Big Data in Banking – Demystified

There is a lot of discussion about big data these days, and along with it, the 3 V’s; velocity, variety, and volume. But I get two major questions from bankers I talk to on the subject. First, what does this really mean to the day-to-day operations of banking, and second, what should I do about it? Although each question is complex and a discourse in itself, let me take a few moments to begin to answer these questions.

The three V’s of big data are abstract technical concepts, but when translated into the customer experience, they are all very important. I’ll discuss each one briefly.

Velocity

Data moves fast. Each client at the bank is generating thousands or even hundreds of thousands of data points a day. From mobile location to transactions to human interactions, you don’t know how many of those data points are actually relevant and you don’t have a lot of time to decide. Add those to the millions of external data points such as stock market prices, interest rates, macroeconomic data, and competitive moves; and you need the right infrastructure to make sense of it all.

Big data systems are designed for this kind of velocity – analyzing the data quickly, sometimes in memory and sometimes while discarding the streaming data after analysis. What do these data points mean to the customer experience? It depends. You need the systems to decide and react quickly when new data becomes available.

Variety

Data comes from everywhere. And of course, it doesn’t look the same from source to source – or even from moment to moment. Social media, transaction data, text notes from call center reps, and even in some cases, audio and video signals can provide valuable information about your customers.

Of course with a wide variety of data sources comes a wide variety of data types. And in some cases, that data is unstructured. Big data systems allow you to capture this unstructured data and create structure from it. Using text analytics, you can get an additional, valuable dimension to help classify your customers’ behavior and react appropriately.

Volume

One thing is clear about big data – it is big. And growing. The amount of data we are storing about our customers – not using necessarily, but storing – is huge. Discussions of petabytes, zetabytes, and even yottabytes are now common place. Big data applications allow for this storage and analysis.

In order to understand your customers in a way your competition can’t, you need to capture this data. Big data analytic techniques include sophisticated variable reduction and clustering algorithms to help make some sense out of the potentially hundreds of thousands of variables generated by modern data systems. In the coming years, banks that can tap this data and use it to make better decisions will have the competitive advantage.

Big data is here to stay. Of course, it’s critical to plan big data technology acquisitions to focus on making better decisions to improve the customer experience. It’s not just a technical discussion – it’s now a mission critical part of your technology architecture.