Not all data scientists are created equal
There are two common misperceptions about data science that exist in many organisations today, and which are holding them back from unlocking the full value of the discipline for themselves and their customers. The first of these is that data science is all about algorithms, big data, artificial intelligence and machine learning. And the second is that all data scientists are created equal.
Allow me, at the outset to dispel these myths. Yes, data science does harness all those seemingly mystical components like algorithms and AI to deliver the outcomes required of it, but ultimately, the field and its practitioners exist to solve problems.
The second misperception is possibly the one with the potential to do the most harm, both to the business and to the data science practitioners it employs. The field of data science is incredibly broad and encompasses a massive array of people. Most are highly trained and very intelligent. But that doesn’t make them all the same. In fact, the value of data science lies largely in the diversity of skills, personalities and temperaments that so-called data scientists can bring to an organisation. And understanding this simple fact – that not every data scientist is the socially dysfunctional, back-room-based tech-head, that Hollywood would have us believe – is key to any organisation’s ability to unlock the full potential of data science to solve its business problems.
In fact, understanding this true nature of data science, and the diversity of those who practice it, is also key, in the first place, to actually establishing what many of the problems are in the business that you need them to solve. That’s because, as with every professional field, different data scientists have different ways of seeing the world, and different ways of approaching what they do.
Some have a keen understanding of, and insight into, the nature of people and the way in which their particular field of science might enhance the experience or solve the problems of the customer.
Others have the ability to see the big systems picture, and love nothing more than to use the technology and data at their disposal to engineers solutions that are 100% business or processes efficiency focused.
Putting any of these scientists or engineers in a role that might require their skills, but doesn’t fit their mindsets, personalities, or non-technical strengths is effectively reducing their efficiency and diminishing the full value that they have to offer the business.
Rather, the key to consistently value-adding data science is to understand and define the specific problem your business needs solved and then put the appropriate data science specialist, with the relevant technical and ‘soft’ skills to work on solving that problem. So, it is unlikely that pulling a data engineer with a great head for systems design, but no real interest in the day-to-day volatility and changeability of customer service, into a front office position is going to deliver the results you need or solve the customer experience challenges you’ve identified.
Likewise, shoe-horning a data science practitioner with a real passion for solving customer problems, into a back-office role where they have no part to play in identifying customer challenges, and never get to actually see how the work they do positively impacts on the customer experience, is going to ultimately result in the slow death of any passion or creativity that individual could have brought to your customer culture.
Of course, for many businesses, this vital need to define the problem before employing the correct data scientist to solve it can be something of a Catch-22 situation. That’s because, clearly defining a problem is often one of the hardest parts of solving it. And if the solution is going to require the specialist skills of a data scientist, it’s likely that a data scientist will need to be part of the process of figuring out the problem in the first place.
The solution to this obviously cannot be to go out and employ the data scientist you think you need and hope that they end up being the right fit once the problem is pinpointed. Rather, working around, or avoiding, this Catch-22 situation is as simple as ensuring that your business understands, from the outset, the importance of having a fully diverse data science team on board. As far as your budget will allow, this team should be populated with data science specialists across as many fields as possible, but also who have diverse personalities and preferences in terms of the business focus areas in which they prefer to operate.
Prioritising the establishment of such a team, even if it is as small as two or three individuals initially, is the ideal way of ensuring full alignment of the technical, analytical and business requirements for effective problem solving through data science. It also ensures that when there are problems that need to be identified, with a view to bringing specialist data skills on board, you have the professional data-based input you need to avoid making the quintessential mistake that so many organisations still do. Which is to respond to a business challenge with the words “We need to bring in a data scientist.”