota entitled "Clinical Versus Mechanical Predictions: A Meta-Analysis," which showed that predictions of human behavior are significantly more accurate when based on a data model than when based on an expert opinion. According to McAfee, the evidence proves that businesses need to reorient their decision-making processes to make them less dependent on resident experts.
"This tells me that the data-driven approach is clearly the right approach," he said, "but it's going to be difficult because the HiPPOs are not going to go quietly."
Highly paid executives should still be part of the decision-making process, because data can't do everything, McAfee said. It still takes a person to decide what questions should be asked of the data.
For example, data-driven decision making can tell a retailer who its best customers are and what types of advertising will increase their purchases. But it can't tell the business what its growth targets should be or where it should rank against its competitors. These are the decisions executives should handle. This means executive opinions should be introduced at the beginning of the decision-making process, when questions are more general, rather than the end, McAfee said.
More about the benefits of data-drive decisions
See how data-driven decision making can unleash productivity
Learn the mistakes executives make when they fail to embrace data
Read about one company's data-driven marketing gains
Conference attendee Terry Leahy, former CEO of the British retail chain Tesco, said businesses are most successful when they listen to their customers. There is no more direct way to pay attention to what customers are saying than by using their data to guide marketing campaigns, determine what products to carry and influence decisions.
"Data is the voice of the customer, and you've got to bring that right into the heart of the organization," Leahy said. "In the end, that voice has to take primacy over what the CEO thinks."
But Leahy acknowledged that it will be difficult for organizations to make this a more prominent part of the corporate culture. Many CEOs do not currently allow data to make decisions for them, and they may be resistant to giving it a larger role.
To help reluctant executives see the light of data-driven decisions, Leahy recommends that analysts explain how the data actually tells their customers' stories when presenting findings to top executives. He said analysts often present findings that are heavy on statistics, which either bore or confuse executives, making it even harder for them to recognize the value of data-driven decision making. But executives will understand the stories about customer lifestyles and preferences, which the data will illuminate, Leahy said.