The Wisdom of BI Crowds
A study by Dresner Advisory Services suggests that business users are winning the war with IT over business intelligence. Most new BI projects appear to be small in scope, and to favor emerging BI vendors.
Even as demand grows for data analysts, a new study finds a disconnect between what universities teach and businesses seek.
"The State of Business Intelligence in Academia 2010," sponsored by the industry group Business Intelligence Congress II, includes responses from 173 professors at 129 universities and 339 students at 62 schools, as well as hiring managers. A similar survey in 2009 surveyed only professors. (The full report is available here.) The input of hiring managers and students brought some added perspective to the situation. (The study uses "business intelligence" and "analytics" interchangeably.)
Companies reported that too often, job candidates are adept at the technology, but lack business acumen or vice versa. But they're looking to hire people well-versed in both.
... universities are delivering BI through silos - statistics departments are delivering BI classes, business schools are delivering BI classes, math programs are. But what recruiters are saying is that they need students with a breadth of skill sets and those skills have to be developed across the silos. So there's a disconnect between the programs that exist and the needs of the marketplace. So universities have to change the way they create their BI programs going forward.
And the second big finding was the need for much more realism in the kinds of curriculum that students are exposed to so they're better prepared for real-world jobs.
Teradata's Scott Gnau described the need as:
... people who can be conversant between tools and technology and bring that to real business value.
Big Data has been getting a lot of attention lately, though a recent McKinsey report calls the labor force in this area inadequate. It says the United States faces a shortage of 140,000 to 190,000 people with analytical expertise and 1.5 million managers and analysts with the skills to understand and make decisions based on the data. It projected that the number of highly skilled BI workers by 2018 in the United States will exceed the available work force by as much as 60 percent. It's no wonder, then, that the University of California-San Diego ranked data mining and analytics No. 2 on its top 10 list of hot jobs, after health care IT.
Wixom said the 2009 survey showed that universities were just getting started in BI education. The 2010 survey, however, found that of the 129 universities studied, only three offered an undergraduate degree in BI and 12 offered a graduate degree.
Part of the problem is that people aren't clear about what BI is - Is it technology? Is it statistics? Is it business? It's all of the above, and for programs to be successful, Wixom says, they must be truly interdisciplinary. But gaining that level of cooperation is not as easy as it sounds.
There's also a problem of finding qualified faculty and providing more guidance about what the curriculum should include. For most business schools, which generally lack faculty with deep BI experience, there's a single line in the curriculum that says "business intelligence," with no more guidance than that, Wixom says. Data experts from the North American Conference for the Association of Information Systems are expected to gather in August to work on those curriculum guidelines.
And there's the issue of gaining real-world data sets for students to work on. Company data tends to be proprietary, what Gnau called part of their "secret sauce," so it's quite the trick to get companies to provide real data for the classroom.
Gnau pointed to a number of factors coming together to drive interest in data analysis: Companies have completed their big ERP and back-office standardization projects and are looking to move on to the next big project, the technology has become cheaper; and companies have more data available to them, including social media, machine-generated data, GPS and others. Even companies that have used BI in the past are finding it's no longer limited to past behavior, but increasingly can be used to predict the future. And with the barrier to entry with this technology ever lower, it's no wonder that even smaller companies are getting on board.
Industry demand and the information from surveys such as this are pushing universities to get this worked out, Wixom says, and she's hopeful that the August conference will be a step toward more guidance and more sharing among universities. But she says for students to come out of universities with the real-world experience businesses seek will require companies to step up and join the universities to make that happen.