Do you still have an unfilled vacancy for that data scientist you think you need? Could it be that trying to a hire an individual data scientist is where your problem lies? McKinsey has predicted that, by 2018, the US alone may face a 50-60% gap between supply and demand for data science talent.
So, am I counselling you to give up on data science and look elsewhere for help to grow your business? Absolutely not. Rather, from my own experience as a leader in this field and from coaching analytics leaders, I’ve discovered that the team is more important than the individual. When it comes to this recruitment challenge, your solution may lie in thinking team, not individual star.
Many businesses still struggle to recruit the highly-trained analytics talent they need. Data scientist is a challenging role. Such a demanding mix of skills is required, from high-end technical expertise (in data languages and machine learning), through strong numerate skills (in maths or statistics) to the business skills to apply their learning (eg, communication and influencing).
It’s perhaps not a surprise, then, that recruiting managers can struggle to find such an ideal candidate. A few of the infographics on data scientists you can find online attest to how many “hats” such a professional is expected to wear.
But is the answer to keep seeking that perfectly-rounded data scientist? Is success really about one individual? A number of my clients have proved another approach can work better. By putting the focus on the team, not just the individual, they more easily recruit teams that comprise all the skills they need.
Bob Rogers, chief data scientist at Intel big data solutions, agrees. In an interview for Forbes magazine, he said: “Think in terms of data science teams with diverse capabilities that can complement each other, rather than seeking to hire individuals that can do it all…In fact, we’re looking for all kinds of skills and backgrounds as we look to build our team at Intel - from programmers to those with creativity, curiosity and those with great communication skills.”
He adds, “it's rare to find a 'data unicorn' that can do it all and we're not spending our time recruiting for such a talent. We build out teams to reflect a variety of backgrounds and experience which brings greater insight to our data analytics work."
What do psychographic profiling tools like Belbin team roles or Myers Briggs-type indicators have to do with such a techie conversation? From my perspective of creating and leading customer analytics teams for over 13 years, they represent two different mindsets when facing this recruitment challenge.
Taking a classic Myers Briggs approach focusses you on precisely categorising the individual into one of 16 potential personality types. Once they emerge from your analysis, categorised as anything from ENTJ (driven leaders) to ISFP (sensitive artists), you’ve put them in a box. You may understand them better, but you risk thinking of them as an individual and focusing on what they can do alone. That is rarely what is needed in today’s fast-changing collaborative businesses.
One of the benefits of the alternative Belbin team roles approach is that people are scored against all nine potential roles. Rather than seeing people as being a Plant (creative thinker), Completer Finisher (conscientious perfecter), etc, you see them as having strengths and weaknesses across all nine areas. That puts your focus on pulling together a team with complementary skills. The descriptions of such roles encourage that approach as well. As a Plant myself, I can attest to the greater power of partnering with a Completer Finisher on projects.
So, how does this apply to Data Science leaders looking to hire the talent they need? Hopefully, it serves as a reminder that, in many areas of business and sports, leaders have seen the power of assembling high-performing teams, not just star player individuals. Since Google first coined the job title data scientist, a range of associated job titles have started to emerge on the jobs market: data engineer, data architect, lead scientist, data programmer are all starting to reveal the need for supporting roles. Is it possible that your business requirement, even your data science requirement, could be best spread across a team rather than one specialist?
From an IT perspective, Maloy Manna shared on Data Science Central blog the benefit of data science teams including roles like:
Having worked with many of the UK’s leading financial services firms, for that commercial world, I’d stress the benefit of roles including:
However you see the skills needed to make up your data science team, I hope you can see the potential of complementary team members rather than that ideal “unicorn”.
My work with large analytics functions has shown how well this approach can work. Skilled coders complement statisticians, with both being keen to learn from the other. Adding individuals with strong business analysis and communication skills to more technical teams, or artistic data visualisers to teams with the other data skills, can strengthen and invigorate those teams. As Aristotle said, “the whole is greater than the sum of its parts”. The approach of Belbin team roles acknowledges that.
So, back to your recruitment challenge. Could it be time for you to assemble a multi-skilled teams benefitting from the synergy of different roles? They just might make more business impact than the ideal data scientists you wanted to hire in the first place.
Related articles: "DNA of a data scientist"