Big data starts with the fact that there is a lot more information floating around than ever before, and it is being put to newer usages. IDC thinks we will produce 44 zettabytes of data by 2020. However, organisations have made huge progress in tracking and analysing data. From management information systems (MIS) and decision support systems (DSS) to a vocabulary of business intelligence (BI), we are now in the world of Decision Sciences.
However, there lies a big challenge too. Organisations realise that data-driven decision making is not about adopting a new technology or hiring some analysts. It is about finding the right talent who can convert data into actionable insights. This gap is a much discussed topic. Some organisations hire experts, who know everything about the discipline — “Superman” model. However, this model is hard to scale and these experts are difficult to find. Other organisations are fixating on machines and automation as the way of scaling analytics — “Robocop” model, which is not flexible.
At Mu Sigma, by right talent, we imply a new breed of employees — a blend of business analysts, applied mathematicians, and programmers. But while most firms separate these into three distinct roles, we need them combined into one — the Decision Scientist. This is the “Iron-Man” model where the Decision Scientist — human is equipped with processes, technologies and platforms.
A Decision Scientist accelerates the journey from data to decisions. He has an acute understanding of the business and can define and solve business problems using data engineering, data science and decision science. To be able to do all this, a Decision Scientist needs to incorporate below fundamental shifts in his approach to solving business problems.
Learning over knowing: Rather than competing on knowledge that constantly breaks down, it’s better to compete on an ability to learn. However, the best approach will be to compete on the speed and adaptability of learning.
Fail fast, cheap and often: Decision Sciences industry demands innovation at speed. This in turn raises the cost of experimentation. If we can bring down this cost — throwing many, many darts at the dartboard within the same budget and timeframe, then the probability of success will rise dramatically.
Interdisciplinary perspective: In an environment of complexity, business problems don’t follow boundaries. For eg, an inventory problem at a manufacturer is not necessarily a supply chain problem alone. It will have roots in the customer ordering patterns, choice of customer segments, products mix that one offers, etc. Isolated approaches will not work now, an interdisciplinary approach is needed.
In today’s changing business environment, it’s all about harmonising data and decisions. Many firms or executives misguidedly end up prioritising data analytics over the need for better decision-making. With evolving customer demands, it is important to stay relevant. Data and analytics can help understanding these demands better. However, one must keep in mind that creative problem-solving is as important. It’s time for a cultural and a mindset shift — to understand that decision matters just as much as data.
Guest Author
The author is head of Products & Design at Mu Sigma
