The right skill setsWhile a majority of data scientists major in a technical field such as mathematics, statistics or computer science, a one size fits all approach will not suffice when appointing the best individual for the business.
The candidate will be required to possess an additional set of skills depending on the industry or role they will be filling. Typically, data scientists need to have a mixture of technical and “business”-related skills – they need to “number crunch” and explain what they have done (including the value of it) to the layman.
For example, a data scientist in the financial services industry would require some basic knowledge of how a business operates, in order to truly understand the business problems, they are hired to solve.
Alternatively, for an individual specialising in agriculture where data would predominantly be utilised to increase operational efficiency of farming and predict weather patterns – a more risk-return analysis, process improvement and agri-based skill set would be relevant.
Seetharam, says this is one of the reasons why there is currently a shortage of data scientists globally - do these elusive people even exist, and if they do, where can businesses find them?
According to IBM, the demand for data scientists will soar 28% by 2020, which is already of an incredibly high growth rate. The company further stated that by 2020, the number of jobs for all US data professionals will increase by 364,000 openings to 2,720,000.
Although South Africa is still lagging, we can expect similar trends in the future. It becomes clear that we will have the demand, but is there a supply of data scientists to choose from?
Hiring at the right levelA business that is starting out in its big data journey is better suited appointing a less experienced scientist who would collaborate with data experts within the organisation while refining their skills. While this may seem counterintuitive, the balancing act remains to find someone who is willing to learn, and skilled enough to make a contribution to your business.
Furthermore, given the challenges of finding the right talent, it would also be viable for organisations to invest in candidates at university level and groom them as future data scientists for their particular organisations.
Upskilling internallyBusinesses who find it challenging to attract data scientists externally should consider training existing employees who already have the advantage of understanding the business, provided they have some aptitude for quantitative-type work. Again, this is somewhat of a risky tactic, but given the skills shortage we face, it might be the most viable option for smaller businesses.
“Once the ideal individual has been appointed, the most important consideration for businesses should be retention, given the high demand for this rare skill set in the industry. In order for the data scientist to produce actionable insights and enjoy problem solving, they have to be invested in the organisation for the long-term,” concludes Seetharam.