By M.A. Kishen Kumar
One of the earliest chronicled usages of the phrase ‘big data’ has been traced back to a 1999 presentation by former SGI chief scientist John Mashey, titled “Big Data and the next wave of InfraStress”. Circa 2012, big data is the next wave of business opportunity.
Clearly, the availability of data itself has never been a matter of genuine concern. And evidently, we have seen enough value in data equity to commit resources to manage the increasing influx more efficiently. But unleveraged or even underleveraged data adds to business cost without contributing to business utility. So if data is at an inflection point today, it’s because we finally have the tools, methods and technologies to unlock new opportunities from it.
Big opportunity for banks: unstructured big data
Banking has always been a data-driven business. Not surprising then, that according to Gartner, it is the most active industry in making big data inquiries (25%), followed by services (15%) and manufacturing (15%). As consumer banking relationships get fragmented across multiple channels and as data volumes increase exponentially, banks will have to invest in big data capabilities to manage volume as well as to extract new value.
Banks collect data in petabytes, deep inside which lies the key to consumer behavior and preferences. The ability of the big data model to parse petabytes at a more granular level will deliver next generation customer analytics capabilities based on real-time insight and action delivery. This will reenergize the existing next-best-action paradigm and enhance customer relationships and revenue growth.
Banks have been maintaining their data in huge warehouses for some time now. Since this data is highly structured and historical in nature, the analysis thereof essentially provides us with a clearer view of the past. The difference that big data brings to the table is an ability to pick up streaming data from various sources as well as the ability to create predictive models and analyses on a real-time basis. As banks gain clearer visibility into the present as well as the past, they would be able to predict the future more accurately, just as VAR or Stress Test analyses do in the context of a bank’s portfolio. So, big data will complement existing data warehouses in a big way.
But, that’s only the tip of the iceberg. In my view, the real new opportunity that big data will deliver to banks is of unlocking the as yet unleveraged potential of unstructured data. For decades, IT has had the capability to mine structured and static databases; but it is estimated that 80 percent of the data held by the average financial institution is unstructured. Harnessing the functionality of big data in analyzing unstructured data elements like images, videos, documents and emails will enable banks to explore new dimensions in insight generation and decision making.
Will external social big data make internal banking transaction data less important, even irrelevant?
Proprietary data is almost always transactional in nature and, hence, historic. By enabling a more rigorous manipulation of these data sets to arrive at more refined insights, big data marks a definite step forward. But as a tipping point technology, there is much more that big data has to offer beyond just parsing enterprise data.
In a world becoming increasingly social, social business intelligence will be one of the most important determinants of competitive advantage in the future. Quite fortuitously, the highly decentralized structure and the sheer data volumes of the social ecosystem make it the perfect discipline for the big data paradigm. For data-intensive businesses like banking, social media represents the new frontier in data analytics.
Social business intelligence is about accessing and understanding what customers don’t reveal while interacting or transacting with their banks. It is therefore a brand new opportunity that can be leveraged to improve existing processes and build customer relationships. Insights derived from analyzing community conversations can add contextual relevance to communication and marketing strategies to address consumer expectations and perceptions. Social networks can also be a rich source of competitor information that can help inform a bank’s strategic decision making process. Banks can also benefit from using the crowdsourcing feature of social networks as a powerful tool to target product development at the financial lifecycle and future needs of its customer segments.
For example, a leading global financial services firm’s credit card program launch on social media did not get an overwhelming response but it was able to tap into the pulse of the community and use the feedback to restructure the offer accordingly. When a major international bank launched a new product, they were able to offer a combination of mass standardization and individual customization as well as target specific products at specific customers.
Will big data bring psychographics into the equation?
Going forward, the increasing integration of big data enabled analytic applications into social networks will provide banks with access to an entirely new range of qualitative insights into trends, sentiments and behavior. Now, having evolved their predictive models from the perfectly structured to the unstructured, from the purely transactional to the behavioral, there still is a lateral opportunity for the banking industry to push the data analytics envelope.
What if banks could look at key customer segments through a lens that focuses beyond their transactional or in-category behavior? What if product development could be influenced by a customer’s worldview or personality type rather than age, income, past history or current behavior? What if psychographics is the key to creating the most perfect targeting model yet?
Today it is possible to subscribe to data available in the marketplace from providers like Yahoo or Google, which can then be mapped to key customer segments.
Using big data analytics it is possible to do a psychographic analysis of the data to understand customers and prospects from a point of view of their interests, activities and opinions. The patterns emerging from this analysis can then be used to inform the product design and delivery process to ensure that the final offering suits the customer’s lifestyle and psyche.
In wealth management, for example, the choice of solution would be defined by the extrapolated risk appetite of the customer rather than by the conventional approach of assigning debt or equity based on age. Even credit score calculations, currently derived from purely quantitative considerations like transaction history, payment behavior and income, can possibly be reassessed from the additional perspective of a customer’s unique behavioral characteristics and traits.
The inevitable challenges
In terms of technology, big data has the capability to handle both data volume and real time data management to move the practice of data analytics far beyond the merely transactional. But technology can only enable the process up to a certain point. The effort requires a multidisciplinary approach, involving skills from the social sciences, psychology and ethnography, to name a few. It is these analytical skills that will help identify, isolate and articulate insights in the patterns delivered by technology. And since these have, thus far, never been the skill sets associated with the data analytics space, talent acquisition will be the biggest challenge in realizing the full potential of big data. Banks and other enterprises keen on leveraging this opportunity must gear up to fill a big demand for investigative and diagnostic data professionals, or “data doctors”, if you will.
M.A. Kishen Kumar is AVP, Consulting & Systems Integration, Infosys Ltd.