The Promise of Big Data
Can aggregating and analyzing information transform businesses?
September 18 2013 by CJ Prince
There is no question that Big Data promises big reward for companies that successfully harness its power. Opportunities abound for sales and revenue growth, price optimization, recruitment, customer service improvement, productivity spikes—across every department and function in the business.
“There is evidence that points to growth rates being about 20 percent higher in companies that apply Big Data strategies, compared with companies that don’t,” Microsoft Dynamics General Manager Christian Pedersen told attendees gathered for a Big Data roundtable discussion held at the New York Stock Exchange in partnership with Microsoft Dynamics.
Yet the precise path to realizing those opportunities is not as clear, attendees agreed. CEOs of companies large and small continue to struggle to get their arms around both the concept and the practical application of Big Data. Even those who have invested in technology and systems for aggregating data are not sure which initiatives will yield the most fruit. Others are stymied by organizational siloes that don’t allow for the easy sharing of data across the enterprise.
But some early adopters provide solid examples of how to harness data for strategic business objectives. Pedersen cited Nike’s use of sensors in its shoes, which allows runners to measure their pace and workout distances. “It gives the consumer a really good experience, but that’s not why Nike did it. Nike executed the concept because they want to have more information about their consumers,” he said. The sensors collect data about how the shoes are utilized, what kinds of surfaces they’re being run on, how often they’re being worn, and they transmit that information back to Nike. “Once the consumers have used their shoes long enough, they can proactively reach out and sell them new shoes.”
Data analyzed strategically can help companies better anticipate the value of future investments. Aetna, for example, is helping corporate customers better forecast health issues among its employee base, explained Michael Palmer, head of Aetna Innovation Labs. One-third of the U.S. population is afflicted with metabolic syndrome, a condition characterized by factors such as large waist circumference, high blood sugar and elevated blood pressure, among others. Employees with metabolic syndrome are approximately 1.6 times as costly as those without it. “So instead of costing your employer $4,000 a year, you cost them $6,000. That’s a big deal,” Palmer noted. “If you as the employer can get people out of metabolic syndrome, in theory, you can cut your healthcare costs.” By looking at the growth rate within a specific corpoate population and then drilling down to the individual level, Aetna is able to help its large customers predict which wellness programs will give them the biggest return on investment.
Avoiding the Data Deluge
For those just starting out, it’s easy to become overwhelmed by the sheer volume of data, pointed out Farooq Kathwari, CEO of Ethan Allen Interiors. The furniture company collects data about customers from multiple sources, including 2,000 interior designers, who interact with consumers daily and 1,000 truck drivers who deliver the product to consumers’ homes. “We are lucky in the sense that we are a vertically integrated company, so we get information from manufacturing, from retail, from our interior designers, from distribution,” he said. “But the challenge is what do you do with it? How do you prioritize it? If you don’t do that, Big Data can be a burden rather than a benefit.” Kathwari’s strategy is to simplify as much as possible. “It’s about relative importance. God gave us five fingers, so I always say, ‘Give me five priorities,’” whether it’s data, whether it’s product, whatever.”
In general, it can take time—once Big Data systems are in place—until the data can be used strategically, said Jeffrey Sonnenfeld, CEO of the Chief Executive Leadership Institute at Yale, citing the the statistic that while 81 percent of utilities collect data on outages, only 60 percent use it to business advantage. Kevin Burke, president and CEO of Con-Ed, said his utility installed equipment over the past two years to get a better sense of how the transmission system was operating across a large region. “So far the data are being used more forensically than in real-time operations,” he said.
One way to avoid data deluge is to begin with the business priority and then look for the technology and systems to support it, said Jim Taiclet, CEO of American Tower, which operates communications sites around the world, including many in emerging markets that rely on wireless technology for communications. The company identified a top priority, which is to keep its electrical power up 99.8 percent of the time. Sensors were installed on multiple parts of its tower installations, including fuel tanks, which are sometimes the targets of theft. With the sensors, the company can monitor fuel levels in real time. “If they are changing faster than actual usage, then someone’s stealing it and we need to get a truck right out there,” said Taiclet. In the U.S., through its network operations center, the company has tied sensors from fuel towers to the dispatch system, so that trucks can proactively be dispatched to those areas that most need fuel. Thanks to that efficiency, he said, “when we had Hurricane Sandy here, none of our generators ran out of fuel.”
Taiclet pointed out that the technology his company uses isn’t that complicated or costly. Today, Big Data systems can be implemented at a much lower price point than ever before. Pedersen pointed out that, with the advent of cloud computing, companies no longer need thousands of their own servers. Instead, they can remotely harness the power of servers in the cloud. “You get in, you consume the compute power you need; and then, when you don’t need it, you get out. You only pay for the exact time you use it,” he said.
Of course, technology can only do so much. Talent in data analytics is becoming essential for companies that need to ensure the intelligence coming from their data churn is reliable. “We do predictive maintenance on turbo machinery,” said David Cote, CEO of Honeywell. “Part of this is figuring out the vibration monitor. You don’t know what you’re looking for in the beginning. It’s about accumulating piles of data and just trying to figure out what is the predictor of a failure? What signal do I see that says, ‘Vibration has increased. I’m going to have a problem?’ That’s where the math guys come in. For us, it’s going to be a big deal, having those kinds of people.”
Big Data, Big Brother?
The intersection of Big Data and consumer privacy and the potential impact on business, is still relatively unknown—a concern to CEOs who don’t know just what their companies’ legal liabilities will be for aggregating useful data. “Legally, how does this ever get sorted out? Who really owns the data?” posed Cote.
And how much privacy would consumers be willing to forego to allow companies to make their day-to-day lives easier? Utilities employing smart meters to remotely monitor electricity consumption make life more convenient for consumers, but privacy is a major concern, Burke noted. “Because you can tell when somebody’s home, when they’re on vacation, just by by looking at their electric consumption.”
Healthcare is another area with strict privacy laws that don’t allow for certain types of data sharing that could be beneficial to patients. “I think the regulations actually make it harder, in theory, to get better care, because if all that data were available to your full-care team and to you, you’d probably make better decisions. I think we’ll eventually get to the point where folks are willing to share more.”
Meantime, Pedersen’s advice is simple: “If there’s no reason to have personalized data, don’t.”