Enterprises are leverage data science to derive actionable insights from data. But challenges abound
Data science is a recent discipline, one that enterprises are trying to embrace since the Internet became ubiquitous and social media triggered an explosion of data. In the earlier days enterprises relied on the data that was mostly structured and resided in their data centers.
But it is only over the last few years that they are realizing that unstructured data that lies in the public domain can provide them customer insights, if extracted and analyzed properly. That’s when Big Data became the buzzword and enterprises started deploying software that helped them extract and analyze all data, both within and outside their data centers.
The way data science is understood today is that it is the ability to extract insights from both structured and unstructured data by using analytical tools as well as machine learning and deep learning technologies.
Data science is expected to help enterprises mine for insights that can help them increase revenue and profitability, operational efficiency, productivity while enhancing customer experience.
Data science is not merely about deploying technology. Data scientists will have to understand statistics, econometrics and technology. By combining these skills they are able to derive value out of data. Data science is not about merely analyzing data. It’s about predicting future outcomes that can positively impact a business.
While both data science and Big Data seem to mean the same, there is a difference. Big data refers to the vast amounts of unstructured data that mostly lie outside an enterprise’s data center. But data science is an approach that can be applied to even smaller data sets irrespective of whether it is structured or unstructured.
Deploying data science in an enterprise requires data scientists to understand the flow of information across key functions or departments, identify and classify various data assets and then make sense out of the data for predictive insights. Many banks today are deploying data science to identify frauds. Retailers are using data science to anticipate demand, guage customer behaviour and make their supply chains more efficient. These are essentially enterprises that gather and store quality data.
In the Indian context many enterprises are still in the process of cleaning data and those with clean data are still in the early stages of deploying analytics. Enterprises that are ahead in using analytics and deploying data science are mostly in e-retail, food tech and e-aggregator businesses. But moving forward, as competition intensifies from disruptive, new players, enterprises across sectors will be forced to turn to data science for increased efficiencies, agility and customer insights. This will create a massive demand for data scientists not just across the globe, but also in India.
Globally there is already a massive demand for data scientists. A US survey predicts that the demand for data scientists will go up by 40 per cent by 2020.
According to the survey India today has only 10 percent of data scientists available globally. As a result Indian service providers catering to the global markets are losing out on deals where they could have benefited out of cost arbitrage. As more Indian companies deploy data science we are likely to see an acute demand supply gap in this space.