Putting It All Together: How Data Science, Technology and Retail Banking Can Delight Customers


Share on LinkedIn

Who hasn’t had poor experiences with their bank branches? The mere thought of waiting in lines to transact with a teller, sitting with a customer service rep who started work the week before and has no idea who you are, what you need or how best to serve you gives me a stomach ache. Where’s my mobile banking app, please?

However, advances in what the tech industry calls “anticipatory technology” as well as some well-touted new mechanisms like Bluetooth Low Energy (BLE), in-store sensors called “beacons” and modern data management and analytical tools have the potential to make me a delighted customer in a branch.

Imagine this near-future scenario with me: I walk into a local branch of Big Bank because my paycheck and my mortgage payments are not synched up well and I worry I’m going to bounce a check for my biggest investment – my house.

Upon entering the branch, the beacon in the lobby senses the signal coming from my iPhone 8, which has been enabled to communicate a token code – very much like what my retailers send back and forth when I make a credit card or debit card purchase. This token code tells the bank’s systems that Eric Levy has just walked into their branch. The bank’s intelligent concierge immediately queues me up in the teller system. When I approach the teller counter, my information is available to the bank employee.

At the same time, the intelligent concierge system sees all of my customer indicators, which have been artfully prepared by the bank’s data scientists to tell the bank employees key information about my relationship with the bank, all of my accounts, my balances and any opportunities to offer up products and services that people with similar accounts, balances and habits have as well.

I have a habit of keeping lots of money in my checking account (I’m terrible about moving my money over to savings; I’m fearful I’m going to bounce my mortgage check). But due to my overall savings relationships with Big Bank, they’ve segmented me into a group they call “High Potential Savers.” I’m there with lots of other customers who’ve found happiness and contentment in their financial lives by switching over to the bank’s new zero-balance account (ZBA) checking product. It “sweeps” my money over from an interest-bearing money market account to checking as needed, to cover my checks and purchases.

Ralph Kurrensy, a new relationship manager with the bank gets a signal at his desk terminal indicating that someone from the bank’s “High Potential Savers” segment has just walked through the door. With a glance at the pop-up window, he sees my name and some key information about me, including a suggestion to intercept me before I get into the teller line.

Ralph comes out from behind his desk, straightening his tie and greets me by name “Hi, Mr. Levy! My name is Ralph Kurrensy and if you’ll follow me to my desk, I’d be delighted to serve you today.”

I am taken aback that this new face in the bank apparently knows who I am, and eyeing the long teller line, I sigh with relief that my lunch hour might actually include lunch. I follow Ralph over to his desk where his screen now has all of my details pulled up. Being a service-oriented individual, Ralph asks how he can help me today.

I mention my concerns about bouncing my mortgage check and ask pointedly about overdraft protection to keep me from getting into hot water with my mortgage company. Ralph smiles broadly and does a silent cheer – the bank systems correctly predicted that I’m a candidate for the zero-balance checking account.

After patiently hearing me out, Ralph clicks on the ZBA product link on his screen. He turns the monitor to face me to show me the key details of the checking product. The screen lists features and benefits ordered, based upon research the bank has conducted to determine how people in my segment feel about the relative importance of the ZBA product features.

Now I’m smiling broadly. Instead of feeling like a check-kiter asking for some more string, I’ve been offered up both a solution to my big fears and the ability to earn a bit more on my hard-earned money. I no longer have to anxiously pull up my account balance daily to make sure some unexpected hit to my checking account is going to bounce my mortgage check. Gee, this Ralph guy is a genius! And, wow, my impressions of Big Bank have gone from just merely ubiquitous to a company that “gets me.”

Science fiction? Nope. The banking industry already has most of these tools and techniques at their disposal. Apple has already built Smart Bluetooth into their mobile phones, and iBeacons were introduced last year.

My iPhone communicated the token code to the beacon in the bank. This is the same gibberish-looking information that retailers produce when they run your credit or debit card through the store’s merchant systems. Rather than sending the card number to the retailer’s bank, it uses a token code produced by the merchant system. The bank reads this token code and looks it up on a secure database, providing authorization and security for the credit card holder. As the new chip-and-pin debit and credit cards emerge in the latter part of 2015, mobile devices can store and use them safely rather than transmit card information in the open retail environment.

Segmenting me into a customer opportunity group is a somewhat exotic approach using latent class segmentation. Data scientists commonly use this approach that combines behavioral information that the bank has about customers along with a well-designed primary research approach that yields “fuzzy segments.” Unlike other clustering techniques which put each and every person into a single group, the fuzzy segments tell the data scientist the groups to which I most likely belong. I’m 80% likely to be in the “High Potential Savers” segment, 10% likely to be in the “Check Bouncing Deadbeat” segment, and 10% likely to be in the “Chronic Bank Worriers” group. And since I told Ralph to sign me up for the new ZBA checking account on the spot, I’ve provided the bank’s systems with additional information. The data scientists actually can improve my segment classification. Upon future interactions with the bank, this team’s recommendations and observations about me become even more accurate.

The bank’s data warehouse stores these data in a special database tied to the transaction and sales systems. My account information (the account numbers, balances and other transactional data) is “householded.” The bank rolls up my data from an account level to a person-level (me vs. just my account). The bank can apply my information to my family, including my spouse and children, all of whom have their own relationships with the bank – and deserve recognition as part of a household unit with multiple accounts and people. The bank employees actively query the data warehouse as needed. The bank also uses my information passively. Examples of this application include my queuing up for the teller line, and the customer service representative getting pinged with a new product opportunity for the customer walking in the door.

We’ve all been subjected to hype-laden stories about the demise of bricks and mortar banking locations, the “internet of everything” and scary privacy and security concerns. However, many of the tools described in my tale are available. Banks need only to stitch together the solutions. By doing so, data-rich-insights-poor financial institutions can delight customers, improve cross-sell opportunities and increase business results on a systematic basis.

Do you find this tale creepy or exciting?

Republished with author's permission from original post.

Eric Levy
Eric is a marketing research professional with over 20 years of experience with a classic background in brand and advertising development, marketing research design, methods and analysis, organizational development and strategic planning. He's had supplier- and client-side roles with significant P&L, budget, management and client service responsibilities.


Please use comments to add value to the discussion. Maximum one link to an educational blog post or article. We will NOT PUBLISH brief comments like "good post," comments that mainly promote links, or comments with links to companies, products, or services.

Please enter your comment!
Please enter your name here