Microsoft 365 Outage: Outlook, Teams and Copilot Went Down Worldwide — What Happened and What to Do

For more than a day, email simply stopped working for millions of people. On Thursday, November 12, 2026, Microsoft 365 began failing worldwide, and by Friday morning the outage had become one of the biggest cloud disruptions of the year: Outlook would not load, sent mail sat undelivered, Teams calendars broke, and Microsoft 365 Copilot went dark alongside them. Downdetector logged roughly 130,000 Outlook problem reports at the peak, from users in more than 70 countries. Microsoft says it was not a cyberattack. The cause, it says, was an issue during an earlier network capacity upgrade that surfaced after a recent network change — and by Friday afternoon, most services were back.

This is the story of what broke, why it took so long to fix, and what to do the next time your inbox goes silent.

When the Outage Started

The first signs appeared on Thursday afternoon. Microsoft announced it was investigating reports of users unable to access Exchange Online mailboxes in the Americas at around 10:40 a.m. ET on November 12, and within about two hours it had opened a wider incident — tracked as MO1550821 for Microsoft 365 and EX1550543 for Exchange Online — warning that users in any region could be affected. Exchange Online users in the Americas began losing access to their mailboxes at about 16:17 UTC.

From there it spread. By Thursday evening, Downdetector had logged more than 120,000 Outlook problem reports at the peak, with Microsoft 365 reports exceeding 47,000. Reports came in from 74 countries, and the disruption hit GP surgeries, pharmacies, schools, police forces and thousands of small businesses, not just office workers. Microsoft’s own status page showed the scale plainly: this was not one app misbehaving, it was the connective tissue of Microsoft 365 failing at once.

What Broke — and What Kept Working

Data center corridor at night with server racks glowing red with warning lights and a network cable overloaded with light pulses, illustrating the network capacity failure behind the outage

The list of affected services was long, which is exactly why the outage hurt so many organizations. Microsoft listed Outlook on every client — web, desktop and mobile — along with Exchange Online mail delivery, Teams calendars and scheduling, SharePoint features that rely on Exchange, Microsoft 365 Copilot, Microsoft Graph, Planner, To Do, Intune, Purview and Power BI. The Microsoft 365 admin center and parts of Microsoft’s own security stack, including Defender XDR, were also affected.

What kept working matters too, because it shows where the fault line ran. Core Teams messaging and online meetings remained largely functional, and Microsoft said services that do not depend on Exchange Online kept working. The outage hit Microsoft 365 services “that interact with Exchange Online and email services” — a reminder of how much of the modern workplace still runs through the mailbox.

Microsoft’s investigation found that a process on affected servers responsible for establishing secure connections was becoming overwhelmed, causing the servers to stop processing requests. In its Friday afternoon update, the company put it more plainly: “This was a technical issue not related to cyber or AI/agents. The incident was introduced by an issue during an earlier network capacity upgrade that surfaced following a recent network change.” It has promised a full post-incident report once its review is complete. You can follow the official status at Microsoft’s 365 status page.

Why It Took More Than a Day to Fix

Recovery was slow and uneven. On Friday morning, Microsoft reported “early signs of stabilisation,” with traffic being “carefully reintroduced” — but capacity had not returned to normal, and some servers went down again after being brought back. Early Friday, Microsoft said changes intended to improve stability were not working as expected. By around 11:30 p.m. Thursday, more than 18,000 users in the United States were still reporting Outlook problems, down from the peak but far from normal.

The scale of the fix explains the delay. Microsoft had to restore capacity server by server while traffic was reintroduced gradually, and it extended the email timeout window to 48 hours to protect messages in transit during the recovery. By about 3 p.m. ET on Friday, most services were back online, the service health page showed services as “operational,” and Downdetector reports had dropped to roughly 1,600 for Outlook — down from 130,000. Microsoft warned that some users might still see issues with 365 features as servers continued to stabilise through the evening, and people accessing Outlook.com through a web browser could still be affected.

