WEST PALM BEACH, Fla.-Take small steps first, have a long-term focus, and avoid coming up with a lot of data that can't be meaningfully interpreted.
That's just some of the advice offered to Credit Union Journal readers by credit card experts when it comes to using data analytics with a card base. Here's what five people had to say.
1 MAKE A COMMITMENT
BOSTON-Ron Shevlin, senior analyst at the Aite Group, insists that when entering into cards analytics the CU has to make a commitment to building a new business capability around the tools.
It can't be the kind of effort where the credit union puts its toes in the water and then is ready to step back if the results are poor or market conditions turn down. "To be successful in using analytics means making an investment in building a competency around the tools because it will be a multi-year process. It's really about building a roadmap."
Using an outside provider is fine, said Shevlin. "You don't have to build it all in house. But what I have found is that the low performers, when they have predictive models, they don't use as many sources of data as the more sophisticated institutions that have a great deal of success. The lower performers might simply rely on compiled demographic data as opposed to using more behavior data. The good shops also will use purchase data and even social media data."
2 GOALS FIRST, DIVE SECOND
RANCHO CUCAMONGA, Calif.-Michelle Thornton, senior product manager for CO-OP Financial Services, says the CU has to know what its goals are before diving into any analytics program.
"There is a lot of data to wade through and if you are not sure what you are trying to accomplish you can spend a lot of time going through a lot of data and then coming up with just that-a lot of data."
Thornton said to take small steps first. "Start with the most obvious things- increasing the number of transactions from existing cardholders. You don't have to segment down to the 18- to-20-year olds who use their card at Starbucks. Start with the easier things, you are still segmenting."
Once the credit union becomes more comfortable with the tools it can then dive even deeper into the data, added Thornton.
3 BREAKING UP THE PIECES
ST. PETERSBURG, Fla.-The first step of any good analytics effort is to take a deep dive into the card portfolio to see what behavior segments exist, according to Michelle Hillenbrand, director of marketing at Advisors Plus.
"Does the CU have a lot of closed accounts? Does it have a lot of inactive accounts? How are they doing in terms of revolvers?" asked Hillenbrand, who reminded the deep dive includes digging into processor data.
Hillenbrand said Advisors Plus performs this work for credit unions. "Once we break apart the portfolio into pieces, we can determine many things. For example, the credit union may have too many inactive accounts and needs to weed them out, or it may have credit lines that are constrained and needs to create a marketing plan to get these people moving because they won't use their card much for fear of bumping up against their credit line."
But the key throughout, emphasized Hillenbrand, is that whatever marketing plans are created from the findings, they must be in conjunction with the credit union's goals and objectives.
4 UNDERSTAND THE SEGMENT DRIVERS
DES MOINES, Iowa-Any good analytics effort begins with segmenting consumers by behaviors and understanding what is driving those behaviors.
That's the stance of Todd Herren, chief technology officer at The Members Group. "You can then target them appropriately: are you transactor, revolver, inactive, or a pay-down? Within those segments you go deeper to learn what you can do within each group."
For revolvers, the move might be to increase a person's limit, offered Herren. "Not only will they begin to carry more balances, they will feel more loyalty to the credit union because they know the CU trusts them, understands them, and cares about them."
For transactors, the move might then be a balance transfer campaign, said Herren. "If you can drive the right balance transfer campaign to the right people, they will go to revolving balances as well."
5 0FFER THE RIGHT INCENTIVES
TAMPA, Fla.-Among the many advantages of analytics, Barney Moore likes how the tools can help the CU make card incentive programs much more effective.
"I really like cash incentives; earn a point for every dollar you spend. Target cardholders with double or triple points for a certain number of transactions or reaching dollar threshold," said the senior portfolio consultant for CSCU. "Analytics tools allow you to ID a group of cardholders who shop certain merchant categories and tailor incentive offers to their shopping behaviors."
When mobile payments become more accepted among merchants, Moore sees the merchant offers shifting to more real-time discounts, making analytics that much more important.
