7 Questions with Dr. Eric Siegel

Everyone is talking about predictive analytics, artificial intelligence and machine learning. But these buzz words can be overwhelming to the average subject matter expert (SME), who before the world of big data just wanted to do his or her job. In the information age, SME’s have to be more strategic, effective and efficient.

Download DMWay’s Solution White Paper

We asked Dr. Eric Siegel, founder of Predictive Analytics World and author of “Predictive Analytics: The Power to Predict Who Will Click, Buy, Lie, or Die – Revised and Updated,” what the story is with predictive analytics and if it’s going away or time for SMEs to grab the bull by its horns.

DMWay: Why would a company pay for predictive analytics software or a data scientist? Is the benefit really that great?

Siegel: Predictive analytics is booming. It reinvents industries and runs the world. Companies, governments, law enforcement, hospitals and universities are seizing upon its power. These institutions predict the behavior or outcome for each individual.

Why?

Big DataFor good reason: predicting human behavior combats risk, boosts sales, fortifies healthcare, streamlines manufacturing, conquers spam, optimizes social networks, toughens crime fighting, and wins elections.

When it comes to data science or big data, predictive analytics is the key. The ultimate use of data is to learn from it and then predict. The most actionable win from data is prediction, which will more effectively drive organizational operations.

The field of predictive analytics seems to be moving fast. Every few months we take another bound forward. Where do you see the field in five years?

There are two main leaps now taking place. First, the adoption of predictive analytics is widening across sectors and across business functions. More and more industries are taking it on. In the last few years I’ve been commissioned to provide introductory speeches at events in these industries: marketing, market research, e-commerce, environmentalism, financial services, insurance, news media, healthcare, pharmaceuticals, government, human resources, travel, real estate, construction and law. And, within the enterprise, penetration is deepening. Predictive analytics is being adopted more and more intensely across all functions.

Second, the regular practice of data scientists is becoming formalized and automated, such as the most intense part, the pre-processing steps. DMWay is a leader in this realm. It fully automates the greatest bottleneck of the process: preparing the raw input variables for prediction by transforming and selecting them.

What are the benefits to the consumer?

Although an organization’s impetus to deploy predictive analytics usually rests in its bottom line, the value to consumers is also tremendous. In effect, it delivers an anecdote to the information overload we face each day as consumers. Movie, music and book recommendations help us filter through far too many options to review manually. We see only trivial amounts of spam rather than the crippling amounts we’d get without the predictive model that serves as a filter to our inbox. Likewise, more precisely targeted direct mail means less junk mail for us (and fewer trees felled). Google search results are ordered based on what’s predicted as more relevant. Similarly, Facebook orders your feed, thus prioritizing among thousands of recent posts from your contacts. And did I mention predictive analytics fortifies healthcare, streamlines manufacturing, and toughens crime fighting?

Download DMWay’s Solution White Paper

Give me your favorite four examples of what predictive analytics can help predict:

The title of my book ends with those four: “Predictive Analytics: The Power to Predict Who Will Click, Buy, Lie, or Die.” Predicting those behaviors/outcomes optimizes online search, e-commerce, direct marketing, targeted customer retention, law enforcement, disingenuous Amazon reviews, life insurance, and healthcare operations.

But using predictive analytics comes with its share of ethical dilemmas. Can you give me some examples of why people might be cautious about the data (and then predictions) machines are getting about each of us?

I devoted the entire second chapter of my book, “Predictive Analytics,” to the particular ethical issues that arise in the deployment of predictive. Namely that a predictive model may infer about you sensitive data you did not volunteer, such as whether you are pregnant, whether you will quit your job, and whether you will commit a crime.

Can you offer an argument to temper those fears?

My first impulse would be that people should learn more and help spread the word about these pitfalls. Only then, with increased awareness and caution, can fears be allayed. With great power comes great responsibility. The power of prediction is not going to be put to rest, so we need to collectively, as a civilization, take care when deploying it.