If your work email went silent on Thursday, you were far from alone. A major Microsoft outage knocked out Outlook, Teams, Copilot and other Microsoft 365 services for users around the world, and recovery was still underway on Friday morning.

Here’s what happened, what Microsoft has said about the cause and what businesses and everyday users can take away from it.

What went down

Problems began piling up on Thursday, November 12. According to The Guardian, the outage-tracking site Downdetector saw a spike in Microsoft 365 reports at about 6 p.m. GMT (1 p.m. ET), with nearly 130,000 problem reports from 74 countries.

Affected services included Word, Excel, PowerPoint, Outlook and Teams, plus business tools such as SharePoint and OneDrive. Microsoft’s AI assistant, Copilot, was hit as well. The disruption was patchy, with some users reporting problems while others were unaffected.

The scale matters. Microsoft says more than 430 million people use Microsoft 365 apps, and more than 90% of Fortune 500 companies use Copilot, according to The Guardian.

What Microsoft says caused it

Microsoft said there were no indications the outage was caused by a cyberattack.

Instead, the company said a number of servers had become “unexpectedly unavailable” because of a problem that caused them to fail automated health checks, and that part of its infrastructure was “experiencing elevated load.” Microsoft has also suggested the trouble may be linked to settings applied to servers that keep Microsoft 365 running.

Microsoft’s executive vice president of Copilot, Charles Lamanna, posted on X that “we regret that today’s outages caused disruptions for our customers, and we are working to address the issue.”

Microsoft has promised a post-incident report with full details of the root cause once the problem is fixed.

Where recovery stands

Early fixes brought only “marginal success,” Microsoft said at one point. By Friday morning the tone had improved. In an update shortly after 9 a.m. GMT, the company said it was seeing “early signs of stabilization,” though service capacity had not yet returned to normal and health remained variable as traffic was carefully reintroduced.

To protect delayed messages, Microsoft said it had extended the email timeout window to 48 hours. The Washington Post also reported the outage continuing into a second day.

Who felt it

The ripple effects were global. The Guardian reported that the UK’s National Health Service saw disruptions to services including remote therapy sessions, and some pharmacies struggled to reach GP surgeries about prescriptions. Several UK police forces asked people to phone in reports instead of using online forms, and some schools moved attendance reporting to the phone.

In Asia, Indian airlines IndiGo and Akasa Air said their email support was affected, and a Japanese shipping company and a major pharmaceutical firm moved some orders to phone. The outage even reached AI tools from OpenAI and Anthropic that connect to Microsoft 365.

Why outages like this cascade

Mark Gregory, an associate professor at RMIT University in Melbourne, told The Guardian that when some servers fail, users are forced onto a smaller pool of remaining servers. “If you lose part of that pool, then all of the users are trying to access the remaining servers, and that overloads them. And then they go down,” he said. That domino effect helps explain why the recovery has been slow and uneven.

What you can do

For individuals, the best move is patience: delayed emails should still arrive, thanks to the extended timeout window, and Microsoft’s service health page posts updates. For businesses, experts say this is a reminder to keep backup ways to communicate, such as a phone tree or a secondary messaging tool, so one provider’s bad day doesn’t halt everything.

The takeaway

Microsoft’s outage is a reminder of how much of modern work runs through a handful of cloud services. Recovery appears to be moving in the right direction, but until Microsoft publishes its full root-cause report, the big question remains how a server configuration problem managed to ripple across 74 countries.

Image: Apple Macbook by Jay Wennington via StockSnap, CC0 1.0