A Lesson In Mobile Banking Economics

A Wall Street Journal article titled Mobile’s Rise Poses a Riddle for Banks is so off the mark, I simply can’t let it slide. According to the article:

“A waitress from Queens, NY, Ethel Bueno keeps her phone close at all times and frequently logs on to her bank account to check her balance. But for bigger and more-complex transactions, which often require fees, Ms. Bueno prefers to visit a bank teller in person. That means her digital devotion to the bank doesn’t actually generate much revenue, a puzzle firms across the industry are still trying to solve. According to a new study by Bain, mobile [bank] interactions are now 35% of the total, more than any other type. A 2012 Fiserv report found that digital transactions cost on average 17 cents each, compared with 85 cents for an ATM transaction, and $4 for an interaction with a bank teller.” (italics added)

My take: Blind faith in the contention–and numbers–thrown around in this article is sheer stupidity.


If you read the full WSJ article, you’ll find that the two bank execs quoted are anything but puzzled about mobile banking. So how the author of the article concludes that mobile banking is a “puzzle” that banks “are trying to solve” is beyond me. Did anybody mention to the WSJ that the transactions being conducted on a mobile device for free were actually free when they done in a branch or ATM?

If anything, it’s a gift on a silver platter. Who cares if it doesn’t generate revenue? The cost savings that are implied by the channel transaction costs should more than compensate for the fact that the mobile transactions don’t generate revenue.


Assume for a moment that there’s a bank whose customers conduct 1,000,000 transactions per year. The WSJ article didn’t include estimates for call center transaction costs, so let’s assume they’re somewhere between the ATM and branch, at $2.50 per transaction.

Now assume the following transaction allocation, first before the implementation of digital banking (online and mobile) banking, and then now, with 35% of the transaction volume going to mobile (and for argument’s sake, another 15% coming from online transactions).

    Before After
  Per transaction cost Transaction allocation Channel costs Transaction allocation Channel costs
Digital $         0.17 0% $                  – 50% $        85,000
ATM $         0.85 10% $        85,000 8% $        68,000
Call center $         2.50 25% $      625,000 10% $      250,000
Branch $         4.00 65% $  2,600,000 32% $  1,280,000
TOTAL     $  3,310,000   $  1,683,000

In the Before scenario, based on the assumed transaction allocation, total costs are ~$3.3 million. In the After scenario, with digital transactions at 50% of all transactions (35% for mobile, 15% for online), the increased allocation had to come from somewhere, so ATM and call center percentages declined slightly. And with the impending death of branches (hey, that’s what all of you think, not me), the % of transactions going to branches declines from 65% to 32%.

Overall, the cost reduction from this jump in mobile transactions–based on the lower cost per transaction–should produce a 50% reduction in transaction costs. Good luck finding a bank whose has reduced their operating costs by 50% because of this explosion in mobile banking transactions.


But clearly, what’s happened, is that digital transactions have not fully displaced transactions in other channels, but have added to the overall transaction volume.

Now here’s what’s important, and what way too many bankers either can’t understand, or refuse to understand: It doesn’t matter how much less expensive digital transactions are than transactions in other channels if the digital transactions don’t displace those other channel transactions.

If digital transactions are additive, then even at just $0.17 per transaction, every digital transaction adds to the bank’s cost structure and is a drag on profitability. Conclusion: Per transaction costs by channel are meaningless if channel migration doesn’t occur.


There’s another problem with those channel transaction costs: They’re averages of all the transactions that occur in the channel. Some branch transactions will cost much less than $4 per transaction, and some will cost more.

If the branch (as well as call center and ATM) channel transactions that are displaced by digital transactions are the lower cost transactions, then the potential cost savings is even further reduced.


From a cost perspective, there are two key elements in mobile banking economics: 1) How many transactions are displaced from higher cost transactions, and 2) How many new transactions are added?

If the migration from higher cost channels doesn’t occur, of if the migration is from online to mobile (which is entirely possible), then there’s little economic benefit. If many new transactions are added, then even with a low cost per transaction, overall costs go up.

