Millions of Americans sat down at their desks on Thursday, November 12, 2026, and found that their email simply would not work. Outlook would not open. Messages would not send. Meetings would not load. Within hours, what looked like a routine glitch had become one of the biggest Microsoft 365 outages of the year, knocking out email, chat, and cloud services for users across the United States and around the world, and dragging on for more than five hours before Microsoft began to bring its systems back.
The outage hit at the heart of the American workday. Microsoft 365 is the software that runs offices, schools, hospitals, and government agencies across the country, and when its core services go down, work stops. According to the outage tracker Downdetector, reports of Outlook problems peaked at 138,058 around 1 p.m. Eastern time on Thursday, and the disruption spread to 74 countries within about an hour, according to the BBC. By the time Microsoft began restoring service, the incident had become a global event, and a reminder of just how much of daily life now depends on a handful of cloud services.
What Happened
The outage began on Thursday afternoon, Eastern time. According to Vietnam News, Microsoft determined that the disruption started at 1:26 p.m. Eastern time on Thursday, November 12, although the company said it had begun receiving and investigating reports of Exchange Online mailbox connectivity issues in the Americas earlier in the day. Within minutes, reports flooded in from users who could not connect to their Outlook mailboxes, and the problem quickly spread beyond email.
Microsoft acknowledged the issue in a post on X on Thursday, saying it was aware of and investigating reports of users experiencing issues with Microsoft 365 services, primarily involving Exchange Online. Exchange Online is the cloud service that sits behind Outlook email, and it is used by businesses, schools, and government agencies around the world. When Exchange Online fails, email stops working, and so do many of the services that depend on it.
Reports of problems climbed steeply through the afternoon. According to India Today, reports of Outlook problems on Downdetector peaked at about 1:30 a.m. India time on Friday, which was 1 p.m. Eastern time on Thursday, at 138,058 reports. In the United States, Downdetector logged around 97,000 Outlook reports at one point, Reuters reported. The tracker also recorded more than 38,000 reports for Microsoft 365 itself and about 4,500 for Teams. By about 4:30 a.m. India time, the count of Outlook reports had fallen to roughly 34,490, and by 5:47 p.m. Eastern time, reports of problems with Microsoft 365 had fallen to nearly 13,000, with 763 reports for Teams, according to the Korea Times.
The outage lasted for hours. Microsoft 365, Outlook, Teams, and other Microsoft applications remained disrupted for thousands of users for more than five hours on Thursday, the Korea Times reported, and some accounts put the total disruption at seven hours. For many users, the worst of the outage came during the working day, when email and chat are most essential, and the slow recovery meant that some people were still having trouble well into the evening.
Which Services Were Affected
The outage reached far beyond Outlook. Microsoft told users in its service health advisory that customers might be unable to connect to multiple Microsoft 365 services, and it listed Exchange Online, Microsoft Teams, SharePoint Online, Microsoft 365 Copilot, Microsoft Defender XDR, Planner, To Do, Intune, and Microsoft Defender for Office 365 as potentially affected. The company also said that multiple Purview services, including authorization, automatic labelling, data loss prevention, communication compliance, and data classification, were impacted.
Users reported trouble across the Microsoft ecosystem. According to one account of the outage, people had problems with Outlook, Teams, Word, Excel, PowerPoint, SharePoint, OneDrive, and Microsoft’s Copilot assistant. Some users could not access their recently used files in Microsoft 365 apps, and some pages in the Microsoft 365 admin center failed to load. The Microsoft Store and the Azure cloud platform were also reported as affected in some accounts of the disruption.
The outage also spilled over into other companies’ services. The Korea Times reported that the disruption hit connectors used by OpenAI’s and Anthropic’s AI services. OpenAI said it was investigating an issue with its Microsoft connectors in which new connections might be broken and tools were not working as expected. Anthropic said it had identified an issue with an upstream provider affecting its Microsoft 365 connector. The knock-on effect was a reminder that modern software is deeply interconnected, and that a failure in one company’s cloud can ripple through the products of many others.
