The Data Scientist: A Sexy Hybrid Role


Countless blogs and articles have proclaimed “data scientist” the hottest job title of the year. Indeed, Harvard Business Review called it the “sexiest job of the 21st century”. Hyperbole or not, it appears that this will not just be a one-year phenomenon. A 2011 paper from McKinsey Global Institute identified a significant talent gap through 2018: as much as 190,000 deep analytical roles and 1.5M data-aware managers in the US alone.

It remains to be seen whether these estimates are accurate, as two years is a long time in tech and the tools available to address data analysis are making leaps and bounds, particularly in terms of usability. Nonetheless, a fundamental shift in the culture of decision-making in large organizations is happening. The digital economy has an implicit assumption of data-driven decisions wherever possible, and data is no respecter of hierarchies. Consequently, we are seeing a data-driven approach reshaping organizations, communication styles and the attributes of successful managers.

Meantime, another set of assumptions seems to be getting in the way of establishing a smooth glide path to that future of which proponents speak so glowingly: the issue of where to find and develop those skills. I like to remind my own team members that “even the tech business is 10% technology and 90% psychology”, so this is an important topic for the graduating classes of our universities, as well as for hiring managers in the organizations that would benefit from enhanced capabilities in the field of data science.

There is plenty of debate about what a “data scientist” really is, and there are even good arguments – such as those advanced by Douglas Merrill, CEO of ZestFinance – that what we need are data artists. However the more I learn about the topic, the more I think that we have been looking in the wrong faculty buildings at our educational establishments.

Because it is often viewed as a “tech problem”, too many leaders are still looking exclusively to their own IT departments internally and, for recruiting, to the Computer Science departments of academia. However, it turns out that the skills most relevant to identifying and elevating those most valuable insights are not solely developed in classes on programming, or principles of systems, or even on statistical analysis and modeling. Many of the drivers for demand in data science relate to the analysis of human behaviors and propensities, whether individually or in groups.

Fortunately there is a much larger pool of resource with skills in understanding, analyzing and quantifying those types of pattern: they are graduates in the social sciences. Of course, the most obvious candidates among members of those graduating classes will be the ones who have combined studies in both the “softer” ends of that spectrum (e.g. behavioral economics) with the “harder” (e.g. quantitative economics). It is also important to remember that psychology, sociology and political science majors are intensely analytical and often – especially at graduate level – strongly quantitative. Their skills form a good basis for equipping team members in this most important new function at the business/technology interface.

So, be demanding. But certainly don’t be shy about broadening the scope of relevant domain expertise when looking for your future data scientists. Look beyond IT or deep mathematics skills. The best candidates – and the best technology platforms – will cover the gap before you know it.

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>