Microsoft Blames a Network Upgrade for Its Day-Long Outlook Outage. Businesses Can Claim Up to 50% Back.

For most of Thursday afternoon and a good chunk of Friday, millions of people opened Outlook and got nothing. Teams calls dropped, Copilot spun in circles, and SharePoint links timed out. Now Microsoft has finally said what broke, and as UNILAD Tech reported, the explanation is far more boring than the cyberattack theories that flooded social media.

It wasn’t hackers. It wasn’t AI agents. It was a network upgrade that went sideways. And if your business pays for Microsoft 365, that boring answer comes with something useful attached: you may be owed money back, but only if you ask for it.

How the outage unfolded

The first sign of trouble showed up on Downdetector at around 12:39 p.m. ET. Fifteen minutes later, the Microsoft 365 Status account on X confirmed it was “investigating reports of users experiencing issues accessing Exchange Online mailboxes through multiple connection methods in the Americas.” By 1:09 p.m., Downdetector had more than 129,000 error reports, and the problem had spread well beyond email.

That 2:12 p.m. update was the first real clue. Microsoft said it had “paused all deployments,” was moving traffic off the struggling infrastructure and restarting servers. Crucially, it said the strain wasn’t driven by more people using the service. It came from “a number of servers being unexpectedly unavailable; straining the remaining infrastructure.”

What actually broke

In its Friday statement, Microsoft was blunt: “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.”

In plain English, an upgrade made earlier left a hidden problem in the network. It sat there quietly until a later change triggered it. That kind of delayed fault is a nightmare to diagnose, because the change that sets it off often looks harmless on its own.

Mahmoud Elkhodr, a senior lecturer in information and communication technologies at CQUniversity Australia, offered a likely explanation for why things spiraled the way they did. Writing for The Conversation, he pointed at automated health checks. These are systems that constantly ping servers and pull out any that look unwell. Normally that protects a service. But if the checking system itself is getting bad information, it can start removing healthy servers. The ones left behind then have to absorb far more work than they’re built for, and they fall over too.

That lines up with what Microsoft said early on, when it mentioned that some servers were failing automated health checks and pointed to a component responsible for processing server health. Elkhodr was careful to note that until Microsoft publishes its full post-incident review, this is “an inference drawn from Microsoft’s own updates rather than a settled account.”

Why it took so long to fix

If the fix was found Thursday night, why were people still locked out on Friday? Professor James Ferryman from the University of Reading explained it neatly to the BBC: “This kind of failure snowballs. One server falls over, its work gets passed to the next one, and meanwhile huge numbers of phones and laptops keep trying to get back in. You can’t just turn it all back on at once. You have to bring it back bit by bit.”

Think of every Outlook app on every laptop and phone hammering the door at the same moment. If Microsoft had flipped everything back on at once, that flood of reconnections could have knocked the servers straight back over. So engineers drained traffic, restarted, patched and returned infrastructure to service in stages. Microsoft admitted this “took longer” but said it “helped ensure a stable restoration.”

Did my emails disappear?

Almost certainly not. Mail servers are designed to keep retrying when they can’t reach a recipient. Exchange Online normally keeps trying for 24 hours before it gives up and sends the sender a bounce message. During this outage, Microsoft doubled that window to 48 hours so queued messages wouldn’t expire while engineers worked.

That means most email sent to you on Thursday and Friday should have arrived late, often in bursts as your mailbox server came back online. If something important never showed up, the sender should have received a non-delivery report. If you’re catching up on a messy inbox, our guide on getting around Outlook quickly covers search, filters and focused inbox tricks that make the cleanup faster.

Who got hit

Microsoft says more than 450 million commercial seats use Microsoft 365, so the ripple effect was huge. Some of the more notable examples reported so far:

Rep. Josh Gottheimer of New Jersey, who worked at Microsoft before joining Congress, told CNBC he’d asked for a briefing. “Is there something that went on that’s nefarious here, or is this just something else?” he said, pointing to the size, length and timing of the outage. Microsoft’s stock, which dipped early Friday, ended the day largely flat.

How much did it cost?

Nobody knows yet, and you should be skeptical of anyone giving you a precise number. Some early estimates simply multiplied a guessed number of affected workers by an hourly wage. For context, insurer Parametrix estimated that the July 2024 CrowdStrike incident, which crashed about 8.5 million Windows machines, cost Fortune 500 companies around $5.4 billion. This outage was a different animal. Laptops kept working, but the tools people use to talk to each other didn’t.

Ashwin Pal, a risk advisory partner at RSM Australia, told ABC News the hidden cost is the cleanup: “It’s not just the initial downtime, it’s more actually about then the catch up effort because that costs money.”

How to claim a service credit

This is the part most businesses will miss. Microsoft’s service level agreement promises 99.9% monthly uptime for Exchange Online and most Microsoft 365 services. When it falls short, eligible customers can get a percentage of that service’s monthly fee back. Here’s how the tiers work out in a 30-day month like November:

If your whole organization was locked out of email for 18 to 24 hours, that would land you in the 50% tier for Exchange Online. If only some of your staff were affected, your number will be lower, because Microsoft calculates uptime across all your users. Either way, it’s worth filing. Here’s what Microsoft asks for:

• Who can claim: Business, education and enterprise tenants on paid plans. Consumer Outlook.com and Microsoft 365 Personal or Family users aren’t eligible.
• Where to file: In the Microsoft 365 admin center, open a support request and ask for an SLA credit for the November 12 incident. Partners can file on behalf of customers through Partner Center.
• What to include: Your tenant details, the services affected, and evidence of how the outage hit your operations.
• The catch: According to Microsoft’s partner guidance, evidence emails “must come from the domain of the affected tenant.” Email from a personal address isn’t accepted.
• It isn’t automatic: No claim, no credit. The credit applies to the affected service, not your entire Microsoft 365 bill.

Is this becoming a pattern?

It’s not the first rough patch for Microsoft 365 this year. A January outage knocked out Outlook, Teams and other services for hours, and an August incident was tied to core authentication configuration. Elkhodr’s broader point is that so much of the world now runs on a few shared platforms that we should plan for digital outages the way we plan for storms.

For a business, that means having a second way to reach staff and customers that doesn’t run through the same provider, whether that’s a phone tree, a group text or a status page hosted somewhere else. The same logic applies at home, as we covered in our look at what still works when the network goes dark.

About this article: GeekBlog covers U.S. technology news, AI, phones, smartwatches and gaming. Every story is written and checked under our Editorial Policy. Spotted a mistake or have a story tip? Contact our editors.