Assembling a group of talented people with diverse skills is the best way to meet your data science needs, says Deloitte director.
Big Data Talent War: 7 Ways To Win
(click image for larger view and for slideshow)
Having trouble finding a qualified data scientist, a multitalented individual skilled at data manipulation and modeling, programming, and interface design, and also fluent in the language of business?
Good luck with that. A more pragmatic approach is to build a data science team of members with particular skills that help your organization analyze and visualize data, said David Steier, director of information management for business consulting firm Deloitte.
In a phone interview with InformationWeek, Steier gave his perspective on the reported shortage of data scientists, and how organizations can successfully assemble an analytics team, even when talent is hard to find.
"In order to have an impact in analytics in the real world, especially in the business world, you need to combine a group of people who can work together," said Steier, a computer scientist by training, with a PhD from Carnegie Mellon University.
Steier said it's true that data scientists adept in a variety of technical and business disciplines are in short supply.
"You don't find all of these skills in one person. It's very rare to find even a few of them in one person," he said.
"Not every person who can handle the math is willing to engage with customers, or deal with the situation of reality," said Steier, adding that analytics whizzes may "prefer the cleanness of numbers" to social interaction.
And that's where the team approach can help, although the added expense of hiring multiple team members, each with a unique skill set, may be less appealing to cost-cutting companies.
"It's not going to be a cheap endeavor," Steier said.
Then again, if money is tight, an organization can build all or part of its data science team overseas where labor costs are lower. This approach is particularly effective when a company is seeking people with analytics skills.
"Frankly, the educational system is probably turning out more data science-ready analytics folks in countries other than the U.S," said Steier, who adds that Deloitte hires a "significant group" of people who do analytics work in India and China.
Teamwork is essential, however, regardless of whether the group is shoehorned in a small room or dispersed around the globe. In fact, managing a data science team is "not for the faint of heart," said Steier.
"You really have to commit to having business people work alongside the folks with the technical skills, and for them to try a variety of approaches," he adds.
In the long run, the data scientist shortage will likely correct itself, Steier believes. One reason is that salaries are rising -- always a viable way to attract talent -- and schools are providing more classes with a data science angle.
"We're seeing a number of universities offering courses in business analytics," said Steier. "The enrollment in these is rising pretty rapidly, so you already see market forces at work trying to balance out supply and demand."
Another factor that may play a major role in ending the shortage: Advanced tools that allow business users without analytics chops to pull insights from big data.
"I liken it to the old problem of telephone operators," Steier said. "I'm sure if they did projections a long time ago, they figured out that half the workforce would need to be operating a switchboard."
What happened, however, was that much of the switchboard operators' work was eventually automated. The data science profession could follow a similar path.
Predictive analysis is getting faster, more accurate and more accessible. Combined with big data, it's driving a new age of experiments. Also in the new, all-digital Advanced Analytics issue of InformationWeek: Are project management offices a waste of money? (Free registration required.)
6 Tools to Protect Big DataMost IT teams have their conventional databases covered in terms of security and business continuity. But as we enter the era of big data, Hadoop, and NoSQL, protection schemes need to evolve. In fact, big data could drive the next big security strategy shift.
Big Data Brings Big Security ProblemsWhy should big data be more difficult to secure? In a word, variety. But the business won’t wait to use it to predict customer behavior, find correlations across disparate data sources, predict fraud or financial risk, and more.