Banks are working a “spaghetti” tech construction that’s making a sport of “know-how Jenga” for IT managers updating their techniques.
Making adjustments to IT techniques has large knock-on results, with upgrades risking downtime. However banks are keen to just accept a sure degree of downtime – this can solely change when the price of downtime within the type of buyer compensation and fines is increased than the price of reaching 99.999% availability.
Within the aftermath of a three-day outage at Barclays, bosses on the UK’s largest banks had been pressured to clarify to MPs why their digital techniques had been unavailable for a mixed 800 hours up to now two years.
Barclays CEO Vim Maru instructed the MPs the reason for the latest outage was a software program drawback in a essential module of its UK mainframe working system, which “brought about progressively extreme degradation of mainframe processing efficiency”.
Information obtained from banks by MPs on the Treasury Committee revealed not less than 158 banking IT failures between January 2023 and February 2025.
System adjustments had been the key reason for issues, adopted by third-party points, bugs and {hardware} failures.
Chris Skinner, fintech business professional and CEO at The Finanser, mentioned the construction at banks, with many various gamers and suppliers, together with a must repeatedly replace techniques, creates a Jenga-like drawback.
“There’s a spaghetti construction that I name Bankenstein – a complete load of items stitched collectively to work, but when one half dies, the entire system goes down. You see the identical subject with airways and others. Programs go down as a consequence of one piece not working – it’s like one large know-how Jenga,” he instructed Pc Weekly.
Skinner mentioned the advanced legacy surroundings has been supplemented with a fancy multi-supplier surroundings, making “the myriad of dependencies way more advanced”.
Info obtained from the banks revealed that Barclays is anticipated to pay out as much as £12.5m to clients in compensation.
The appropriate value of downtime
In keeping with one senior IT skilled within the banking sector, who wished to stay nameless, all of it comes right down to a easy query: how a lot will it value to make sure 99.999% availability, and is it cheaper simply to pay compensation when issues go flawed?
The IT professional, who has labored at quite a lot of the UK’s largest banks, mentioned: “It appears to be like like accepting the danger of outages is extra financial than working ultra-high availability techniques. If you happen to can recuperate a key system comparatively rapidly, and that’s cheaper than working at ultra-high availability, it could make extra sense commercially to just accept the danger of outages and depend on swift restoration as an alternative.”
It appears to be like like accepting the danger of outages is extra financial than working ultra-high availability techniques Banking IT professional
The banking IT professional recommended the excessive expense of constructing and working techniques on the “5 nines” (99.999%) uptime commonplace means there’s a value discount incentive to chill out this for some techniques.
“Barclays expects to pay extra in compensation than the opposite banks, however the estimate is insignificant compared with the annual IT finances. The banks may also have inside prices, and possibly fines along with compensation, however these will not be talked about within the letters,” he mentioned.
“If the banks can estimate the potential value of fines and compensation for outages in comparison with the price of increased uptime targets, it’s doable to work out the optimum business uptime goal for every system or service. “My guess is this might be decrease than the ‘5 nines’ goal,” mentioned the professional.
Balancing prices
Fives nines availability would quantity to 5 minutes’ downtime per 12 months. 4 nines (99.99%) uptime would permit nearly an hour of outages per 12 months. Three nines (99.9%), in the meantime, would permit nearly 9 hours of outages.
Every further 9 can add important IT value.
“I do know the banks want for 100% reliability and availability of techniques, however that is by no means more likely to occur as it could be technically unimaginable and economically unviable,” the banking supply instructed Pc Weekly. “The very best defence for patrons is to have accounts with a number of banks and hope they don’t all break on the identical time.”
The banks want for 100% reliability and availability of techniques, however that is by no means more likely to occur as it could be technically unimaginable and economically unviable Banking IT professional
Following the reasons from banking bosses, Treasury choose committee chair Meg Hillier mentioned: “For households and people residing paycheck to paycheck, shedding entry to banking companies on payday generally is a terrifying expertise. Even when rectified comparatively rapidly, it could actually trigger actual panic, which is why we wished to get a correct understanding of why unplanned banking outages occur and the way banks and constructing societies reply.
“The very fact there have been sufficient outages to fill a complete month throughout the final two years reveals clients’ frustrations are utterly legitimate. The fact is that this information reveals even essentially the most profitable banks and constructing societies hit technical glitches. What’s essential is that they react swiftly and guarantee clients are saved knowledgeable all through.”
Regulators have imposed main fines for essentially the most critical outages. In 2018, whereas migrating to a brand new core banking system, TSB skilled main issues. Over a five-day interval, customers had been locked out, skilled cash disappearing, and a few had been even in a position to see different clients’ accounts.
The UK regulator fined TSB practically £50m for its failures. The financial institution additionally paid £32.7m in redress to clients who suffered detriment. Its then CEO, Paul Pester, fell on his sword, leaving the corporate quickly after the catastrophe. The UK Prudential Regulation Authority additionally fined TSB’s former CIO, Carlos Abarca, £81,620 for his half within the catastrophic migration.
