Gone are the days when an MBA degree was your ticket to the corporate world. Basic knowledge of marketing, HR, finance, macro and micro economic models, logistics and supply chain management, analysing a simple data set and reading case studies did guarantee you a job earlier! The ability to present this limited data set in the form of attractive power point presentations mostly acted as the “cherry on the cake”.
So what has changed in the present scenario? Nothing and everything! We are in an era where data is the new oil of digital economy. Today’s world requires deep domain expertise integrated with soft skills. The digital era, with an overflow of data, requires a professional who can collect and organise the data in a meaningful structure, analyse it through smart algorithms and arrive at a business/ social solution.
Recently, at the launch of the Manipal Global Academy of Data Science, a senior colleague from an insurance multinational talked about how they are piloting an algorithm to not only reduce the time to settle vehicle accident claims for customers but also costs and human error significantly by shortening the claim-survey cycle. All that they did was mine their images repository and presto all new images could be analysed without the need for a surveyor. The end result: delighted customers, quicker settlement time and substantial amount of cost saving for the company. This is the power of data science!
So what is data science? It is the science of transforming data for decision making. It is an iterative and non-linear process that involves the following stages:
Stage 1: Asking the right questions — this is the most critical role in data science. The training as a business analyst adds a lot of value here.
Stage 2: Collecting and structuring the data — you need basic computer science skills to trawl the web, query databases and clean the data. There are multiple tools to do this and it helps to train on these tools.
Stage 3: Exploring the data — it is essential to get familiar with the data, identify anomalies and patterns and develop a hypothesis. Common sense and an understanding of the domain helps.
Stage 4: Modelling the data — competencies that are used to model the data are applied statistics and machine learning. So if you didn’t like stats, it’s a good time to start liking it now!
Stage 5: Communicate the data — it is important to visualise your data and tell the story. You need training on the various visualisation principles, tools, good presentation skills, and oral and written communication skills.
Stage 6: Implementation — of the solution arrived at is something a business analyst enables. So professionals who can’t engage with stakeholders to implement the solutions for the betterment of the business are ineffective as data science consultants.
Stage 1 & 6: From conceptualisation to implementation — the data scientist is a person who understands the domain (banking/ healthcare/ retail, etc.) well and has the knowledge of various insights for framing the problem correctly. From asking the right questions to implementing the business solution, a data scientist has to know it all. Domain understanding, overview of business linkages, etc.
Guest Author
The author is senior vice president, Manipal Global