The timing could hardly have been worse for Microsoft. The outage came just days after the company’s November Patch Tuesday, which fixed 1,094 flaws including an Outlook zero-day — a week in which Outlook was already in the security headlines for the wrong reasons. And it landed while Microsoft is pushing Copilot deeper into everyday work, which meant the outage took the AI assistant down with the email it is meant to help manage. Meanwhile, the competition is not standing still: Anthropic’s Claude Sheets, Meetings, Whiteboards and Data suite is aimed squarely at Microsoft’s office turf, while IBM and Microsoft’s Imagine Enterprise platform is rolling out 6,000 prebuilt AI agents for regulated industries — which makes a day without Copilot a day the rivals can point to.

Is It a Cyberattack? No — and Why That Question Matters

Every major cloud outage now triggers the same first question, and Microsoft answered it quickly: it said it was “not aware of any malicious activity,” and a senior Microsoft executive confirmed no security breach was involved. The cause was an internal network change, not an intruder.

That distinction matters beyond reassurance. A cyberattack would mean stolen data, compromised accounts and a very different recovery — password resets, forensic audits, regulatory disclosures. A capacity failure means the data was never at risk, only unreachable. The lesson for IT teams is different in each case, and this time the lesson is about resilience: when email, calendars, documents and AI assistance all depend on the same cloud, a single infrastructure fault can take the whole workday with it. The OpenAI Daybreak red-team and defense work published the same week is a useful reminder that the industry is thinking hard about AI-era threats — but this outage was a reminder that the oldest failure mode, a bad network change, is still the one that bites. A detailed timeline of the incident is available in this outage advisory and this SourceIT timeline.

What to Do During a Microsoft 365 Outage

Professional at a desk holding a smartphone showing a green service status checkmark, with a laptop and checklist nearby, illustrating how to check Microsoft 365 service status during an outage

If your inbox goes dark again, the first step is to confirm it is Microsoft, not you. Check Microsoft’s official status page at status.cloud.microsoft, or the Service Health page in the Microsoft 365 admin center if you can reach it. Administrators can follow the incident by its ID — MO1550821 for the wider Microsoft 365 incident, EX1550543 for Exchange Online. Downdetector is a useful second opinion, but Microsoft’s own status page is the source of truth.

Second, do not keep hammering the send button. During this outage, Microsoft extended the email timeout window to 48 hours specifically so messages would not bounce while mail flow recovered. Queued mail will generally go through once service returns; resending the same message repeatedly just adds to the backlog.

Third, have a fallback ready before you need one. Keep a secondary channel — a personal email address, a phone tree, or a chat app outside the Microsoft ecosystem — for time-critical messages, and make sure your team knows when to switch to it. The organizations that coped best this week were the ones that did not have to improvise that plan mid-outage.

Finally, once service is back, verify before you trust it. Check that sent mail actually left, that calendar invites went out, and that shared documents synced. Early recovery is when the odd message slips through the cracks, and a quick spot-check beats discovering a missing email on Monday.

What to Watch Next

Three questions remain open. First, what exactly went wrong in the network capacity upgrade — Microsoft’s post-incident report will be the final word, and it will show whether this was a one-off error or a process gap that could recur. Second, how quickly the last pockets of degradation clear, and whether mail flow fully normalises over the weekend. And third, what Microsoft changes afterwards: the company has promised a full report, and the industry will be watching for concrete changes to how it tests network changes before they reach production.

The broader signal is hard to miss. Microsoft 365 is now the operating system of the working day for hundreds of millions of people — and when it goes down, it is not an IT problem, it is a business-continuity problem. The companies that treat it that way, with a tested fallback plan and a clear way to check official status, will be the ones that keep working next time the cloud blinks.

Conclusion

The Microsoft 365 outage of November 12–13, 2026, took Outlook, Teams scheduling, SharePoint and Copilot offline for more than a day, peaked at roughly 130,000 user reports, and was caused by a network capacity upgrade issue — not a cyberattack. Most services were restored by Friday afternoon, with a full post-incident report still to come. For users, the takeaway is simple: know where to check the official status, give queued mail time to flow, and have a backup way to reach your team. The cloud is reliable until it isn’t, and the next outage will not send a calendar invite.

Leave a Comment

Your email address will not be published. Required fields are marked *

Scroll to Top