Q&A: Big Data Past, Present, and Future
From the Spring 2012 OPEN magazine cover story, "The Big Data Machine."
Anitesh Barua is a professor of information, risk, and operations management and associate director of the Center for Research in Electronic Commerce. He spoke with Steve Brooks about data's competitive advantage, what's ahead for Big Data, and how McCombs is preparing students for the jobs it's creating.
Many analysts say we’ve entered the era of Big Data—sets of data that are too large for conventional database software to handle. Why is this significant for businesses?
In the past, when a limited amount of data was available from point-of-sale systems, most companies didn’t do much with the data. About 2 percent of data from checkout scanners got analyzed. Today, it’s like an avalanche. We’ve changed the units by which we measure data. Suddenly, terabytes seem small. There will be a big thrust on business analysis and business intelligence for the foreseeable future. Today, businesses are realizing it’s a lever for competitive advantage.
How can data be a lever for competitive advantage?
If you and I start an online video rental company, we don’t have data to start with. Whereas Netflix is sitting on billions of customer reviews that they can mine. They have a huge advantage over anybody starting out. The same is true with Amazon. They can dig into their database and pull out hundreds of thousands of people like you and predict what you’re likely to buy. Smart companies are leveraging the big data they’re sitting on to make decisions at every point in the value chain. Not just on price, but they’re making operational decisions, too.
Can you give some examples of using data to make operational decisions?
Harrah’s is one of the best-known examples of a company that runs on data. It’s in the casino business, which is surprising. You don’t expect deep analytics. Yet this company started analyzing its data. Now, it can even answer questions like, “Why were you playing that particular slot machine? Was it because nothing else was available, or did you figure you would have better odds?” They can combine that with how valuable a client you are. Then, when they buy slot machines, they take into account what kinds their valuable customers like to play at.
Or take Dell. What’s been talked about is their just-in-time inventory, but what’s equally amazing is how they sell. Say, Dell finds a great deal on some less-than-stellar processor. They’ve bought 100,000 of them. Now, they need to sell 100,000 boxes. Who should they target? If you are a customer who’s bought from them before, they categorize you. If you are hankering for the latest and greatest, they won’t send you offer. Instead, they say, “Let’s pull up a large number of customers from our database who are likely to buy a low-tech box from us. Let’s send them an offer with a certain price point and see how they react. If the reaction is less than favorable, we sweeten the deal. We send it off again and again, until we hit the sweet spot.” It’s a sales strategy happening in real time based on customer data.
What role has social media played in this explosion of data?
The amount of data now available just went up several orders of magnitude because of social media. But it’s not really a brand-new concept. Companies tracking customer behavior online were making smart marketing decisions before social media. But now we reveal so much about ourselves in social media, it’s a new opportunity for companies to tap into that knowledge.
Social media is interesting because it’s not just one way. You can see interactions between individuals in humongous groups. Before, we could collect your browsing habits. Now, I suddenly get to see you as you live through the day.
What’s ahead for Big Data?
The question is, how long will this competitive advantage last? Best practices spread more rapidly today than they did 10 to 15 years ago. As a result, we see more and more companies making inroads with effective data analysis. I don’t know whether, 10 to 15 years from now, they’ll still have that competitive advantage. But companies live for the next 10 years. So for the next 10 years, we’ll see a huge rush of investments in Big Data.
The McKinsey Global Institute projects that U.S. companies will need 1.5 million workers with data analysis skills over the next few years. What is McCombs doing to prepare today’s students for these jobs?
One of the biggest challenges corporations will face is the shortage of human capital in this area. We want to be one of the premier suppliers of human capital. We’ve been offering courses relating to business intelligence and data analytics for quite a while now, and we are developing a master’s program in business analytics.
Where we differ [from other programs] is that our students learn how to apply these ideas in a business context. Our students have to do real-world projects and present the results to clients: How can we retain 5 percent more customers? What levers do we push? Can you tell us from the data?
Ultimately, it’s not about the software that does the data mining. Every company can buy that. Where it makes a big difference is in the connection between the data and business performance, how to create more value for my customers, how to become more efficient operationally. We try to connect those dots.
These jobs cannot be offshored easily. You can offshore pure technology jobs. What you cannot offshore is somebody’s job who’s a liaison between business decision-makers, technology users and technicians. We want to position our students in that space.