Competing on Analytics

How fact-based decisions and business intelligence drive performance

Companies have long used business intelligence for specific applications, but often these initiatives were too narrow to affect corporate performance. Now, however, some leading organizations have begun to base their competitive strategies on the sophisticated analysis of business data.

Tom Davenport, author, professor and director of research for the School of Executive Education at Babson College in Boston, has taught at Harvard Business School, the University of Chicago, Dartmouth's Tuck School of Business and the University of Texas at Austin. As an educator, Davenport has researched companies that compete on the basis of their analytic prowess, hoping to quantify how enterprise-level business intelligence and fact-based decision making can provide a competitive advantage.

We caught up with Davenport during his recent visit to SAS' Cary, N.C., world headquarters.

Can you provide an overview of the business intelligence research you've been doing?

Davenport: Our research team is determining how companies compete with analytics. We've talked to 20 large organizations so far – such as Proctor & Gamble, Bank of America, Wachovia, Harrah's. These are all organizations who are pretty sophisticated in their use of BI.

We're starting to see that there really is a class of companies that have identified analytics and BI as the primary bases upon which they're competing – the way they manage customers more effectively, manage global supply chains, manage their financial performance and so on. The leading companies are definitely competing on analytics and are pursuing it quite aggressively.

What do you mean by "competing on analytics"?

Davenport: Historically, "business as usual" has meant taking whatever customers that came, managing supply chains on a just-in-case basis and looking backwards on financials to understand what happened last month. The companies who are competing on analytics have gotten rid of this definition of business as usual. Competing on analytics is really about making the most of these key processes within your organization.

Can you expound on the factors that these companies use to differentiate themselves?

Davenport: Companies can differentiate themselves through analytics in a variety of processes, but there are three major ones:

• Customers: Companies can either take any customer that comes or they can choose the most profitable customers. The companies that compete on analytics would obviously focus on the profitable customers.
• Supply chain: They can have a lot of inventory around or they can very carefully manage a global supply chain so everything arrives just when it's needed. These companies can then route around any problems or contingencies that happen in the supply chain.
• Financial performance: Companies can look back at what happened last month or they can model and anticipate their performance – try to understand the drivers of profitability and financial performance.

The companies that compete on analytics would take the more predictive, aggressive approach to these processes.

What type of analytics is most effective for companies trying to compete on analytics?

Davenport: Descriptive analytics have become something that anybody can do; there's not much differentiation there. The companies that are focused on analytical competition are oriented to a much higher level of analytics: predictive modeling, optimization techniques and, in some cases, even the automation of decisions – a higher level of analytical capabilities than we've been used to in the business world.

In your writings, you describe some organizations as having an "analytical culture." Can you explain that concept for us?

Davenport: Just about anybody these days can buy the technology to do sophisticated analysis; almost everybody has a lot of data and a data warehouse. So what really distinguishes the analytical competitors from people competing on more traditional factors is this idea of an analytical culture. It's the secret sauce behind analytical competition. It basically means that senior executives have said, "We're going to use data and sophisticated statistics in our decision making; we're not going to be satisfied with the same kind of information that other organizations have gotten." And they're relentless about pushing this through many aspects of their organizations, targeting the kinds of processes and behaviors that it takes to be successful.

How can you integrate analytics into the decision-making processes at the executive level?

Davenport: It's very important for there to be a close, iterative relationship between business analysts and decision makers. Senior executives have to give the analysts some sense of the decisions they are faced with, what data might be relevant and so on.

Some of the organizations that we've been researching are quite aware of this need for collaboration between executives and business analysts. For example, at Bank of America, executives realized the bank's future depends quite heavily on what they call the Three I's: information, intelligence and ideas. The bank has an executive development program that educates its leaders on the aspects of the Three I's that they need to know about, so they can do some of the analysis themselves and communicate more effectively with the business analysts.

How are successful companies broadening their use of analytics while still driving toward common goals?

Davenport: One of the difficult balances for organizations to maintain when they're competing on analytics is this idea of focus versus breadth of application. When we interviewed Gary Loveman, the CEO of Harrah's, he said he agreed his company was competing on analytics, but Harrah's is very careful to focus its efforts on specific, high-impact areas of the business. For Harrah's, that focus needed to be on customer service, for its loyalty programs primarily. Other organizations said the key focus is on financial performance or supply chain. I suspect that, eventually, as companies get really good at this sort of thing, they can broaden it out even further. But right now, in the early stages of analytical competition, there needs to be a focus.

If you would like to receive a copy of Tom Davenport's complete Research Report, please register at: www.sas.com/davenportstudy.

This independent research study is being conducted through the Babson Working Knowledge Research Center, and is co-sponsored by SAS and Intel. Working together, SAS and Intel are delivering scalable enterprise business intelligence and predictive analytics on Intel^® Itanium^® 2 processor-based servers to help you predict outcomes and make more effective decisions throughout your enterprise. SAS takes you Beyond BI^™ by making it easy to put The Power to Know^® in the hands of everyone. 

Bio: Kelly LeVoyer is editorial director of sascom magazine.

Tom Davenport, author, professor and director of research for the School of Executive Education at Babson College in Boston.