by Jessica Tsai
Tuesday, October 30, 2007
ORLANDO, FLA. -- Walking on stage here yesterday at SPSS' Directions 2007 North American Conference, author Tom Davenport sported a Boston Red Sox cap and used the 2007 World Series Champions as an example of how predictive analytics can give organizations a competitive advantage.
"The Oakland A's had analytics and no money," Davenport said, referring to A's general manager Billy Beane, who introduced the power of mathematics and statistical analysis to the day-to-day operations of running a major league baseball team. "The Yankees had money and no analytics," he added. "The Red Sox have both money and analytics," which he believed contributed to the team's second championship in four years. Not without taking a few additional jabs at Yankees fans in the audience, Davenport, as part of his presentation, "Competing on Analytics: How Fact-Based Decisions and Business Intelligence Drive Performance," proceeded to emphasize the importance of predictive analytics. His formula, he said, could be broken down using the acronym DELTA:
- Data: Before any company can launch into effectively utilizing predictive analytics, that company needs good, hygienic data. For most companies, getting data isn't the problem -- the problem is actually doing something with that data. (Along the same lines, SPSS invited Chicago-based comedy troupe Second City to entertain audiences between keynote presentations. During one skit, an actor reiterated a problem common to many organizations: The only action taken on his massive amount of data was that it was being stored. Worse still, according to Jack Noonan, SPSS' chief executive officer, by 2010 or 2011, data will double every few hours. In today's competitive world, you can't rely on your gut instinct just by looking at the data anymore, Noonan says.)
- Enterprise: Siloed operations are out, Davenport told the crowd. In order to function as a whole, organizations need to act as a whole. The age-old dilemma of a sales-and-marketing disconnect is just one example of how organizations are failing to communicate. Davenport said that Excel spreadsheets are the analytical equivalent of rolls of duct tape: Sure, they're useful, but you'd never want to build a house with them. Organizations, for example, should focus on maintaining one customer database for the entire enterprise, Davenport said, not separate data stores for each product or service. Companies who have been doing well with this type of integration, according to Davenport, include Procter & Gamble with its Global Analytics and Royal Bank of Canada (RBC), which in the 1970s began using technology to analyze its customer data.
- Leadership: Davenport shared an anecdote about Gary Loveman, chief executive officer of Harrah's, the casino operator. Loveman has three ways his employees can get fired, according to Davenport: 1) Steal; 2) Harass women; or 3) Fail to use a control group. He always asks his employees, "Do we think or do we know?" pushing heavily for analytics-based decision-making. Organizations need to have a strong leader who believes in the company's initiatives before they can embark on ventures that often turn out to be costly and risky. Furthermore, leaders often require hardcore data and logical justification. Therefore, before setting out on the "Prove it!" path, Davenport warned, have a discussion and make sure your leader will believe in the project by the time it's deployed. Davenport described an example when Amazon.com Chief Executive Officer Jeff Bezos fired 10 out of 12 employees because they failed to use empirical data.
- Targets: Companies have to decide where they want to target their technology in order to focus their organization's initiatives, Davenport said: Start from a target and then evolve from there. He described how Wal-Mart wanted to target its customer relations and therefore focused on linking internal analytics from its retail portals out to its suppliers.
- Action: Davenport admitted that he's gone back and forth over whether the "A" in his mnemonic should stand for "analytics" or "action," but finally decided that no use will come of analytics if no action is taken. So, with effective action, organizations should embed analytics. Seeing a high attrition rate of human resource employees, for example, a particular company began analyzing the characteristics of these individuals. The company found that those who were planning on leaving were more likely to not sign up for 401(k) plans. To minimize the high cost of hiring and training new employees, the company could embark on two options in response to the analytics: 1) Make signing up for the 401(k) nearly mandatory, or 2) Don't hire those who decline to sign up for a 401(k) plan.
Davenport also made clear how important first steps really are. To employ predictive analytics, he said, test on a small scale using a controlled experiment. That way, you'll see concrete results more easily. Then, be willing to change your current process and behaviors. It's not easy to have a scientific discussion, Davenport noted, but scientific evidence is critical in the pursuit of change.
Companies will benefit in the long run as they are increasingly able to make proactive, instead of reactive, decisions. With predictive analytics, there can be real-time action, which is often the most effective. At Harrah's, for instance, when a customer is losing, the casino will provide a $20 buffet coupon to lift her spirits, thereby reversing (or, even better, preventing) any negative attitude toward the casino.
Regardless of how a company chooses to use predictive analytics, Davenport advised everyone to get started. "There's not much time to spare because somebody's going to become your analytics competitor," he told the crowd. But getting started means looking for other options, as well, he said; in that sense, he seconded the opinion of SPSS CEO Noonan: Predictive analytics, Noonan says, "is not a silver bullet. It's a journey that moves across an organization."