Wills, for example, took a circuitous path to the role of data scientist. After graduating from Duke University with a bachelor's degree in math, he pursued a graduate degree in operations research at the University of Texas on and off while working for a series of companies before dropping out to take a job at Google in 2007. (He notes that he did eventually complete that master's degree.) Wills worked at Google as a statistician and then as a software engineer before moving to Cloudera and assuming his data science title.
In short, big data folks seem to be jacks of all trades and masters of none, and their greatest skill may be the ability to serve as the "glue" in an organization, says Wills. "You can take someone who maybe is not the world's greatest software engineer [nor] the world's greatest statistician, but they have the communications skills to talk to people on both sides" as well as to the marketing team and C-level executives, he explains.
"These are people who cut across IT, software development, app development and analytics," Wills adds, noting that he thinks such professionals are rising in prominence. "I'm seeing a shift in value that companies are assigning to these people," he says.
Sacheti, too, keeps his eye out for people like that. "We are finding there are a lot more who are flexible in learning new skills, willing to do iterative design and agile thinking," he says.
Roberts agrees. "The innate characteristics of people, like a predisposition to curiosity, can be more predictive of someone's performance in a role than them having a degree in, say, IT or IS or CS," she says.
Wanted: Relentless, Scientific Temperament
Until recently, creativity, curiosity and communications skills haven't typically been emphasized in IT departments, which may be why many employers aren't looking to their IT operations staffs to find people to spearhead big data projects.
The IIA sees data science as resting on three legs: technological (IT, systems, hardware and software), quantitative (statistics, math, modeling and algorithms) and business (domain knowledge), according to Phillips. "The professionals we see who are successful come from the quantitative side," he says. "They know about the technology, but they aren't running the technology. They rely on IT to give them the tools."
Big data also demands a scientific temperament, says Wills. "When we talk about data science, it's really an experiment-driven process," he explains. "You're usually trying lots of different things, and you have to be OK with failure in a pretty big way." Wills goes on to say that there's a "certain kind of relentlessness you need in the personality of someone who does this kind of work."