We -- and I'm speaking universally here -- love to talk about big-data . It's great and glorious in its transformative capabilities, we all agree. But hold on a second, now, let's not get ahead of ourselves.
If you're talking about the technology backing the big-data revolution -- the high-performance computing and advanced analytics software, for example -- then, OK, yes, talk away. But you've got to stop a minute on the organizational challenges that come with big-data, Andrew McAfee, author and principal research scientist at MIT, told attendees at his keynote address at the SAS Global Forum Executive Conference taking place in San Francisco this week.
"I'm not confident that a lot of companies are going to be able to make the switch in running themselves effectively in this era of big-data," McAfee said.
His reasoning: The HiPPOs aren't going quietly.
"The what?" you're probably wondering, as I was.
In McAfee lingo, a HiPPO is the "highest-paid person's opinion." Even companies that consider themselves data-driven businesses typically operate as such today: They gather relevant data, do their analysis, then turn to the HiPPO in the room and have that person make the decision. HiPPOs rely on "what's in between their ears," their guts, or their instincts. They've gotten there because of their track records, their resumes... because they can command the room, McAfee said.
Now, enter the pencil-necked geeks with their analytics software, said McAfee (with the greatest of affection). The real big-data challenge is in letting the geeks -- the data -- really drive the decision making, even if it doesn't correspond with the HiPPO's intuition.
This is a common theme we hear, but McAfee warns that the hostility from HiPPOs is not to be underestimated. Resistance can be nasty, he said, citing as one example the now-famous Nate Silver and his spot-on accuracy in using poll data to predict the 2012 presidential election. In advance of election night, traditional election pundits -- i.e., the HiPPOs -- minced no words on how utterly wrong they thought Silver and his numbers would be. Whoops.
It's a bit confounding that we listen to HiPPOs at all any more, but alas, they lurk everywhere. "When we've got data, and build decent models, you'll always get a better judgment than if you rely on a HiPPO," McAfee said.
But McAfee says we shouldn't wish HiPPOs extinct. We should consider them in new roles. These are the experts who know the organization's biggest challenges and know what questions to ask. He quoted Voltaire: "Judge a man by his questions, not by his answers."
In addition, the geeks should be using HiPPO predictions, judgments, and forecasts in their algorithm, he suggested. This is the polar opposite of what happens today at most organizations. Geeks, who aren't known for their subtlety, he noted, need to play more nicely. They've got to say, "OK HiPPO, tell us what you think will happen, and we'll build that into our models."
Only then, McAfee suggested, will organizations truly move themselves to a data-dominated decision making style.
Gut vs. the data. Which side of this divide does your organization stand? Share below.