How Big Was the Outage
By the numbers, this was one of the largest Microsoft outages in recent memory. Downdetector said the disruption reached 74 countries in roughly an hour, according to the Free Press Journal, and users from 72 nations reported problems with Teams, Copilot, OneDrive, Azure, and other services, according to NDTV Profit. Global Outlook complaints peaked at nearly 140,000 around 18:00 GMT on Thursday, more than 2,000 times the normal level, according to the BBC’s account of the tracker’s data. By early Friday GMT, Downdetector put the number of affected users at above 28,000, down sharply from the peak.
It is important to understand what Downdetector’s numbers mean. The tracker counts reports from users, not the total number of people affected. A single report can represent an entire office, school, or company whose email has stopped working, and many affected users never report a problem at all. The real number of people affected was almost certainly far larger than the number of reports, but Microsoft has not published a figure for how many users were impacted, and no official count exists.
The outage also hit India hard. The Free Press Journal reported that Downdetector logged more than 800 problem reports from India by 9 a.m. India time on Friday, and that the disruption had not been fully resolved at the time of writing. In the Philippines, Downdetector data showed that 37 percent of Outlook reports concerned the app, 36 percent the website, and 14 percent trouble receiving messages, according to Philstar Tech. For Teams, 68 percent of reports concerned the app, 19 percent the website, and 10 percent login problems.
What Microsoft Said Caused It
Microsoft has not yet confirmed a final cause for the outage, and the company’s own updates show an investigation that was still evolving hours after the disruption began. But the company did give users a picture of what was going wrong. According to Microsoft’s updates, a group of servers had become unavailable after failing automated health checks, which pushed extra load onto the servers that were still running. The company said a core process on the affected servers was being overwhelmed, causing a loss of capacity.
In a later update, Microsoft said its investigation had identified a process responsible for establishing secure connections on the affected servers as a key factor behind the disruption. The company said it was reviewing diagnostic data, limiting traffic, and rebuilding its servers using a previous software configuration. Microsoft also said it had paused deployments and configuration changes while it checked whether a recent update had played a part. No final cause had been confirmed in the reports reviewed at the time of writing.
Microsoft pointed to servers that had become unexpectedly unavailable under elevated load, and said the problem was not an increase in traffic, according to the Korea Times. That distinction matters. It suggests that the outage was caused by something going wrong inside Microsoft’s own infrastructure, rather than by a surge of users overwhelming the system. The company has not said whether a software update, a configuration change, or a hardware failure was responsible, and a full explanation may not come until the investigation is complete.
The Recovery Effort
Microsoft’s recovery effort began on Thursday afternoon and continued into Friday. The company said it was returning part of the affected servers to service and that they had stayed stable so far. In a post on X, Microsoft said it was bringing some of the affected servers back online and that they remained stable. But the company was careful to warn users that the recovery was not yet secure. In an update at 8:06 a.m. Philippine time on Friday, Microsoft said that the servers it had returned to service could become unresponsive again once they received traffic.
As of 6 p.m. Eastern time on Thursday, Microsoft said it was still working to return affected servers to service and had observed limited recovery as users gradually reconnected, the Korea Times reported. The company said recovery would be gradual as capacity returned and clients reconnected. That gradual pace is typical of a major cloud outage. When a service has been down for hours, millions of devices try to reconnect at once, and that flood of traffic can overwhelm servers that are only just coming back online.
Some services recovered faster than others. Microsoft told users that Teams had been restored in North America, citing the Associated Press, although some people still saw connection problems. Outlook, which had been the hardest hit, took longer to come back, and reports of problems were still coming in well into the evening. By early Friday, the number of Downdetector reports had fallen sharply from the peak, but the outage had not been fully resolved at the time many news reports were written.
The Ripple Effects
An outage of this size does more than stop email. It disrupts the routines of millions of people, and it raises questions about how much the world now depends on a small number of cloud providers. When Microsoft 365 goes down, it is not just one company’s problem. It is a problem for every business, school, hospital, and government office that uses it, and for every person whose working day runs through Outlook and Teams.
