Next Steps for Analytics in the Big-Data Era

Nov 23, 2009 (10:11 AM EST)

Read the Original Article at

With interest in analytics exploding along with the data that makes predictive analysis possible, there's no doubt that data-driven decision-making is the way forward for most organizations. But do these firms have the talent they need to pull important metrics and key performance indicators (KPIs) out of their data? And how prepared are executives to draw insights and make quick decisions based on their analyses? With CIO surveys, salary surveys and job listings all pointing toward surging interest in analytics, software vendors, commercial interests and universities are scrambling to develop the needed talent.

  • A soon-to-be-published report on "The Current State of Analytics in the Corporation" has been released by the Business School at Villanova University. Using research gathered for the report, the university has revised its curriculum to better support data-oriented decision-making.

  • On December 14-15, the 2,000-member Teradata University Network is planning a Business Intelligence Congress in Phoenix. The goal of the event is to set top-10 agendas for research and course development on analytics.

  • On December 9, Fordham University and IBM will gather leaders from academia, the venture capital community, government and the health care industry for a two-hour discussion on the coming wave of new jobs that will require analytics skills.

Reacting to growing interest in analytics in recent years, the Business School at Villanova University formed a Business Analytics Strategic Interest Group in late 2007. The group, which counts a dozen corporations, two integrators and two software companies among its members, provided feedback on changing the school's curriculum with data-centric decision-making in mind. For example, a new course on analytics and risk assessment was added for undergraduates and the statistics courses required at both the undergrad and MBA levels were updated to be more practical and applied.

"From the point of view of business, what leaders really need is to be able to understand data, and we think these changes will give students a stronger preparation for the kinds of analyses they'll face in business," says Matthew Liberatore, a professor at Villanova who co-authored the report.

The University is also contemplating adding a minor in analytics because it could "supercharge" any business major, Liberatore asserts. "[Such a degree] would help them understand risk, what the possible outcomes are and what the range of possible rewards are for decisions within their disciplines," he says.

Setting a New Agenda

Back in 2001, when the Teradata University Network was first organized, most business schools hadn't given data warehousing much attention. They were just beginning to peer into database issues related to database design, normalization and the writing of SQL queries. Today, business intelligence and analytics studies are fast replacing more generic Information Systems (IS) course work at universities.

"Between the dot-com failures early in the decade and the offshoring trend, enrollments in Information Systems have been declining," says Michael Goul, professor and chair of the Department of Information Systems, W. P. Carey School of Business, Arizona State University. "As faculty started to look for non-commodity IT skills that students would need, BI became the hot topic because it crosses the business world with the IT world. The feeling is that it's not something that companies will quickly take offshore."

The first-ever Business Intelligence Congress has been organized to explore ways academia can provide theory-based methods and design sciences to advance BI and analytics while also giving industry a platform to steer research toward the most significant challenges. The network, which now counts more than 2,000 faculty members and 1,000 institutions from 75 countries among its members, has already done informal surveys on the topic and Goul says the group has several prospective topics on the agenda.

"Some schools are weaving business intelligence study into general business courses while others have developed targeted BI-specific courses, so that's one topic we want to discuss," he says. "We're also going to have panels on buzz topics like sensor networks, managing large volumes of data, and in-database analytics and what implications that has for the analytics development life cycle."

Keynote addresses at the Business Intelligence Congress will be offered by Tom Davenport, the well-known researcher and author of "Competing on Analytics: The New Science of Winning"; Hugh Watson, management information system professor at the University of Georgia and one of the founders of the Teradata University Network; and Scott Gnau, Teradata's chief development officer.

Next-Wave Jobs

To consider the next wave of jobs requiring analytic talents, IBM has invited a gaggle of business, government and academic leaders to a December 9 RSVP event at a Fordham University conference hall in Manhattan. Attendees will include Kamal Bherwani, CIO, New York Health and Human Services; Evangelos Simoudis, managing director, Trident Capital; Jonathan Bowles, director, Center for an Urban Future; Ambuj Goyal, general manager, Business Analytics and Process Optimization, IBM; and the deans of the Fordham College of Business Administration and Fordham Graduate School of Business Administration.

According to a recent study by IBM, 83 percent of executives ranked business analytics -- "the ability to see patterns in vast amounts of data and extract actionable insight" -- as their top priority. IBM's report suggests that if training and education programs don't follow, "we may see a skills shortage as the economy rebounds and the technology needs of both the private sector and government agencies increases."

For those who can't wait for the next generation of college graduates to meet current demands, analytics expert Neil Raden suggests that corporations set up self-study programs aimed at developing the required math skills. "There is nothing so special about most 'advanced analytics' that someone with adequate training could not do," Raden writes in this blog post . "I'd urge companies to set up such a plan with an eye to ripening experts over a two- to three-year period with time at work to study."

The upshot of growing academic interest in analytics seems to be that more data-savvy analysts and business leaders are on their way to the job market. But for the problems practitioners may face over the next few years, Raden suggests that with a bit of continuing education, the best candidates may be the business-experienced people "standing right in front of you."