System Downtime Costs Banks $309M an Hour 

System Downtime Costs Financial Firms $309M a Year 

With global losses nearing $600 billion, corporate treasuries must prepare for sudden system downtime.


Whether it’s an internal resource or a critical service provider, system downtime is becoming an increasingly costly norm.

The global cost of downtime for the Global 2000 is expected to reach $600 billion by the end of the year, according to research published by Splunk, a data platform provider.

“That works out to an average of $300 million per organization each year,” according to the study. “The expenses add up quickly: Every minute of downtime costs $15,000, or over $900,000 an hour.”

Among survey respondents for the report, 268 financial services executives estimated average annual downtime costs of $309 million or about $35,000 per hour, including contractual costs, service-level agreement breaches, security-related issues and lost revenue. Losses for companies in other industry sectors will vary.

“This is the highest across all industries covered,” wrote Charles Adriaenssens, a financial services industry advisor for EMEA at Splunk, in a company blog post. “It also showed the greatest increase over the past two years.”

As clients increasingly expect always-on digital access, reducing downtime has become a fast-growing priority for most organizations, he added. “Customer issues are not only frustrating for those who experience them but also, over time, can reduce customer confidence, damage brand integrity, and contribute to churn.”

The New Normal

In October 2026, organizations experienced 2,550 weekly cyberattacks on average, up 32% year over year, according to Check Point Research.

That number will likely increase as AI-powered hacking tools continue to proliferate. For example, the FBI released a cyber threat alert on Nov. 13 warning of the growing use of an AI-powered autonomous penetration framework called ARTEX.

The FBI estimates that hackers have used the open-source AI model since at least August to attack multiple sectors, including defense, energy, financial services and semiconductors.

The model enables users with little technical expertise to automate and execute every step of the hacking process, from initial vulnerability scans to gaining privileged access, in as little as three hours across multiple targets simultaneously.

ARTEX’s developer removed it from the GitHub open-source repository in October, but the FBI noted that it has continued to proliferate through community forks, localized variations, archived copies and commercial versions. In early November, researchers identified more than 1,600 publicly exposed ARTEX consoles. This does not include instances behind firewalls or on private networks, which could significantly increase the number, according to the alert.

However, none of the most recent service outages have been attributed to ARTEX, but rather to technical bugs, software defects and internal updates.

How Treasurers Can Plan

Operational resilience is not only critical for cyberattacks but also for all types of events.

“Plan for loss of visibility: When balances and pending items go stale, cash positioning breaks down, so treasurers need prior-day and intraday data through more than one channel,” said Jeffrey Gapusan, managing director, banking, payments and fintech at global AI consultancy SIA Partners. “While it is a lost art, balancing a ledger like an old-style checkbook register can save hours of anxiety and frustration.” 

What should corporate treasuries do if their bank acknowledged a payroll or supplier file before losing access to the bank’s portal and doesn’t want to risk making duplicate payments from a secondary bank?

“It’s a matter of “trust but verify,” Thomas Hunt, Treasury Practice at the Association of Financial Professionals, told Global Finance.

He suggests using known contact information to call their relationship manager, client services, or treasury service representatives, or using a backup connection method that both parties arranged earlier, and not using email from an untrusted account.

“When trusted contact is established, the corp/bank then discusses an inventory of what still needs to be completed, including file cutoff times and whether the Fed will likely extend cutoff windows, etc.,” said Hunt. “If a line of credit can be used, bank reserves, pooling/consideration of all cash accounts, etc., can be used to accommodate the size of the file. At a minimum, this becomes a point-in-time inventory of what the bank can support.”

Once systems are back online, companies likely have a limited chance of recouping their financial losses from an outage.

“Bank agreements typically cap liability and offer fee credits, not compensation for consequential loss,” said Gapusan. “The [Consumer Finance Protection Bureau’s] standing guidance says customers can ask their bank to reverse late and overdraft fees caused by an outage. To my knowledge, no such program has been announced, but firms should ask. Regarding cyber insurance, coverage will depend on whether this is classified as a covered event. A bank’s non-malicious technical failure most likely falls outside many policies, and dependent business interruption coverage and waiting periods will vary.”

The best practice, he added, is to document recovery steps and costs, then promptly notify banks and brokers to preserve rights.

Rob Daly covers the economy and fintech. He can be reached at rdaly@gfmag.com

arrow-chevron-right-redarrow-chevron-rightbutton-arrow-left-greybutton-arrow-left-red-400button-arrow-left-red-500button-arrow-left-red-600button-arrow-left-whitebutton-arrow-right-greybutton-arrow-right-red-400button-arrow-right-red-500button-arrow-right-red-600button-arrow-right-whitecaret-downcaret-rightclosecloseemailfacebook-square-holdfacebookhamburger-newhamburgerinstagramlinkedin-square-1linkedinpauseplaysearch-outlinesearchsubscribe-digitalsubscribe-printtwitter-square-holdtwitteryoutube