Banks sit upon vast reserves of customer data, the likes of which would make even Google salivate, but in the early years of the big data hubbub, traditional depository institutions at first struggled to get their arms around the concept. Now, by zeroing in on one or two very clear and narrow goals, banks are learning to make big data work for their business.

It hasn’t come easy, though. The sheer scope and magnitude has been one significant hurdle. Even the name itself, “big data,” sort of implies that putting it to work is going to be a large and unwieldy process – beginning with creating relationships between various data points.

“What we’ve seen a lot is the data sits in different spots, so banks have access to a lot of data, but many times it’s in different places, so it’s hard to create the relationships between the data,” said Gerald R. Gagne, a member of the Boston-based firm Wolf & Co. Data integrity is another issue, he said, and so is data warehousing.

But as banks have begun to work through some of those obstacles – and as vendors have emerged to help them do so – financial institutions are beginning to deploy big data into practical applications within the industry.

Those applications usually tend to take shape under one of three broad umbrellas: revenue generation, cost reduction or risk management, said Phani Nagarjuna, the chief analytics officer of Sutherland and the CEO of the data analytics firm Nuevora, which was recently acquired by Sutherland.

In particular, he sees an opportunity for banks to use big data to enhance the customer experience, helping to better inform bankers about how customers might like to be contacted and what kinds of products and services they might actually need.

“When you really look at knowing which is the next best product to introduce to a customer it’s a great opportunity for banks,” he said. “Growth is not going to come to the banks from external factors. Growth is going to come from innovating and building up your footprint.”

Small, Quick Wins

The key to making big data work for your bank is starting with a narrow vision of what you hope to accomplish with that information, said Eric Esfahanian, senior vice president and general manager at the Boston-based Gryphon Networks.

“I think the biggest caution is moving towards the light of big data without a very clear understanding of what they hope to accomplish with it,” he told Banker & Tradesman. “Everybody’s drawn to the light of big data and analytics, innovation, disruption, all those great terms, but the reality is, you have to use and apply this concept of big data in a very narrow way at first in order to figure out, What is my ROI?”

Gryphon launched in 1998 as a software firm that initially helped companies comply with “do not call” lists, particularly where their non-centralized or remote sales staff might be concerned. The firm had to keep a log of every call made through its system essentially until the end of time because any of its clients could be subject to auditing. The company realized that a few of its clients were using that call-level data to better understand what their sales reps were doing in the field, what was working and what wasn’t.

Today, that kind of activity is Gryphon’s bread and butter, and the company services about a dozen banks, including the $1.9 billion Belmont Savings Bank, in doing just that.

Gathering that call-level information allows the bank to better determine which calls and what kind of behaviors were resulting in customers’ following up with the bank and making appointments, said Morgan Cambern, director of retail operations.

“There is so much data to mine in there, that’s part of the reason we started to focus in on a few very specific things,” she said.

Or as COO Hal Tovin said, it’s about catching people doing the right thing – as opposed to doing the wrong thing – and then reinforcing that behavior throughout the organization.

He also said, “I learned a long time ago, start with a very focused goal and expand from there, versus trying to be everything to everyone.”

One of the major takeaways, particularly for community banks, might be not to dive into the deep end of the big data pool – start in the shallows.

“We say, ‘focus on small, quick wins.’ Don’t buy everything at once. Start very small and very narrow and once you get a feel for the toolset, the key performance indicators and the results, over time, you will want to get more sophisticated,” Esfahanian said. “Retail banks especially are particularly susceptive to this because in many cases they are tech laggards, so when they finally get religion on this, they feel like they’ve got to move fast.”