According to the Bain study, mobile transactions now account for 35% of all transactions. In other words, a minority of consumers (what, ~25%?)–who presumably are not mobile-only customers–are now accounting for a hugely disproportionate percentage of all banking transactions. That tells me that this minority must be adding a huge number of new transactions to the mix. Which implies that mobile banking isn’t saving banks anything, even with the huge disparity in per transaction costs.


The riddle (i.e, the challenge) that banks face–if there is one–has nothing to do with mobile banking. It has to do with the other channels. What can banks do to drive transactions out of those channels and into the lower costs channels?


8 thoughts on “A Lesson In Mobile Banking Economics

  1. Thank you Ron. It is amazing how many banking consultants throw around the cost of a transaction without ever mentioning that there isn’t a 1:1 displacement value like you have done.

    Another variable that is often missed is that even if you factor in the right displacement %’s and add to the overall number of transactions as you have, the math assumes that the bank will take actions to reflect the drop in costly transactions.

    In other words, while the call center may have reduced transactions, the cost of qualified call center agents may have increased because, instead of looking at a computer screen and reading off balances, they now need to be technicians who understand how the mobile banking app works on each type of device. In addition, if a bank has reduced branch transactions of 30%, how many banks have reduced their teller staff AND branches by 30%? How about none.

    The impact is that the per transaction cost of traditional channels is going up as transaction volume is decreasing since the lower number of transactions are being spread across only a marginally lower fixed cost.

    Until banks actually reduce their costs by eliminating the fixed and variable costs in less used channels, the benefits will only appear in articles by the WSJ where disillusioned banking execs give numbers they have read as opposed to numbers they are experiencing.

    1. Jim: Spot on. Here’s another factor that I didn’t even bother to bring up: Development costs. Call center and branch functionality may have been developed long ago, and those development costs were amortized long ago. Building new mobile functionality takes $.

      1. I agree that the costs don’t go away immediately (branches, office space, staff, etc), however I think FI’s need to get creative in how these fixed or semi-variable baseline costs are allocated and recognized, so as to provide more incentive to those trying to develop lower cost options.

        1. Don’t get me wrong, Frank. I’m a huge proponent of the mobile channel. But the transaction cost estimates are a sham at worst, and misleading at best. In addition, the premise of the WSJ that there’s some kind of riddle the industry is dealing with is a bit far-fetched.

          1. There are cost savings moving current transactions to digital channels. The impact, however, is nowhere near the level mentioned by some consultancies (Bain is not the sole culprit). And the savings can be zero if no adjustment is made to the legacy fixed costs since there are only nominal variable cost savings by eliminating a check cashed or deposit made. It’s time to eliminate tellers and close branches.

    2. “In addition, if a bank has reduced branch transactions of 30%, how many banks have reduced their teller staff AND branches by 30%? How about none.”

      I hate to be a bit of a naysayer for that statement but I was one of the tellers from BB&T who was laid off at the end of the summer when BB&T decided to do a reduction of the workforce, in large part, due to the mobile app changing the traffic flow in the branches. There were a lot of tellers and relationship bankers who were laid off this year from across the company not just in the Triad area. From what I have been able to tell while looking for another teller position there aren’t that many banks still hiring tellers. They are switching over to teller/bankers positions where half your time is spent in an office or at a desk doing a bankers job.They are streamlining their branches so they are reducing their workforce by simply making their current employees multitaskers and all new employees multitaskers.

  2. It seems I somehow missed the part that accounts for the cost of losing a customer once he/she finds out that his or her peers are doing all kinds of things with their finances that they aren’t doing or simply can’t as a result of the inconvenience, the inability or the excessive cost of doing so using the more traditional technology (or lack thereof).

    1. You are right Rick. There is no human version of real time and proactive digital money management combined with alerts and notifications or a human alternative for 24/7 deposit taking or P2P payments or even a 30 second personal loan like the one offered by mBank. Once the mass market realizes digital’s full potential in banking, branches will be vacated more than today.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s