True "analytical competitors" are using data analysis to wring maximum value from their businesses' processes, thereby driving new competitive advantages. Thomas H. Davenport, an expert in the field of analytics and an acclaimed consultant, analyst and author, recently researched and wrote about these innovative and driven businesspeople in his newest book, co-written with Jeanne Harris, Competing on Analytics: The New Science of Winning (Harvard Business School Press).

Thomas H. Davenport, President's Distinguished Professor in Management and Information Technology at Babson College, says businesses must change their approach to become true "analytical competitors."

Davenport holds the title of President's Distinguished Professor in Management and Information Technology at Babson College. He has written more than 100 articles and 12 bestselling books, and he leads research and consulting initiatives with Babson, the Concours Group and Waterstone Analytics.

Teradata Magazine recently visited with Davenport to discuss his research on the goals and tactics of analytical competitors.

Q Many organizations use business intelligence [BI] to support decision making. Are most of these initiatives applied too narrowly to affect corporate performance?

A Most companies have sprinkled business intelligence all over the place. It hasn't been a visible factor in their competitive capability. Analytical activities must be the most important thing you do—or as I call it, your "distinctive capability."

Q So to compete on analytics, companies need a new approach?

A Yes! Companies must ask, "What are we already better at than anyone else in our industry?" or "What do we want to be better at than anyone else in our industry?" Then they can use analytics and, to a lesser degree, the other aspects of BI to support that capability.

Q Tell us about your research on analytical competitors.

A I interviewed 32 companies to discuss their use of quantitative, fact-based analysis. After this initial research, I spoke with representatives of an additional two dozen organizations and surveyed 400 other businesses. From this assessment, it became clear that no more than 10 percent of large, sophisticated companies are currently using analytics as a significant portion of their competitive strategy. Of those, only half can be defined as true analytical competitors—companies that build their businesses on the ability to collect, analyze and act on data.

Q What characteristics define an analytical competitor?

A They tend to use modeling and optimization technologies to look beyond rudimentary data or statistics. This lets them understand business trends, predict customer behavior and correlate operational costs with financial performance. Analytical competitors also operate their data initiatives under a single, coherent approach. The programs share leadership, technology and tools. Finally, these companies have executive support and passion for analytics.

Q Executive leadership isn't enough, though. Do companies also need to build an analytics culture throughout their work force?

A Yes. You have to be prepared to push the analytics culture throughout the organization. One of my favorite examples is the Boston Red Sox. I'm a big Red Sox fan. The Boston baseball franchise is staffed by expert statisticians and a very analytical general manager. In the 2003 season, they made it to the American League Championship Series. They were ahead in the last game, 5-2, and their best pitcher, Pedro Martinez, was pitching. The organization had volumes of analysis showing that Martinez's performance declined markedly after he threw 105 pitches. Around pitch 106, manager Grady Little asked Martinez if he was ready to hit the showers. Martinez said "No," and Little let him stay. Martinez got shelled, and the Red Sox lost. Little, who didn't respect the team's analytics, lost his job.

Q What are some of the technologies and tools needed for an organization to become an analytical competitor?

A Obviously, you need a lot of data. If you don't have it, getting it can be rather time-consuming. Putting the data in a [data] warehouse is a good idea. What most people have done incorrectly with warehouses is they gather too much data. They don't have a clear set of ideas about what they're going to do with the data when they start, so it becomes an "enterprise data landfill." For technologies, you need very sophisticated analytical software and some pretty capable hardware—64-bit processors and big data warehouse technologies like Teradata's.

Q How can an organization take steps to become an analytical competitor?

A If you have support in the executive suite, you can go the full-steam-ahead path: Hire the analytics people you need, build the databases and begin analysis. If not, you have to take the "prove it" path: Determine what would help your senior managers understand that an analytical strategy would be appropriate and do that. It might be a pilot, or it might be figuring out what your competitors are doing. The key is to start the discussion and to say: "We're not very analytical now. What would it take you to want to become more analytical?"

Q How long does it take for an organization to begin using analytics as a competitive advantage?

A If you've got really high-level support, you can do it pretty quickly. One leading company started to see some pretty dramatic returns after a year or 18 months. But they already had a loyalty program in place, so they had the data. If you don't have the data, it may take a few years. One bank that I studied had a five-year plan that addressed questions such as how to get the data in place, how to change processes and how to change the culture. A five-year plan seems like a long time, but to completely turn around your strategy, it's not unreasonable.

Q Looking forward, what's next for analytical competitors?

A They're moving toward more real-time analyses and actions, where they can shape a customer's experience as it is happening. Certain industries, such as financial services, are continually pushing the frontier on metrics. They already use credit scores to support decision making, but they want to go further. Can they, for example, identify people with bad credit scores who might be good customers? And vice versa?

Analytical competitors also combine analytics with knowledge management—collecting data, conducting experiments and learning from what they've done. And there is more automation of decision making. These companies believe that the best decisions come from the combination of humans and automated systems.

Q Any final words of advice for companies that want to compete on analytics?

A Be willing to take action on the results of your analytics. I've flown 4.5 million miles on a certain airline. In some industries, a customer like me would be treated like a king. Yet there are industries that refuse to grant their best customers better service, and on this airline I'm treated as badly as anyone else. Gathering the data is useless if you refuse to take action on the analytic results. T

Cheryl D. Krivda writes about high technology and business practices.

Photograph by Kenneth Scott.

Teradata Magazine-June 2007