kentoh - Fotolia
Among enterprise users of tools for predictive analytics, the general consensus is that data beats gut feelings for driving business decision-making almost all the time. But in the political realm, that calculation isn't so straightforward.
One big reason Donald Trump gained the upper hand on Hillary Clinton in the 2016 presidential election is that Clinton's campaign based key strategic and tactical decisions largely on data analytics, while Trump appealed to voters on a more visceral level -- seemingly without a concerted analytics effort to direct his campaigning, at least initially.
"When you adopt an analytical approach, you're looking at things as they are today, but what emotion can do is change that by changing people's behavior," said Pradeep Mutalik, an associate research scientist at the Yale Center for Medical Informatics and an election blogger for Quanta Magazine.
Mutalik said Trump created a so-called reality distortion field, a term originally coined to describe how Apple co-founder Steve Jobs was able to shape technology development plans and schedules to his ideals and convince others that the seemingly impossible could be done. Whether you agree with Trump or not, his personality radiates passion, just as Jobs' did, Mutalik added. That clearly had an effect in energizing his supporters.
The divide between the approaches of the two campaigns wasn't entirely neat and clean. Early on, there was plenty of talk about how unsophisticated the Trump campaign was when it came to using data; Trump himself called big data "overrated." Later, though, details emerged that painted a somewhat different picture. Trump's campaign eventually invested heavily in an analytics operation that used tools for predictive analytics to guide messaging, fundraising and campaign stops.
Pradeep Mutalikassociate research scientist, Yale Center for Medical Informatics
That said, there's little doubt the Clinton campaign put predictive analytics much closer to the center of its operations. Virtually every decision, from donor outreach to messaging changes, was run through predictive models to test its likely efficacy.
The election result doesn't mean that a candidate more focused on leveraging analytics can never beat a more emotion-driven one. In practice, analytics and emotion aren't necessarily mutually exclusive. But Mutalik said Trump's victory "exposes some cracks in the purely data-driven approach" -- cracks that corporate analytics teams would be wise to keep in mind as well.
Presidential election spotlights predictive analytics weaknesses
Speed up predictive modeling efforts for a bigger business payoff
Predictive analytics requires one part science, one part art