Over the last decade, businesses have undergone a transformation. The dynamic landscape has given rise to ill-defined and continuously shifting business problems. For example: how are my customer preferences changing? Who should I target? What is the right pricing model for different products?
In order to address such problems, organisations across industries are shifting from a knowledge-based to a learning-based approach. They can no longer rely on gut or past experience. They have to rely on the latest available information to infer and learn from, before making decisions.
This trend has been enhanced by the data age, where organisations are progressively looking to leverage data and institutionalise data driven decision making.
It is for this reason that many have gone on to term data as the economy’s new oil. However, as the volume of data grows, organisations will need to democratise the use of analytics and make data driven decision making intrinsic to their culture.
Ironically, even with so much data available, its true value has not been fully unleashed. One reason for this could be the severe talent shortage in the analytics realm. Research Organisation Gartner predicts that by 2015, Big Data will create 4.4 million jobs globally. Another alarming projection suggests only one-third of these positions will be filled. Make no mistake; we are not referring to people who at the core possess just one kind of skill. Rather, we are referring to people who are able to cull meaningful insights and findings from chunks of disparate data and information sources to enable better decisions.”
Potent skills ‘Decisions’ is the operative word here. Although there is a lot of hype and talk around Big Data, organisations need to focus on the journey from Big Data to Big Decisions.
What is imperative for companies to understand is that as they embark on the journey of making data-driven decisions — even the typical skills associated with the now popular term ‘Data scientist’ may not be sufficient.
While a lot has been written about data scientists, we believe that the skills that they bring to the table need to be augmented. Just data, math and technology skills are not sufficient. One needs to take an interdisciplinary approach comprising a potent combination of skills such as applied math, business acumen, technology, design thinking, behavioural sciences as well as the ability to work with people. In this regard, what organisations really need is professionals from ‘Decision Sciences’.
Whether you are an organisation with an in-house analytics team or an analytics provider; having the right analytical talent is fairly mandated. By coalescing a deep understanding of math and technology, data scientists create models to generate insights.
But just the understanding of statistical techniques / methods and technological prowess would not result in uncovering meaningful and insightful correlations. They also need to be able to bring business perspective and define problems correctly; using consultative, principle based thinking to provide necessary hypotheses — something only a decision scientist do.
In that sense, “building” such a talent pool should always take precedence over just acquiring it from outside.
Acquiring the raw talent is just half the battle won. It is imperative to equip this raw talent with necessary learning that is continuous and holistic.
Organisations currently possess the analytical tools, techniques and technologies with capabilities getting enriched at an astounding pace.
Hence imparting this learning to employees via a combination of formal and informal trainings, conferences and seminars and then testing the learning will help create a robust analytics ecosystem of the future.
In essence, decision scientists have the right intellectual current. They have the ability to unify logic and emotion.
Beyond number crunching Decision scientists need to be professionals with not just the right skill-set and mind-set, but also the right business understanding and context to develop insights and facilitate analytics consumption. This understanding and orientation to business can be achieved through exposure to real-world business problems and hands-on project experience.
In addition, it is imperative that the decision scientist engages with multiple stakeholders— in a consultative fashion — to drive discovery-driven analytics and agile experimentation, while having an interdisciplinary focus. The business analyst, the applied mathematician and the programmer need to come together in one individual.
Cross-industry and cross-domain learning can provide a fertile ground for innovation. Conventionally, specific analytics practices or techniques, from specific business functions, were used to derive meaningful insights. This practice is not in vogue today as companies are quickly learning the ropes of collaboration and cross-pollination of ideas. For example, yield optimisation, the predominant analytics technique of the airline industry, can be used across the online advertising industry; survival modelling techniques for life science can also be used in financial risk analytics.
Hence, analytics companies should build an institution that trains and acquaints raw talent with business operations across industry and domains. In sum, companies that are ready to build and sustain data-driven decision making should ensure that their Decision Sciences workforce possesses certain key traits.
The writer is with Mu Sigma, a decision sciences and analytics firm