The glitch that left SunTrust customers without access to mobile and online banking for two days is just the latest digital snafu to afflict the banking industry. Upgrades by banks or their vendors that went haywire have resulted in multiple outages this past year.
So it begs the question: If there’s such potential for an upgrade to go wrong, why aren’t banks more prepared for the eventuality, instead of scrambling to get service back up and running?
Industry experts bear some bad news about the state of technology in banking: Banks can’t really be proactive enough to prevent the problems.
Working against banks are several factors, they say: Legacy code and infrastructure can create issues, while the increased need to continuously update systems has shortened the time for testing and extra scrutiny. Other factors include the increasing complexity of banking systems, which can include multiple vendors providing different services.
Eric Baryol, principal at Levarge IT Consulting in Sacramento, Calif., said each banking outage is unique because every bank’s digital ecosystem is different. But, generally speaking, outages are often tied to legacy infrastructure and code.
“They are built by different people at different times,” Baryol said about banking digital systems. “So there could be underlying software systems that are legacy systems, and [engineers and developers] are bolting on new interfaces and continuing to move them forward. They need to attach and bolt on these systems, and this introduces more complexity.”
Other causes for errors: The banking infrastructure housed in a data center can have an unknown weak spot in the system, “a single point of failure” that engineers did not foresee, Baryol said. Human error is another cause, such as typing in the wrong code.
Emmett Higdon, director of digital banking at Javelin Strategy, said there are a couple of contributing factors to why banking systems go down. One is the pace of changes in banking technology and systems.
“In years past, if a bank updated a mobile banking app once or twice a year, it was a big major change, all done in one batch,” Higdon said. “We now see banks doing incremental changes to assets, multiple times a month.”
As banks move from big package releases to an almost continuous stream of updates, that increases the chance of an error being introduced, said Higdon. The timeline for testing could be compressed compared with years past due to these constant upgrades.
A second factor is that digital banking systems are now more complex and interconnected. In a modern digital banking infrastructure, you have many different connected systems, such as debit transactions and a reward system, he said.
Another factor is that a banking ecosystem will have multiple vendors plugged in providing different services or features. This further increases the complexity of a banking infrastructure and likelihood of a ripple effect if something goes wrong.
To prevent these outages, Baryol said banks need to have built-redundancies in their systems (and most of them do), but also banks need to test any upgrades multiple times or do a test run before implementing any changes in a real-world setting.
Higdon agreed that multiple testing needs to be done, and he said that banks need to concentrate on making changes on one platform at a time. One week, there will be an upgrade to mobile. Another week, an upgrade for desktop.
SunTrust first reported its mobile and online banking services were down Sunday morning. The banking platforms were once again accessible to customers Tuesday morning.
"We appreciate your continued patience as we restore our online and mobile banking services," the company tweeted late Tuesday morning. "Some clients may have intermittent access, and we continue to monitor performance."
The official SunTrust Twitter account informed customers that the bank “will make sure you do not incur any fees or penalties due to this outage.”
Still, as
The bank put three apologies out in a row on social media.
The last: “We recognize the frustration our digital banking outage has caused,” read the bank’s Tweet. “We are dedicated to fixing this problem ASAP.”