For businesses, the outage was a test of their contingency plans. Companies that had backup email systems, alternative chat tools, or offline ways of working were able to keep going. Companies that relied entirely on Microsoft 365 had to wait. The incident is likely to prompt many IT departments to review their plans for the next outage, because in the cloud era, the question is not whether a major service will go down, but when, and for how long.
What Users Could Do During the Outage
For most users, there was little to do but wait. Microsoft’s own advice was to check the Microsoft 365 service health page and the company’s status posts on X for updates, and to use the incident codes EX1550543 and MO1550821 in the admin center for more detail. IT administrators could use the admin center to see the status of their own organization’s services and to communicate with their users about the disruption.
Some users found that switching between the Outlook app and the Outlook website made a difference, since the two can connect through different paths. Others found that messages sent during the outage were delayed rather than lost, and that they arrived once service was restored. But for many people, the only practical step was to switch to a phone call, a text message, or another chat app, and to let colleagues know that email was unreliable.
There are a few sensible steps worth taking after any major outage. First, check whether any messages sent during the disruption actually went through, and resend anything that did not. Second, check your calendar, since meeting invitations and updates sent during the outage may have been delayed or duplicated. Third, if you use two-factor authentication or sign-in codes delivered by email, be aware that those messages may have been delayed as well. And finally, if your workplace has an IT team, let them know about any problems that persist after Microsoft says service is restored, because individual accounts can sometimes need a fresh sign-in or a restart to reconnect cleanly.
Frequently Asked Questions
What caused the Microsoft 365 outage on November 13, 2026?
Microsoft has not confirmed a final cause. The company said a group of servers became unavailable after failing automated health checks, which pushed extra load onto the remaining servers, and that a core process on the affected servers was being overwhelmed, causing a loss of capacity. Microsoft later said its investigation had identified a process responsible for establishing secure connections on the affected servers as a key factor. The company said it was reviewing diagnostic data and rebuilding servers using a previous software configuration, and it had paused deployments and configuration changes while it checked whether a recent update played a part.
Which Microsoft services were affected by the outage?
Microsoft listed Exchange Online, Microsoft Teams, SharePoint Online, Microsoft 365 Copilot, Microsoft Defender XDR, Planner, To Do, Intune, and Microsoft Defender for Office 365 as potentially affected, along with multiple Purview services. Users also reported trouble with Outlook, Word, Excel, PowerPoint, OneDrive, and the Microsoft 365 admin center. The outage also affected connectors used by OpenAI’s and Anthropic’s AI services. Outlook and Exchange Online were the hardest hit, with Downdetector reports for Outlook peaking at 138,058 around 1 p.m. Eastern time on Thursday.
How long did the Microsoft 365 outage last?
The disruption began at about 1:26 p.m. Eastern time on Thursday, November 12, according to Microsoft’s own account, and Microsoft 365, Outlook, and Teams remained disrupted for more than five hours, with some accounts putting the total at seven hours. Microsoft began bringing servers back online during the afternoon and evening, and by early Friday the number of Downdetector reports had fallen sharply from the peak. Microsoft said recovery would be gradual and had not given a specific time for full restoration at the time of writing.
How many people were affected by the outage?
There is no official count of the total number of people affected. Downdetector, which counts user reports rather than affected users, logged a peak of about 138,058 Outlook reports around 1 p.m. Eastern time, more than 38,000 reports for Microsoft 365, and about 4,500 for Teams. The disruption spread to 74 countries within about an hour, and global Outlook complaints peaked at nearly 140,000, more than 2,000 times the normal level, according to the BBC. The real number of affected users was almost certainly far larger than the number of reports, since one report can represent a whole office or company.
Related reading
- How the Internet Was Born in America: The ARPANET Story
- How the U.S. Power Grid Delivers Electricity Across America
- GPS: How American Satellite Power Guides the World
Sources: Philstar Tech, The Free Press Journal, Kahaan Kya Hua, The Korea Times, Vietnam News, NDTV Profit, Jaano Junction.