Credit unions aren't always aware of the immense amount of data they have in their various systems. But more important, they need to know how to make their data useful.
That’s where data analytics comes in, says Dean Nolan, vice president of products and marketing at Saylent Technologies. He defines data analytics as the deep analysis of information credit unions already have on hand.
“Good use of data can increase member satisfaction and help credit unions reach their marketing goals,” Nolan says.
But that simple definition can trigger simple assumptions. “Many credit unions typically will think of applying data to direct mail promotions,” says Dan Lozier, senior director of client relations at The Members Group. “The question for them is, ‘How can we make all this data useful?’ ”
He says credit unions haven’t been as sophisticated as big card issuers in churning through data to serve consumers in the most effective ways.
“A key to that is understanding consumer behaviors, as well as the more traditional demographics such as age, sex, and income,” Lozier says.
“You need to ask where transactions occur: Does a member use his debit card primarily for groceries and gas? Does she travel or order online a lot? You have to reach each member differently with unique offers that apply specifically to them.”
“Analytics not only involves getting a better grasp on member data, but also analyzing your approaches to them,” adds Kim Brackhan, vice president of sales for Experian. “What worked and what didn’t? You need to look at members’ activities and behaviors to determine what to offer and the best time to offer a loan or other product.”
She says the data source mix for all of this is vast—loan application data, demographic information, credit data and scores, and behaviors. “Within those sources are characteristics and trends that often go unnoticed.”
Trying to gain efficiencies through data analysis is a big consideration: The ability to make quick decisions with less manpower adds to the bottom line, Brackhan says.
‘Get the loans that got away’
The first questions IQR Consulting asks credit union clients that inquire about data analytics, says Karan Bhalla, the company’s managing director: What problems are you facing and what do you expect from this service?
He says credit unions can use data analytics to address concerns such as slow membership growth and underperforming loans. “Once we determine a client’s needs, data analytics has three steps: collect the data, convert it into usable information, and apply the information to concrete actions.”
IQR uses both “on us” data—the credit union’s own data, such as loan originations, payment histories, and other member activity— and “off us” data the company supplies, such as credit bureau data, general economic information, and seasonal trend reports.
“In most cases, credit unions have more data than they think they have,” Bhalla says. “We help clean and better manage that data. We want to make sure that the analytics process is based on good information.”
Lozier says some credit unions approach The Members Group because of flat-line performance in certain product categories. “Many times the problem is that they have continued doing generic promotions that simply no longer work. They need to move on to more targeted, sophisticated approaches.
“The right offers and timing lower costs because you spend less to reach the most receptive people,” he continues. “With a more targeted approach, the potential return is much higher due to an increased response rate and the use of fewer marketing resources.”
Key to determining the most receptive members is intensive segmentation, “which is more than three or four general categories, such as age, sex, income, etc.,” Lozier says. “There are segments within segments, and different behavior patterns exist within each. You’re looking for two main bits of information: Who are the right people to market an offer to—and who are the wrong people?”
“Big data” describes the exponential growth, availability, and use of information. It can serve as the basis for innovation, differentiation, and growth.
It’s important to focus on the ever-increasing volume, variety, and velocity of information that forms big data. The challenge is in determining the relevance of this data and creating value from it.
Contributors to big data, according to the 2013-2014 CUNA Environmental Scan, include:
Nolan offers an example of how appealing to the right segment can work. “We helped one California credit union track card users who were doing less than $250 in signature transactions each month. It offered these members $20 in gas vouchers if they would increase their signature transactions to $500 or more per month. Card users became habituated to increased use, and the credit union saw a 93% increase in card use that was sustained for more than 90 days after the end of the campaign.”
Data analytics vendors understand credit unions are concerned with credit risk. “One credit union came to us and said, ‘We want to offer more loans to more members, but safely. How can we do this without running substantial risk?’ ” Brackhan recalls. “We looked at their data and determined they really couldn’t [lend any deeper]. So we looked at loans they had previously approved but not booked, and found that those loans were performing well.
“Our advice was to go after those types of loans and book them,” she continues. “Don’t lend deeper; get the loans that got away.”
Know what you want
What keeps credit unions from exploring data analytics? Lozier cites two main culprits:
Bhalla says that when a would-be client cites a lack of resources for not pursuing data analytics, “We suggest breaking a big project into small components. By solving a small problem, it reassures a client, who is then better able to take the next step.”
Sometimes, clients simply believe what has worked in the past will continue to work in the future. But that’s not usually the case.
“Say your credit union goes from 5,000 to 10,000 members due to successful marketing efforts,” Bhalla says. “Now you have a different set of problems and concerns.”
Whatever the objective, Lozier says, using data analytics correctly is a continuous process. “Say you run a targeted campaign directed at a specific group of members. You measure results by which segments produced the greatest response, then incorporate what you’ve learned into the next campaign. You constantly refine.”
Sometimes the proof in the data analytics pudding reveals itself in member responses, he says. “We’ve seen member comments like, ‘You knew what I wanted before I asked for it,’ and ‘I never thought of 'X product' until you suggested it.’
“The idea is to stay ahead of people, no matter what their stage in life,” Lozier continues. “Data analytics isn’t magic. It’s a disciplined, scientific process that not only can tell you who to market to and how, but also help you identify risky accounts so you can prevent loss.”
But Nolan cites two bad practices that using data analytics can invite: Overestimating or underestimating its effects. “Overestimating means taking a ‘boil-the-ocean’ approach that seeks solutions to every marketing or membership problem a credit union has. Underestimating occurs when a credit union seeks a one-size-fits-all solution that really doesn’t meet its unique needs.”
Credit unions need an “ask” when pursuing data analytics, Brackhan says: “a goal, a plan for implementing it, and a team to carry it out. Typically, what may deter a credit union from pursuing data analytics are concerns about resources and time, and the ability to prioritize objectives. So build your plan ahead of time.”
PATRICK TOTTY is a freelance-writer based in Larkspur, Calif.