I’ve always gotten a kick out of the way certain catchphrases emerge in the IT realm every couple of years, and how IT vendors rush to spin their product stories around whatever catchphrase seems to be enjoying the most hype at the time. Ten years ago, it was all about the emergence of “utility computing.” When that became blasé, everybody jumped on the “green computing” bandwagon. That morphed into the “cloud computing” frenzy, which now has also become kind of ho-hum. Fortunately for the IT marketing crowd, “Big Data” swooped in to capture the imagination of an industry that seems to suffer from a communal attention deficit disorder.
Then there are the technologies that don’t get as much hype, but are a lot more tangible and are actual game changers. I would put predictive analytics in that category, and I would list Eric Siegel among the most intelligent voices on the topic. In fact, if he doesn’t top that list, he’s awfully close.
Siegel is a former Columbia University professor, founder of Predictive Analytics World, and author of the new book, “Predictive Analytics: The Power to Predict Who Will Click, Buy, Lie, or Die.” In an email interview with Siegel last week, I asked him whether the emergence of Big Data has had an impact on predictive analytics, or if it’s just a more recent catchphrase for something that’s been an element of predictive analytics for some time. Siegel said what’s key is channeling all the excitement:
"Big Data" suffers from a lack of agreed definition, but suffice it to say everyone is excited because there is more and more data—a humongous amount. Predictive analytics is the most powerful solution to the "too much data" problem. Excitement over "Big Data" begs the question of what's to be done with all of it. Answer: The most actionable thing to get from data is attained by learning from it to render predictions for each individual, since those predictions directly inform and drive each of the millions of per-person decisions organizations make every day. That is predictive analytics in action.
Referring to the use of predictive analytics to gain customer intelligence, I asked Siegel how and where the line should be drawn to avoid violating customer privacy. He said there’s no clear line to be drawn:
With predictive analytics, organizations gain power by predicting potent, yet—in some cases—sensitive insights about individuals. The fact is, predictive technology reveals a future often considered private. These predictions are derived from existing data, almost as if creating new information out of thin air. Examples include Hewlett-Packard inferring an employee's intent to resign; retailer Target deducing a customer's pregnancy; and law enforcement in Oregon and Pennsylvania foretelling a convict's future repeat offense.
We're only at the very beginning of the debate process around this, in that both sides understand relatively little about the other side's perspective. There is no clear line to be drawn, but rather a lengthy collaborative process of debating and exploring that needs to take place. Activists and citizens have some catching up to do: Only by learning more about how data exerts predictive power can the societal issues be identified and addressed.
I asked Siegel how predictive analytics is affecting the mobile app market. He said the power to predict adds value to all kinds of activities:
Here are a few that people may experience on their mobile devices:
And here are a few other examples that are not apps for the people holding the phone, but help improve their experience:
Finally, I mentioned to Siegel that I co-wrote the book, “Spy the Lie: Former CIA Officers Teach You How to Detect Deception,” which is based on a methodology for detecting deception that was developed within the CIA, and explains how it can be employed in daily life. In that context, I asked him to share his insights on how predictive analytics technology can be used to predict who will lie. His response:
Wait, are you a spy or something? Anyway, as with medical diagnosis and assessing the risk of an applicant for insurance coverage, predictive analytics augments the state-of-the-art to improve—by way of machine learning methods—the ability to assess an individual's risk based on the collection of known characteristics of that individual.
In the case of lie detection: