The Instant Intelligence of Big Data: Improving the Analytical Value Chain

<< back Page 2 of 2

Big data analytics has endowed banks with the capability to identify risk triggers in real-time, across multiple channels, and take the appropriate steps required to then protect their customers – again in real-time. It also enables the detection and prevention of account takeovers and insurance fraud  possibilities that resonate well with the inherent risk intolerance of the banking community. That being said, the scope of big data extends far beyond fraud and risk management; it has the potential to generate new opportunities in product-line innovation, enable proactive decision making, identify new levers of profitability and create seamless omni-channel banking experience for customers.  

The capability of big data solutions, then, to handle data in larger volumes, wider varieties and increasing velocity is no doubt a significant evolution from conventional business intelligence systems that discount unstructured, streaming and even a great chunk of non-enterprise data. But its greatest disruptive value lies in its ability to enable actionable insights in real-time and deliver analytic value to time-critical applications, processes and businesses.

However, the disruptive value that can be derived from any big data program is largely a function of the versatility of the underlying solution. It is therefore absolutely critical for businesses to deploy solutions that are enterprise-wide and scalable, support low-latency datasets and offer comprehensive data compute and built-in analytic capabilities – all underscored by the emphasis on real-time decision making.  Discover (data), develop (insights) and action (decisions) are the three key stages of the analytics life cycle. Accordingly, analytics solutions must be evaluated along these parameters to ensure that they fit enterprise needs for data aggregation, data manipulation, decision making and execution. And that is where the men start to separate from the boys.

Conventional Point Solutions 

Conventional point solutions have the ability to discover data in real-time across internal systems and external sources and perhaps a limited number of structured and unstructured datasets – which is mostly just analytic hygiene. But they take far too long to integrate new sources of data: research says that best in class companies take as long as 12 days to do this; the others could take five times as long! It’s the same thing all over again with insight generation. Business analysts, who must typically depend on technical staff to build a new solution, are cramped by the long times to insight. The limitation of having to work with standard visualization dashboards further compounds their problem. Finally, insight without timely action is of little value. One of the biggest challenges that enterprises encounter with their conventional solutions is that they lack an efficient mechanism to execute insights in real time. Part of the problem is that all the stakeholders of a collective decision rarely make it to the same place at the same time.

The good news is that these issues may soon be a thing of the past. Emerging analytic platforms provide features that overcome the weaknesses of conventional solutions by enabling enterprises to quickly develop and leverage industry-specific big data insights. For instance, enabling users to configure data filters can accelerate the aggregation process and ensure that only relevant information is extracted from every source. This enhances development productivity as well as accelerates the process of data source integration. And a visual approach to building transformations – complete with preconfigured components and predefined rules – can reduce a business’ dependence on technical resources.

Time to insight can be reduced dramatically with inbuilt algorithms which business users can leverage to perform text analytics, sentiment analysis, and more. This means operationalization of insights – quickly, efficiently and through self-service. Additional key capabilities for analytic platforms include the ability to integrate tightly with enterprise business processes including operations, line of business and support functions. Most of all, real-time collaboration among decision makers, across functions and geographies, is critical in order to facilitate faster insights and better decisions, which can then operationalized in real-time by an integrated workflow.

It has been established that big data has the potential to disrupt and transform the practice of business analytics as we know it. But the scale of disruption and transformation is entirely dependent on the choice of big data solution and its relevance to an enterprise's business. Because, without the right solution, there is always the possibility that enterprises may well leverage big data and yet have nothing more than a newer rear view mirror. 

<< back Page 2 of 2