NWS

Microsoft Outage Disrupts Global Workflows; Businesses Face Billions in Losses

A global Microsoft 365 outage disrupted Outlook, Teams, SharePoint and Copilot, leaving businesses, schools and public services struggling to function. As engineers battled repeated server failures, the cause remained unconfirmed.

Microsoft's Redmond campus in Washington, US. Its dominant position in workplace software and cloud services makes the company a central pillar of the world's digital economy. Photo: Wikimedia Commons / Coolcaesar.
Microsoft's Redmond campus in Washington, US. Its dominant position in workplace software and cloud services makes the company a central pillar of the world's digital economy. Photo: Wikimedia Commons / Coolcaesar.

Microsoft’s Outlook, Teams and Copilot services failed across the world from Thursday (November 12) and were still unreliable on Friday (November 13) afternoon. So far, Microsoft has confirmed only partial recovery for some users and has given no timeline for full restoration. Experts estimate the outage could cost businesses billions of dollars. Microsoft has not disclosed its own loss figure or identified a root cause.

Users began losing access to Outlook mail through Exchange Online at 16:17 UTC on Thursday, equivalent to 9.47 pm in India. Exchange Online is the cloud service that stores and delivers email, calendars and contacts for Microsoft 365 customers. When Microsoft opened its incident record, it was already investigating reports from users who could not access mailboxes on the web, desktop and mobile. The company initially described the problem as confined to the Americas, but over time it became clear the trouble was global.

Why one failure spread so far

Many Microsoft 365 tools depend on Exchange Online, so its failure spread outward. Teams, SharePoint, Copilot and Microsoft Defender users in all regions were affected, although features that do not depend on Exchange Online reportedly kept working. A separate Dynamics 365 incident locked contact-centre agents out of Microsoft’s workspace for just under eight hours on Thursday evening. OpenAI and Anthropic also reported problems with their Microsoft 365 connectors.

Microsoft has not published a root cause. By Thursday evening, it had determined that the elevated load was not caused by increased traffic. The problem was a shortage of working server capacity, not a surge of users. Microsoft said the affected infrastructure was experiencing elevated load across multiple regions. It later said a core process that sets up secure connections on the affected servers was becoming overloaded.

Problem not contained

Microsoft's repair attempts repeatedly failed. The company announced a “gradual recovery” just after 9 am AEST on Friday, but an hour later said restored servers were becoming unresponsive again. Its update said restored servers “can become unresponsive again once they receive traffic.” Earlier, Microsoft said its adjustments were bringing partial recovery in some regions but “aren’t resulting in sustained recovery.” In practical terms, servers that were switched back on looked healthy until real users returned, after which they struggled again.

Microsoft is examining a recent change. It was investigating whether a recent change to its network settings, made around the time the incident began, had contributed to the failure. Still, it has not established whether reversing that change will resolve the problem. Executive vice-president Charles Lamanna said there was “no indication” that malicious or cyber activity caused the outage. Microsoft says it will publish a post-incident report on the root cause and safeguards once the disruption is under control.

Costs to businesses

No one has estimated the total losses so far, and Microsoft has given no figure for lost revenue, customer compensation, or added costs. There is no verified global estimate. The numbers that exist come from outside Microsoft, mainly from Australia, and they measure different things. The widest estimate is the headline one. The ABC reported that the outage is expected to cost Australian businesses billions of dollars and that the financial impact is expected to last weeks. That figure covers all sectors and includes knock-on effects such as delayed invoices and lost sales that may become apparent even after the outage ends.

The hourly figures narrow the lens to one sector. Ashwin Pal, a risk advisory partner at RSM Australia, said professional-services firms such as law and accounting practices could lose hundreds of thousands of dollars an hour. The ABC’s live coverage put the figure at more than $100,000 for every hour the outage lasts. The two reports give different estimates, so the range is the safer way to state it. With the disruption running more than 15 hours, even the lower figure compounds quickly.

The scale of exposure is the third number. Pal said the impact was amplified by the 3.9 million businesses worldwide that use the Microsoft suite, many with no backup systems. That is why small firms feel it most. Tracey Tucker, who owns Queensland business Regal Shower Screens, said the lost day would cost her at least $30,000, with further flow-on effects.

Corporate calendars slipped as well. Swiss luxury group Richemont’s earnings were slightly delayed on Friday, and electric truck maker Workhorse moved its third-quarter results from November 12 to 16, citing the outage. Markets showed no clear verdict. Microsoft shares closed on November 12 at $523, up 0.36%, and traded lower after hours, but that move cannot automatically be attributed to the outage. Analysts said Microsoft’s subscription model and the high cost of switching limit the chance of an immediate customer exodus.

For scale, the 2024 CrowdStrike failure was estimated to have caused substantial financial losses within days. One insurer estimated direct losses to Fortune 500 companies alone at more than $5 billion. A comparable global estimate for this outage has not yet appeared.

Businesses, schools and governments

In the United States, workers at some companies went home early on Thursday. In Australia, businesses could not print invoices, and some schools could not hold online classes on Friday. The ABC reported that some people missed job interviews because they could not connect to virtual meetings, and that distribution centres supplying supermarkets experienced problems. Stores lost email-based sales, patients lost access to telehealth services, and long-distance schools shut for the day. Local governments asked residents to call instead of email.

In India, Downdetector reports of Outlook problems climbed to nearly 4,000 around 10 am on Friday and eased to 2,634 by 12:07 pm IST, still well above normal. Microsoft's updates did not include India-specific figures or a confirmed list of affected organisations.

A repeat warning

This is the second major Microsoft 365 outage this year, after one on January 22. It differs in kind from the 2024 CrowdStrike failure, which crashed computers, cancelled flights and disrupted hospitals around the globe. That incident began with a faulty update to third-party software on customers’ own machines, while this one, on present evidence, began inside Microsoft’s own cloud. Experts say it underscores the risks of companies relying on a handful of tech giants.

Last Edited on

Authors

Author
NWS North America desk

NWS North America Desk

Know More