The Harvard Business Review has called it the ‘‘sexiest job of the 21st century’’, and with roles at the top of the field commanding almost $300,000 a year it’s no wonder that data scientists are in increasing demand by banks, businesses and consultancies.
But what is it and what makes one good at it?
Data science as a profession is, in large part, a product of its time and place. We’ve had access to analytics for years, but today’s technology is changing the game.
Cloud storage now means it is cheap to for companies to keep volumes of data on someone else’s hardware, while powerful processors mean that ‘‘what’’ we can analyse is easier than ever.
‘‘People are now thinking of new ways of using and leveraging that data,’’ says Steve Psichalos, a partner in data and analytics at EY.
If we want to run models to predict our customers’ behaviour, for example, we’ve got better data than ever to turn to, and easy access to powerful technology to crunch’ the results.
‘‘Things you couldn’t test previously, or models that took days to run can actually be run very quickly now,’’ says Psichalos.
The role of the data scientist is partly to run the programs or develop the models, but also to understand what businesses are trying to do with their data.
‘‘They work out how they can use the data that’s available, through the analytics, to help solve business problems,’’ Psichalos says.
The result means that data science is not simply a technical role, There’s a strong problem solving component to the job, and a good dose of creativity for those who do it well.
‘‘To me, there’s a mix of art and science involved,’’ says Psichalos.
The best data scientists can think creatively about businesses and systems, but are also quite comfortable with the technology, so they can connect the two and build the models.
For UNSW undergraduate Jacky Koh, it’s a convergence of requirements that means his own future looks bright.
‘‘Data science is definitely where I plan on going,’’ he says. ‘‘It utilises statistics, which I like very much, and it can be used to improve whatever field I’m interested in. In my opinion you need to understand the problem, then have some hacking skills, and then have skills in maths and statistics.’’
Koh sees his skills as a future data scientist could be used in healthcare, or to solve problems like road congestion or pollution.
His studies at UNSW in actuarial studies and science have helped, but so has doing a number of MOOKs (Massive Open Online Courses) to get more hands-on experience.
‘‘I learned the necessary skills that way, then got random data sets online and played around with them,’’ he says.
For now, he’s brushing up on the missing link in his skill set – experience working with businesses – via a hard won paid internship with EY’s data and analytics team.
‘‘You need to develop skills to clearly communicate the benefits of whatever you’re proposing as a data scientist. It’s working with people, not just computers,’’ he says.