How To Become A Data MacGyver (Mullet Not Included)

Credit unions are mining myriad informational resources to peel back the mystery surrounding member behaviors, wants, and intentions.

 
 

Big data has made big headlines over the past couple of years. Much has been made of the idea that — with an endless stream of information and enough computing power — analysts can establish links and draw conclusions for just about any question imaginable.

But from a practical perspective, data collection should be less like trawling with a net and more like fishing off a pier. Sure, there are millions of fish out there, but which ones are you after? How many do you need? And what are you doing with the ones you do catch?

CU QUICK FACTS

HERITAGE FCU
data as of 12.31.13

  • HQ: Newburgh, IN
  • ASSETS: $454.3M
  • MEMBERS: 45,113
  • BRANCHES: 7
  • 12-MO SHARE GROWTH: 5.22%
  • 12-MO LOAN GROWTH: 6.88%
  • ROA: 0.67%

These were the type of questions Heritage Federal Credit Union ($454.3M, Newburgh, IN) struggled with for years before adopting a succinct data strategy in 2007.

“Our leaders and staff have always been focused on making decisions that represent and work well for all of our members,” says Steven Bugg, the credit union’schief marketing and member service officer, speaking on a 2014 Callahan & Associates webinar.“So we started first with a focus on the membership before we got into gathering data and working through a cultivation of our culture.”

Once the credit union did begin to prioritize data, it soon realized it already had a wealth of internal and external information at its fingertips, including third-party membership surveys and institutional performance comparisons, core and ancillary system resources, credit bureau reports, and even some demographic and lifestyle data.

Standard services and activities like bill pay and ACH began to play dual roles by allowing the institution to see whether members were making mortgage, auto, or credit card payments with other institutions and use specific campaigns — such as targeted messages through online banking — to go after them. And the addition of a marketing customer information file (MCIF) system in 2009 is helping employees do a better job of recording member notes and other internal findings. 

“It doesn’t cost a lot to get this information,” Bugg says. “In fact, we found a lot of it was already at our fingertips.”

For Heritage, the challenge was not in collecting data but in taking data from different sources and effectively weaving it together into narratives, which the credit union could then use to better understand its positioning and inform future goals.

Lesson 1: Face Value = No Value

After adopting its data strategy in 2007, surveys quickly became a main staple of Heritage’s analytic toolbox, Bugg says.

“If we can hit on all of our members’ desired service attributes, we know we’ll drive more of their future business on the loan and deposit side,” he says. “But there’s always two sides to these stories relative to who completed the survey.”

Variability in responses can be an advantage when trying to understand the priorities of a certain group. But to obtain a holistic perspective, it’s important to use multiple data sources to ensure a more accurate median. For example, Heritage uses a third party to collect approximately 400 member surveys per month, each of which ask respondents to rank the credit union on a scale of one to five according to seven core considerations:

  1. Were you treated like a valued member?
  2. Were you helped in a timely manner?
  3. Was your transaction processed accurately?
  4. Were you offered additional products and services?
  5. Were you thanked for your business?
  6. Were you provided with overall quality service?
  7. What is the likelihood you will recommend Heritage to family and friends?

Heritage also uses another third-party to perform quarterly mystery shops not only with its own tellers, platform staff, and call center agents but also with competitors within a five-mile radius.

And although its data collection sources vary, collection methods are similar enough to inform one another.

“The scoring criteria for both our survey and mystery shops is similar, which ensures we can use all of this information to drive our decision forward,” Bugg says.  

Lesson 2: Ask Why, Not What

Credit unions should avoid constricting themselves solely to the traditional financial measurements, groupings, and terminology that regulators, peers, and others rely on. In many cases, these represent the endpoint of a data mining strategy, not the beginning of one.

Heritage’s membership growth is one prime example. This metric is up 4% year-over-year as of fourth quarter 2013, according to Callahan & Associates’ Peer-To-Peer analytics. But according to Heritage’s own research, only a small percentage of those new members joined because of the credit union’s TV advertising campaigns while a full 47% joined because of referrals from friends and family.

“For us, visibility is king,” Bugg says. “We learned to understand our convenience factor [while looking at data].”

In fact, the credit union determined approximately 73% of its members live within five miles of one of its seven branches, with the exception of those in its indirect program, and roughly 14% of new members joined after driving by one of these locations.


Lesson 3: Segment Your Membership

Data analysis has also allowed the credit union to divide its membership into separate groupings based on their locations, behaviors, and priorities (seeking low fees, etc.) as well as their profitability. Once it identified those categories, the credit union was able to establish its own benchmarks and thresholds for each.

“The cookie cutter mentality is gone,” Bugg says.

For example, about 9.8% of Heritage’s members fall into what it classifies as the credit-driven segment, 18- to 34-year-olds with an annual income of $50,000 or more. Heritage knows it must proportionally funnel more time and effort into growing this segment than any other in order to remain profitable long term. It has applied the same mentality to its products and services. The credit union now more aggressively hones in on what it deems active accounts — i.e., those with a debit card and online or mobile banking. And after researching how all members — especially Gen Y — think about their financial relationship, the credit union has replaced checking accounts with debit cards as its main indicator of primary financial institution status.

“We don’t just want a checking account and primary share account because that does us no good,” Bugg says. “Instead, we want our members to be loyal, drive interchange, and pull out that card every day so we can be top of mind in our community.”

Additionally, after discovering that its average mortgage holder has approximately five products and services with the credit union, Heritage not only redoubled its efforts to grow this product line but also adopted what it calls a “strive for five” mission to bring every member up to this desired level of engagement.

So far, these data-driven expectations are paying off. As of fourth quarter 2013, loan growth at Heritage is up 6.9% year-over-year, and the credit union’s average member relationship is more than double the average for comparable-sized peers — $15,284 versus $6,122, respectively.

Lesson 4: Work Different Markets Differently

At Heritage, like many other institutions, brick-and-mortar locations are evolving from transactional hubs to the new focal point of relationship development and sales activity.  

“We want to take advantage of the times when members are right in front of us,” Bugg says. “They’re so busy in their everyday lives. Once they leave, it’s much harder to market to them.”

As the branch roles change, so too has the way Heritage establishes the potential and priorities of its individual locations. In terms of a loyalty index, which examines member satisfaction along with the likelihood to recommend), all Heritage locations score between a high of 97 and a low of 20.

“Interestingly enough, our lowest scoring branch has the highest percentage of members anticipating they want a loan in 2014, at 45.1%,” Bugg says. “And the reverse is true for the branch that scored a 97, they’re at 22.5%.”

By diving deeper into individual market priorities, the credit union is learning how to more effectively funnel resources, such as marketing materials, customized promotions, and cross-sell scripting, toward the products that have the highest likelihood of success at each hub.

“At [our lowest scoring loyalty branch], auto loans rated highest followed by mortgage loans,” Bugg says. “So our strategy there is to focus on those products.”

Now, when drive-through tellers notice a new vehicle in their lane, they check to see if the member financed it through Heritage. If not, they either attempt to recapture the loan directly or make a note in the system so another employee can follow up on the opportunity.

Heritage also places mortgage loan experts throughout its branch network. Given the low-scoring location’s potential for mortgage loans, it made sense to make this the home base for Heritage’s top-level employee in that department. After all, if placed in the right location at the right time “even a single employee can make a difference to your members,” Bugg says.

 

 

 

April 14, 2014


Comments

 
 
 

No comments have been posted yet. Be the first one.