On Thursday, November 12, 2026, a worldwide Microsoft 365 outage began at 1:26 p.m. Eastern (Microsoft incident MO1550821). It affected Exchange Online (Outlook email), Teams, SharePoint Online, Microsoft 365 Copilot and Defender XDR, among other services. As of Microsoft’s 7:00 p.m. Eastern update that evening, the outage was still under way: servers brought back into service were failing again under traffic, and no root cause had been named.
For most offices it started with Outlook, which stopped connecting a little after 1 p.m. Teams calls wouldn’t start, SharePoint files wouldn’t open, and for the next several hours a lot of people found out how much of their workday runs through one company’s servers.
If that was your office, nothing on your end was broken. This post covers what Microsoft has said so far, what’s worth checking now, and a few cheap things that make the next outage a slower afternoon instead of a lost one.
What happened to Microsoft 365 on November 12
Microsoft’s incident notice in the Microsoft 365 admin center (number MO1550821, if you want to look it up) puts the start at 1:26 p.m. Eastern. It names Exchange Online (the service behind Outlook email), Teams, SharePoint Online, Microsoft 365 Copilot and Defender XDR, and the longer list underneath includes Planner, To Do, Intune, Purview, parts of Power BI, Universal Print (Microsoft’s cloud printing service) and, for some setups, Windows 365 Cloud PCs. Earlier the same morning a separate Teams problem had delayed messages in North and South America for about an hour, as BleepingComputer reported. Microsoft hasn’t said whether the two were related.
The notice was updated through the afternoon, and the explanation changed as it went:
- 1:27 p.m. Part of Microsoft’s infrastructure is under heavy load in several regions. Engineers start restarting it and look at moving traffic elsewhere.
- 2:30 p.m. The load isn’t coming from more traffic. A large number of servers have unexpectedly dropped out of service, and the ones still running are carrying their weight. Microsoft pauses all of its own deployments and configuration changes as a precaution.
- 2:58 p.m. The servers are failing the automatic health checks that decide whether a server stays in service, and the fault may be in the part of the server that answers those checks.
- 3:57 p.m. A different way of running the health checks brings one group of servers back, and Microsoft starts applying it more widely. A few recent configuration updates are under review as possible contributors.
- 5:03 p.m. Microsoft says its actions so far “have not provided the anticipated relief.” A core process on the affected servers is being overwhelmed, which makes them stop responding.
- 5:58 p.m. Some servers are back in service and holding steady, with limited recovery in some regions. Microsoft says to expect a gradual return as each program reconnects.
- 7:00 p.m. The servers that came back look healthy at first, then stop responding again once traffic reaches them. Microsoft starts limiting which kinds of traffic can reach the affected infrastructure in some locations, to find out whether one type of request is behind it, and rebuilds a group of servers on an earlier software configuration to see whether that holds.
As we write this on Thursday evening, the notice still reads “Service degradation” and Microsoft hasn’t named a root cause. We’ll update this post when Microsoft confirms the fix and the cause.
How to tell whether it’s Microsoft or you
The first twenty minutes of an outage are usually spent restarting Outlook, then the computer, then the router. There’s a faster way to find out.
- The public status page. status.cloud.microsoft works without signing in, which matters on a day when signing in is the problem.
- Service health in the admin center. Whoever administers your Microsoft 365 can open Health, then Service health, and see each incident with Microsoft’s running updates. Microsoft’s own instructions are here. Admins can also have new incidents emailed to them, though it’s worth sending those to an address that doesn’t live in Microsoft 365.
- Microsoft 365 Status on X. The @MSFT365Status account posts the short version during big incidents.
If Microsoft shows an incident that matches what you’re seeing, stop troubleshooting. Reinstalling Office or resetting passwords in the middle of an outage adds a second problem to the first.
What to check now that it’s coming back
Email that was sent to you during the outage is probably late, not lost. The standard that mail servers follow says a sending server should keep retrying for at least four to five days before giving up (RFC 5321, section 4.5.4.1). Most messages from Thursday afternoon will arrive on their own, out of order. If you were expecting something time-sensitive, a signed contract or a bid, call the sender instead of waiting.
Look in the Outbox. Messages written while Outlook was disconnected may still be sitting there. Open Outlook on each computer and make sure they went.
Check anything that runs on a schedule. Print jobs sent through Universal Print (it was on Microsoft’s list), scan-to-email from the copier, invoices that go out automatically, forms that email a shared mailbox, voicemail that arrives as email. These fail quietly, and nobody notices until a customer asks why they never heard back.
Be suspicious of outage emails. A message saying your mailbox was affected and you need to sign in to restore it should be treated as phishing. A big outage in the news gives scammers a believable story, and nothing about this one requires you to sign in anywhere new.
What Microsoft’s 99.9% uptime guarantee owes you
Microsoft’s Service Level Agreement promises 99.9% uptime for Microsoft 365, measured by the month. That sounds like nearly always, so it’s worth doing the math. A 30-day month has 43,200 minutes, and 0.1% of that is 43 minutes. Thursday’s outage ran for hours, so for many customers November will fall short.
When a month falls short, the remedy is a credit against that month’s bill for the affected service. Two things to know about it. The credit isn’t automatic: someone has to file a claim, so ask whoever you buy Microsoft 365 from how and by when. And it’s a share of a monthly license fee, which for a small office is a modest number next to an afternoon of twenty people not working. It’s worth claiming, and it’s not a plan.
Four things that make the next Microsoft 365 outage easier
Nobody outside Microsoft can keep Microsoft 365 running. What an office controls is how well it works for a few hours without it.
1. A contact list that doesn’t live in Outlook
Key customers, vendors, staff cell numbers and your IT provider’s phone number, printed or saved somewhere that isn’t Microsoft 365. With one, an email outage means picking up the phone. Without one, the phone numbers are locked in the thing that’s down.
2. A second way to talk to each other
If Teams is your phone system and your chat, an outage takes both. Decide ahead of time what the fallback is, whether that’s a separate phone system, a group text, or desk phones that don’t depend on Teams, and tell everyone before the day they need it.
3. Files you can open offline
OneDrive can keep chosen folders on the computer so they open without a connection. Pick the handful of documents you couldn’t work a day without (price lists, the schedule, forms) and mark them to always stay on the device.
4. Somebody whose job it is to say “it’s Microsoft”
Most of the time lost in an outage goes to people each working out separately that it isn’t their computer. One person, in the office or at your IT provider, should check the status page and tell everyone within a few minutes: it’s Microsoft, it isn’t us, use the fallback.
What a backup does and doesn’t do here
A fair question after a day like Thursday is whether a Microsoft 365 backup would have helped. It wouldn’t have: Microsoft hasn’t reported any data lost, and a backup doesn’t send email while the service is down. A backup is for a different day, the one when a mailbox is deleted, a folder is encrypted or someone leaves and takes files with them. You want one, for those reasons, but it wasn’t the missing piece this week.
Where we come in
We manage Microsoft 365 for offices across Central Florida as part of Managed IT, and we install business phone systems too. If Thursday showed you a gap, such as no fallback, nobody watching service health or a copier that stopped scanning to email, our free technology review is a reasonable place to start. And if something still isn’t working now that Microsoft says it’s fixed, send us a message or call.