Mining Data to Make Money

mining-data.pngAt U.S. Bank, big data is not an abstract concept. It is making real money for the Minneapolis-based $371-billion asset bank. The bank tracks transactions and customer online interactions to flag those for a personalized marketing message, people who might be a good prospect for a loan, for example. Recently, a branch manager used such flags to call a business customer who had recently moved a large sum out of the bank. It turned out that the customer, a business owner, was planning to apply for a loan for his business at another bank, hence the money transfer. Instead, the banker was able to persuade him to apply for a loan at U.S. Bank.

“Not only were they able to get that customer the funding that customer needed to expand his business but they were able to give him a line of credit for operating expenses as well,’’ says Richard Martino, senior vice president of enterprise data and analytics at U.S. Bank.

Big data is no longer a futuristic concept for technicians and marketers to fantasize about. Credit card companies have used data analysis to make sophisticated fraud detection and credit offers for years, but such analysis wasn’t widely used by banks for other purposes. That is changing. Even small banks now buy slices of information for targeted marketing campaigns that hit on their core strategies, rather than worrying about compiling and understanding unwieldy databases of information. The cost of such information for a variety of companies is now much more affordable than it used to be and easier to use. Driving this is data warehouses that know everything about you and that you have never heard of such as Acxiom in Little Rock, Arkansas, and credit rating agencies such as Equifax in Atlanta. The sheer amount and scope of information that can be purchased on consumers across the country is unfathomable. Data warehouses can basically find customers for you to target in marketing campaigns based not only on where they live, but what their shopping habits are, their income levels, their debt and how much investable assets they have. You can basically underwrite them before you offer them a loan via an email campaign.

Do you want to know which of your customers are getting ready to buy a home? No sweat. Buy the data. Which ones are shopping for an automobile? Check. Would you like to contact all the people who move within a mile of one of your branches within a week of their move (before they have a chance to sign up with a different bank or tell the post office)? Check again. Data analytics can be used in customer service and relationship management, as well as risk and compliance. The challenge is deciding what data your bank needs, interpreting it correctly, and executing your plan to use that data well.

“Banks and credit unions struggle with this so much,’’ says Patrick Grosserode, director of product management for Deluxe Marketing Services, a division of Deluxe Corp. in Shoreview, Minnesota. “They can’t find the mix that works. It seems like you are trying to boil the ocean.”

Big banks have hundreds of employees who work in data analysis, just trying to figure out what their data says about their customers. Some work in marketing or fraud detection or compliance management. The biggest banks might have 40 people looking at branch analytics, 50 people doing pricing analytics and 100 people doing market analytics, says Sherief Meleis, a managing director at the New York-based financial services consulting firm Novantas, who estimates that banks that use customer analytics to drive marketing, distribution and pricing can improve revenues by 3 to 4 percentage points per year, and earnings by 10 to 15 percentage points per year.

“The big [banks] have armies the size of our company churning through the data,’’ says Equifax’s Retail Banking Leader Brad Jones. “They are myopic. They don’t even know what their companies are doing [with it.] They are just looking at the data.”

To protect consumers, agencies such as the Federal Deposit Insurance Corp. require banks to issue privacy statements to consumers letting them know how their personal information will be used, and letting them opt out of certain kinds of use. According to the Office of the Comptroller of the Currency, Regulation DD—which implemented the Truth in Savings Act—established standards for marketing content for deposit accounts. The Fair Credit Reporting Act spells out the permissible uses of consumer credit reports and prescreened consumer reports. Prescreening consumer report lists is only permitted if a bank is making a firm offer of credit or insurance, for example.

The key to good data marketing is targeting your customers without making them feel like they are being spied upon. In the U.S. Bank example, where the bank was using internal data rather than prescreened consumer report lists, the banker didn’t mention the large transaction to the business owner when he called. He just asked the guy if he was happy with the bank.

Banks can also use outside vendors to drive the marketing campaign. For example, if a credit rating agency can see you are making a car payment every month for a certain amount, they can surmise what your rate is, when you will be buying another car, and tell your bank when it’s time to make a better offer. Banks have a lot more information on their customers than they can possibly sift through. But data compilation is much simpler than it used to be. Core processors such as Fiserv, FIS and Jack Henry & Associates are holding on to much of this data that their banks could use. “The great thing about data analytics these days is, it is much cheaper than it used to be,’’ says Corey Booth, managing partner of The Boston Consulting Group in Boston. Cloud-based solutions can help store data. Start-up technology firms are trying to take advantage of cheaper software. “It’s so hard, but it’s not as hard as it used to be.”

For example, StrategyCorps, which sells a checking account product called BaZing, will use the bank’s own data to help banks determine which of their customer households are profitable based on everything from debit fee income to loans. Profitable customers get the value-added checking account with coupons and other services for free. Unprofitable customers can pay $6 per month for the checking account (or opt for a no-frills account that is free if a minimum deposit amount is met). Banks from $500 million to $750 million in assets typically use BaZing. [Full disclosure: StrategyCorps is partially owned by the same investor who owns Bank Director magazine.]

Grosserode says there are several important issues that banks must sort through first if they are going to use data well. You have to define what you are trying to do. Then, you have to ask yourself if that is even possible to measure. If not, you should stop right there. Ask yourself the next question: Is it possible to generate a profit from this campaign? If there are only seven people you end up going after, that might not be so profitable. Is the segment you want to focus on stable enough not to vanish over time? How will you measure and track your success? Is your success, for example, based on the number of people who click on an online ad or the number of people who actually go through the process of filling out an application online and getting approved? Establish a control, as if you were conducting a scientific study. What might happen if you don’t do this marketing campaign, and how will you know that? Banks that don’t have sizeable marketing departments will probably have to hire a marketing firm to help make decisions and devise a campaign.

What happens if you don’t get into data analytics? After all, it sounds time consuming and expensive. The old fashioned way to generate home mortgage leads is to generate relationships with real estate agents. But those traditional avenues may not cut it anymore. People do a lot more research online than they ever used to, both for auto loans and mortgages. Younger consumers rely on recommendations from friends more than they do real estate agents, says Stephen Ramirez, the CEO of Beyond the Arc, which does data analytics for banks. People will price their mortgages and get offers online. “These innovations are beginning to take root now,’’ he says. “It is even more vitally important to develop the business strategy and match the strategy to your priorities.”

Paul Schaus, the president, CEO and founder of consulting firm CCG Catalyst, says it’s a critical time now for banks to address how they want to use data, and not get left behind. “Technology has allowed us to pinpoint our clients and attract them to our bank. If you don’t do that, your competitors are going to get those customers and you are going to be left with the customers nobody else wants.”


Naomi Snyder


Editor-in-Chief Naomi Snyder is in charge of the editorial coverage at Bank Director. She oversees the magazine and the editorial team’s efforts on the Bank Director website, newsletter and special projects. She has more than two decades of experience in business journalism and spent 15 years as a newspaper reporter. She has a master’s degree in journalism from the University of Illinois and a bachelor’s degree from the University of Michigan.