Microsoft 365 went down on 13 November. It is rare, and it will happen again.

What happened on 13 November

Microsoft says the problems began at about 5:15am Melbourne time on Friday 13 November, which was Thursday lunchtime on the US east coast. Microsoft listed Exchange Online (the service behind Outlook), Microsoft Teams, SharePoint Online, Microsoft 365 Copilot and Defender XDR among the services that could be affected, and warned that users in any region could be caught up in it, according to BleepingComputer. Australian businesses felt it as their working day started: inboxes that would not load, Teams chats that would not send, and files in SharePoint that would not open.

Microsoft's explanation changed during the morning. It first pointed to elevated load, then said the load was not caused by a surge in traffic but by a large number of “servers being unexpectedly unavailable”, which pushed the work onto the servers that were left, as the ABC's live coverage reported. Microsoft paused its deployments, moved traffic away from the affected infrastructure and restarted servers. Its later updates said affected servers were failing automated health checks, and that a different way of checking their health was letting them return to service. By around 9am it was reporting servers coming back and working to speed up the recovery.

Microsoft has not yet published a root cause. For incidents with broad impact it commits to a preliminary post-incident review within 48 hours of resolution and a final one within five business days, published to administrators through Service health.

How rare is a Microsoft 365 outage this big?

Rare, but not as rare as it used to be. Microsoft publishes its worldwide uptime each quarter. From 2022 to the end of 2025 it never fell below 99.9%, and most quarters sat between 99.97% and 99.99%. Then the first quarter of 2026 came in at 99.526%, the lowest figure in Microsoft's published table, after an outage on 22 January that The Register reported lasted around ten hours.

So the honest summary is this. On an ordinary day Microsoft 365 is more reliable than almost anything a small business could run for itself. Twice this year it has not been ordinary. A service that works 99.9% of the time is still allowed around 43 minutes of downtime a month, and when it fails, it fails for millions of businesses at once. Rare and impossible are different things, and a plan is for the difference.

What should a business do when Microsoft 365 is down?

Most of the cost of an outage is not the outage. It is the morning spent troubleshooting something that cannot be fixed from your end, and the clients who did not know why nobody was answering. In order:

• Check before anyone troubleshoots. Microsoft's public status page, status.cloud.microsoft, and Service health in the admin centre tell you within a minute whether the problem is Microsoft's. If Microsoft has not posted anything yet, Downdetector is a useful alternate check: it shows whether other people in Australia are reporting the same problem right now. Treat it as a signal rather than confirmation, because it counts user reports, not confirmed faults. If the problem is Microsoft's, resetting passwords, reinstalling Outlook or restarting the router will not help and may create new problems.
• Use what still works. Outlook keeps a local copy of recent email, and files synced through OneDrive can usually still be opened on the computer they are synced to. Phones, mobile numbers and paper still work. Most teams can do a few useful hours of work offline if someone says so.
• Tell clients early. A short notice on your website, a phone message or a text to key clients stops the same question arriving fifty times. We put an outage notice at the top of our own website this morning for exactly that reason.
• Do not panic about lost email. Mail sent to you during an outage is normally queued and retried by the sender's server, and the email standard expects senders to keep trying for at least four to five days. Expect late, not lost.
• Check afterwards. Once service returns, look for anything you sent that bounced, meetings that did not happen, and messages that arrived out of order.

What the outage should remind every business about its technology

Moving to the cloud moved the servers out of the back room. It did not move the responsibility for keeping the business running. Five things are worth a fresh look this week.

Know what depends on Microsoft 365

Email and Teams are obvious. Less obvious are the files your team keeps in SharePoint, any phones that run through Teams, the other business apps your staff sign into with their Microsoft account, and the AI tools that read your mailbox. Several AI tools that connect to Microsoft 365 reported problems of their own today. Write the list down, because it is the list of things that stop when Microsoft 365 does.

Keep the plan somewhere other than Microsoft 365

A continuity plan saved in SharePoint is no use on the morning SharePoint is down. One printed page in the office and a copy on the owner's phone is enough: who checks the official status, who tells clients and how, how staff reach each other, and what they can keep working on.

Have a second way to reach your team and your clients

A current list of staff mobile numbers, a group text, and the ability to forward your main number to a mobile if your phones depend on the same platform. None of it costs anything, and all of it is hard to organise at 7am on the day you need it.

Understand what the service level agreement covers

Microsoft's service level agreement commits to 99.9% uptime per service each month and offers service credits on your subscription when it misses. A credit on a licence bill does not pay for a morning of lost work across a 20-person team. Continuity is the business's plan to make, not Microsoft's.

Remember that an outage is not data loss

Today was about access. Your email and files were still there, you just could not reach them for a while. Backup protects against a different problem: deletion, ransomware and mistakes, where data really is gone. Both deserve a plan, and they are not the same plan. We have written about why so much data loss comes from backups that fail.

Do not overreact either

The wrong lesson from today is that Microsoft 365 cannot be trusted. For a business of 5 to 50 staff, a second email system kept on standby, or a server back in the comms cupboard, adds cost and complexity and brings its own outages, usually ones with nobody else working on the fix. The right lesson is smaller and cheaper: decide in advance what happens on the rare day the platform is unavailable.

Picture a 20-person accounting practice in Frankston this morning. With no plan, the first hours go on restarting computers and wondering whether it is a virus, while clients email into a queue nobody can see. With one page and ten minutes of preparation, someone checks the status page, a notice goes on the website, the phones are answered, and the team works on what they can until Microsoft recovers. The outage is the same. The morning is not.

Where to start

• Write the one page this week, while today is fresh: what depends on Microsoft 365, who checks the status, who tells clients, and how staff stay in touch.
• Put it on your next review agenda. Continuity is part of business technology management, and it belongs in the same conversation as your roadmap and your quarterly business review.
• Ask your IT provider what they did this morning. If the answer is that they waited for your call, it is worth a conversation about what a managed service should cover. Ours is AgileMANAGED.

We are based in Mornington and work with businesses across Melbourne and the Mornington Peninsula. If you would like a hand writing that one page, that is what a conversation is for, no obligation.

Frequently asked questions