The success of the Oakland Athletics in building a winning baseball team by using unique performance metrics tools, as shown in the movie Moneyball, could be repeated in corporate IT shops, say IT executives and analysts.
Moneyball, which opened last week, is based on a 2003 book of the same name by Michael Lewis. It describes how the Oakland A's general manager Billy Beane eschewed traditional metrics such as RBIs and home runs when evaluating and selecting players for the team.
Instead, he focused on lesser-known and rarely used metrics such as walks plus hits per inning pitched (WHIP), on-base average, and value over replacement player (VORP) when deciding how valuable a player would be to the team
His approach resulted in the creation of an Oakland baseball team that made it to several playoff rounds in the early to mid-2000s even though it had the lowest payroll in Major League Baseball.
"Like professional sports teams, traditional companies must rethink how [such] measures can also drive changes in business processes." said Rita Sallam, an analyst with Gartner.
Enterprises must find ways to modify and optimize their processes by using lessons learned from early adopters of such approaches like the Athletics, Sallam added.
Social Compact, a Washington D.C.-based non-profit, is already applying such methods with considerable success as part of its effort to get businesses to invest in struggling inner cities.
Social Compact collects and analyzes overlooked economic indicators from a community to build a detailed profile of its true economic potential, said Alyssa Lee, president and CEO of the non-profit The profile is far more detailed than can be done using census-derived information, she added.
Social Compact's neighborhood profiles for instance, draw from multiple local sources and includes information contained in building permits, certificates of occupancy, utility usage data, tax assessment figures, bill payment patterns and consumer expenditures within a community. It also looks at secondary sources of income that people in a neighborhood might have, such a second job.
Often such profiles help city planners, community organizers and businesses look beyond negative stereotypes that often come from census data-based evaluations, Lee said. "We create an information base that is representative of the true economic strength of a community," she said.
Social Compact collects data from a variety of public, private and commercial sources within a local community.
The non-profit uses technology from DataFlux, a SAS subsidiary, to standardize, validate and to integrate the data. It also uses the DataFlux tools to develop maps of a local community that pinpoint areas that lack basic amenities such as a grocery store, a pharmacy or a bank.
The techniques described in Moneyball can be applied to key business processes such as vendor selection and portfolio optimization in any organization, Sallam said. "Player selection for instance, is similar to vendor selection. You need to look at quantitative and qualitative measures," when selecting a vendor, she said.
Enterprises can benefit enormously from identifying all of the quantitative and qualitative factors that contribute to success, not just the obvious ones, she said.
For example, a company might assume that a customer is more likely to leave them if they get a better price offer from a rival rather than if they have a negative experience with the company. The reality could be more nuanced. A closer examination of data could show that customers are far more likely to respond to a rival's offer if they have even one bad experience with the company.
Another area that could benefit from data-driven management of the sort espoused by Beane is customer contact centers, according to Merced Systems, a vendor of performance management systems.
Contact centers are "rich in underused data and under-tracked processes, especially compared with functions like finance or manufacturing," the company said in a recent whitepaper.
Often, it said, vital components of a contact center are not measured because the data is scattered across multiple data stores and silos. "As a result, many centers favor those management metrics that are the easiest to get to, rather than those that correlate highly with profitability and customer loyalty.
Examples of uncommon and non-obvious metrics that have a major impact on call center performance include coaching frequency, supervisor effectiveness, bonus calculations and the frequency with which statistics are used to bolster decisions.
"Finding these 'Moneyball Metrics' involves a disciplined process of gathering hard to get data," and correlating it with the desired outcomes, the whitepaper noted.
Even the federal government is taking a leaf from Moneyball.
In a blog post Wednesday, Shelley Metzenbaum, associate director of performance and personnel management at the Office of Management and Budget, called on federal agencies to follow Beane's example.
"Like Beane, who understood that his goal was to win games - not hit the most home runs, government agencies must learn to be clear about what they want to accomplish" and be willing to change processes, Metzenbaum said. Agencies need to look for and identify relevant data and factors that are most likely to create problems and increase costs, Metzenbaum said.
By applying Moneyball-like data analytics, the Department of Transportation discovered that 20% of motor vehicles crashes in 2009 resulted from distracted driving, making that a priority focus for the agency, Metzenbaum said.
Similarly, the Social Security Administration is applying predictive analytics approaches to determine the best criteria to apply in the process of qualifying people for disability payments, she said.
"There may never be a movie about the management of the federal government, but the Administration has been taking its own Moneyball approach to management," Metzenbaum said.
Jaikumar Vijayan covers data security and privacy issues, financial services security and e-voting for Computerworld. Follow Jaikumar on Twitter at @jaivijayan , or subscribe to Jaikumar's RSS feed . His e-mail address is email@example.com .
Read more about bi and analytics in Computerworld's BI and Analytics Topic Center.