How my Bank’s Data Analytics Could have Kept Me as a Customer

By | Tuesday February 18th, 2014

A personal story about my (former) bank, churn and analytics .

I’d like to share a personal and frustrating story with you about the time I switched banks.  It was completely preventable and didn’t require much effort on the bank’s part to retain me.  For my part, it was a long, irritating, costly and difficult process to sever those ties I had established.  In the end, we both lost, but just a little bit of data and analysis would have changed this from a tale of woe to one of viral elation.

Let’s take this step by step with a bit of what happened versus what should have happened.

What happened?

I visited a branch after noticing a reduction in interest payments on my savings accounts.  An account review was done and a temporary adjustment was made. 

What should have happened!

There was no need to visit the branch as the interest rate reduction was proactively handled.  I received an email from the bank explaining the date it was going to happen.  Further, based on my relationship with the bank, they were pleased to offer me a temporary rate increase.

What happened?

During the account review, I enquired about a different credit card product which was more suitable to my spending behaviour and needs.

What should have happened!

A scoring mechanism evaluates customer spend and payment behaviour.  This is matched against all current and concept card products and results in a propensity score for each.  This is made available to a call centre, a branch, or an automated process and is delivered to me in the most appropriate way.

What happened?

After being approved for the new card, I was told it would arrive in 5-7 business days. It didn’t arrive after three weeks, and I phoned the call centre.  I was directed to a website to accept additional terms and conditions.

What should have happened!

The workflow from approval to delivery should have been made available to the branch associate.  The additional terms and conditions would have been detected and resolved.  Internal workflow for delivery should have detected the missing acceptance and triggered an alert to either the branch or a call centre.  I would be contacted, and the issue resolved.

What happened?

After filling out the form on the website, I received a rejection notice two days later directing me to a call centre.  The call centre directed me to a branch for an additional form.  I received another rejection email, called the call centre and was told to go to the branch (yet again) for another form to fill out.

What should have happened?

Had the workflow been made available in the beginning, this would not have occurred.  Even so, the e-mail with the decline could have contained not only the reason for this rejection, but also the appropriate steps for remedy. It could have also supplied the necessary forms, or a link to a customised site for electronic signature.

What happened?

I phoned the call centre, clearly agitated, and canceled the new product.  I also mentioned account closure.  I did some research on other credit card products and called back to the call centre and asked to close my account.  That department is closed, so I must call back on Monday leaving me to think about things over the weekend.

What should have happened!

First, it should have never reached this point.  Speech to text processing of the call could have been analysed with sentiment analytics.  This and other indicators enter a scoring process to provide a propensity score for churn.  High risk customers are handled by a specialty branch with specialised training and treatment strategies.

What happened?

On their mobile application I moved all of my funds from my savings to my transaction account.  I withdrew most of my funds at a branch, and then deposited the cheque at a competitor.

What should have happened?

An alert should have been created showing a significant event in my account.  The details would be routed to a specialised care team to contact me and discuss the nature of my funds movement. 

What happened?

Over the next month BPay events ceased, direct deposits stopped, accounts sat dormant.  Finally, I closed my accounts in branch and withdrew the remaining funds.

What should have happened?

Other event detection analytics would identify the change in transaction behaviour and detect the change in the regular deposits.  Analysis would indicate either a risk (such as job loss) or a change (customer churn).  Email, SMS or direct contact would be initiated to discover the cause and try and resolve any issues.

By not leveraging their data, finding insight and knowledge and taking action, this bank lost me as a customer.  Further, they have had the additional impact of my migration to a competitor, and referring my friends and colleagues. 

Send through your comments if you have had a similar dilemma with your bank.

John Berg is the lead Principal Consulting Architect for Teradata Australia/New Zealand, guiding market leading companies in the region further advancing their lead over the competition. John’s experience spans industry verticals including hospitality, banking, retail, e-commerce and government.

The following two tabs change content below.

John Berg

4 thoughts on “How my Bank’s Data Analytics Could have Kept Me as a Customer

  1. avatarAlan

    You might want to also consider the possibility that as a well informed customer, you don’t represent the a market segment that provides them with a good source of income. Banks get most of their profit from the dis-organized and struggling masses with late payments and high interest rates.

    While it may have kept you as a customer had they taken the appropriate steps, there may not be any financial incentive to do so. It may be cheaper to shed your thin margin than implement a system to keep it.

    I’ve cancelled accounts at many banks for their failure to treat me as an intelligent and respected customer.

  2. avatarWayne Newton

    Nice one John!

    I feel (and have felt) your pain. Sad about your frustration, but the bank should be sadder. They should have not lost you as a customer. You simply “voted with your feet” as I would have done.

  3. avatarJohn Berg

    Thanks for the comments!

    Alan – I wish I could give the bank that kind of credit. I had a total of 8 products with the bank which are no longer there. I don’t have current research, but I recall there was a significant spend to attract 1 new customer to open 1 new account. That spend would go up on a more exponential curve with each additional product. I’d love to take a look at that bank’s data to identify what my segment was in a lifetime value.

    When I worked at a major US bank in the late 1990’s, we had a strategy much like you described in the credit card market. People who paid balance in full, made nearly no money for the card (just interchange transaction fees). We liked customers who were a little ‘irresponsible’: just a little delinquent; carry a reasonable balance; maybe make a unfunded payment or two. The balance was to ensure that they didn’t go even further and default.

    The bottom line is that without the analytical insight, you really have no proof, just a theory.

  4. avatarAlec Gardner

    Happy to report that my bank just got a product increment from me when they realised I was about to buy a car. As I went to collect the bankers cheque I was given an insurance quote that took immediate effect and had so few restrictions I would have been a fool not to have taken it and thus ensure that my new car was comprehansively insured when I drove it away. My bank even called the dealer to get the VIN and registration number as i didnt yet have it.

    They also asked me whether I’d “found the house in Mosman?”. I did a double take and looked at the member of staff to see if I knew her. I have to admit to being unsettled for a millisecond and then pleased as I realised that the intel behind all of this was Teradata! :) Sadly I was not able to report the acquisiiton of a house yet so was offered a mortgage review.

    As I say, I was unsettled for a millisecond but this, after all, is my bank and my primary financial institution – I want them to be able to offer me services that benefit us both. (As it happens, I declined the mortgage offer for now)

    As I left I apologised (again) for causing the GFC! :)
    (refer to one of my previous blogs)


Leave a Reply

Your email address will not be published. Required fields are marked *