The massive data outbreak we have known over the past decade naturally gave birth to a new field of methods and processes, called data science. Data science is a multidisciplinary subject, where come together mathematics expertise, business acumen and technology, and aims at solving complex problems thanks to analytics.Indeed, nowadays businesses accumulate infinite quantities of data as well as Big Data; but without a professional expertise, without the right knowledge and techniques to extract valuable insights from it, it remains untapped and useless. This is why the profession of ‘data scientist’ developed quickly over the past years, and is very sought after D.J. Patil and J. Hammerbacher—back then leading data and analytics efforts respectively at Linkedin and Facebook— coined the word in 2008. Later on, D.J. Patil and academic Tom Davenport in a 2012 Harvard Business Review article will state that this is going to be the sexiest job of 21st century. But what is exactly a data scientist doing all day?
Using data in creative ways to generate business value
Data scientists juggle with both structured or unstructured data and a wide type of databases, understanding how to fish out answers to complex business issues from them. Some people see them as a combination of “hackers”, analysts, communicator and advisers. Companies need them when they want to crunch large amount data, or when they don’t exactly know what to do with their operational and customer data. As a business, you can gain great value from hiring a data scientist: he or she may, as Simplilearnexplains, empower management to make better, data-driven decision, but also identify new opportunities or refine target audiences, for instance.
One of the great asset of data scientists is that, after digging deep into data and extracting the value, they can communicate it. Through graphs and charts, they visualize the data and transmit the information the best way possible. Using modern self-service BI tools will help greatly in combining all the data coming from various sources (web, documents, databases), leaving more time to the data scientist for the sexy part of the job —analysis. Once the analysis is done, these business intelligence solutions assist in creating compelling business dashboards that will enable the data and the scientist tell the whole story behind it.
How do I become a data scientist?
More and more universities are opening data sciences curriculum; but you can also start as a self-learner with online compilation a of podcasts, books, and open-source.
However, data science is also a mindset. To succeed, you need a very wide set of skills, as we have seen earlier: being an analyst, a communicator and an adviser at the same time, while being at ease with technology. For that, data scientists have a lot of curiosity -the one that will push them to always ask new questions, to make new discoveries and learn new things. The same curiosity applies when they dig into the problem to understand it in its core, and break it down into a set of hypotheses subjected to test. That requires motivation and determination because it is often an autodidact path: indeed, data science is so interdisciplinary that it works with techniques and theories coming from statistics, computer science, data mining, visualization, predictive analytics, maths, and many more. Even if universities want to make the best study cursus possible, they cannot prepare student enough to what they will have to deal with in the “real world”. Interestingly, and that also might explain the data scientist shortage businesses face, this position is really learned on-the-job. Maybe an interesting approach would be to create the data scientist from within, and empower every employee with the right data culture and tools.
Which future for data scientists?
SHRM released 2016’s list of hardest jobs to fill, among which appears without surprise data scientist. The massive explosion of data and the equal need to capitalize on it make this special profile very demanded and scarce. A McKinsey Global Institute study even predicted that 1,5 million data managers will be needed by 2018 in the US, and the demand today has raced ahead of the supply. But the hype for this position is just starting, as it was ranked number 1 out of the 25 “Best jobs in America” ranking by Glassdoor in 2016, which promises a certain fortunate future.
In parallel grows the trend that exist already in many other industries where automated systems and machines replace humans. This is a general fear but the sector of data is naturally more subjected to these concerns. As we talked about earlier, many software today exist to help data scientists in doing their job, or opening this possibility to non-technical people to create customized graphs and charts that convey their message. However, we have witnessed over time what a Cyber Trend article stated:“when man and machine work individually they are both inferior to man and machine working together”. The human element remains fundamental for data science in the future.
Data science still has its best days to come. It might take some time for it to be anchored as older jobs like lawyer or doctor, but it is not a hype that will fade away. It will evolve and change at speed light, like any technology. But data science is here to stay, adding value to businesses, digging deep to get insights that would otherwise remain untapped, and help management and governments to make better informed decisions.