Microsoft 365 outage: what it showed about backup and business continuity

Microsoft 365 outage: what it showed about backup and business continuity
In short: Microsoft 365 was disrupted on November 12 and 13, 2026, and for many companies email, files, chat and meetings went down together. You can't prevent a vendor outage, but you can plan for one. That means checking that your backups caught up, testing a restore, setting up a way to communicate outside Microsoft 365, and writing a short outage plan before the next one.
On Thursday, November 12, Microsoft 365 started failing for users around the world. Outlook, Teams, SharePoint, OneDrive and Copilot were all affected, and for many organizations the problems ran into Friday.
The outage is over. The useful question now is what it showed about your setup, and what you'd do differently next time. Below we cover what happened, the two risks it brings up, and five steps to take this week.

What caused the Microsoft 365 outage
The outage began Thursday afternoon and degraded access to numerous Microsoft 365 services into Friday, November 13, according to CIO Dive. Microsoft tracked it in the Microsoft 365 admin center as incident MO1550821, alongside a related Exchange Online incident, EX1550543.
On Thursday, Microsoft pointed to elevated load across multiple regions as a likely contributor to the connection errors. On Friday it gave a fuller explanation: the problem was introduced during an earlier network capacity upgrade and surfaced after a more recent network change. Microsoft said it was not a cyberattack and was not caused by AI agents. It apologized to customers and said it will publish a full post-incident report once its review is complete.
It also wasn't the first this year. On January 22, Microsoft resolved a separate outage that took down Outlook, Teams and other Microsoft 365 services.
Outages and data loss are different problems
It helps to separate two risks that often get lumped together.
An outage is about access. Your data still exists, but you can't reach it or work with it until the service comes back. You can't prevent a vendor outage. You can only plan for it.
Data loss is about recovery. Files get deleted, overwritten, encrypted by ransomware or removed when someone leaves. Microsoft 365 includes retention and recovery features, but they live inside Microsoft's cloud, under the same admin accounts as the data they protect. An independent backup keeps a separate copy outside that environment, so one failure, one compromised admin account or one bad deletion can't take out both.

A good plan covers both: a way to keep working during an outage, and an independent backup you can restore from after something goes wrong.
What the outage exposed
One vendor, many services
Most mid-market companies run email, file storage, chat, video meetings and often sign-in through Microsoft 365. That's efficient, and on most days it's the right choice. But it means one failure reaches almost everything.
In a research note covered by CIO Dive, Gartner said this kind of tightly coupled setup "transforms what should be an isolated infrastructure failure into an enterprisewide operational disruption."
Backups depend on the service too
Cloud backup tools connect to Microsoft 365 to copy your data. When Microsoft 365 is down, new backups can't run until it recovers. Your existing backups stay safe, but once the service returns there's a catch-up period.
That's normal and expected. The risk is not knowing whether the catch-up finished. If nobody checks, a gap can sit unnoticed until the day you need a restore.

Most teams had no plan B
When email and chat go down together, the usual ways people hear about an outage go down with them. One of Gartner's recommendations after this outage was to set up out-of-band communications and fallbacks, meaning a way to reach people that doesn't run through the system that failed. If your managers' first question on Thursday was "how do we reach everyone?", that's the gap to close.
AI tools went down with it
AI assistants that connect to Microsoft 365 also reported connection problems during the outage. Every tool you plug into your Microsoft 365 data inherits the same dependency, and that list is growing fast.
1. Confirm your backups caught up
Open your backup reports for November 12 to 14. Check that every Microsoft 365 job (mailboxes, OneDrive, SharePoint and Teams) completed after the service recovered. If something failed or is still running, rerun it now and note how long the catch-up took. That number belongs in your outage plan.
2. Test a restore
Pick one mailbox and one OneDrive or SharePoint folder and restore them to a test location. Time how long it takes and check that the files open. Until you've restored from a backup, you don't really know it works.
3. Set up an IT outage communication plan
Choose a channel that doesn't depend on Microsoft 365, such as a phone tree or a text group for managers. Keep a current contact list somewhere people can reach without their work email, and decide who sends the first message.
4. Write a one-page outage plan
Keep it short. It should answer four questions:
- Who decides that it's an outage and switches to plan B?
- Who tells customers and partners, and how?
- Which work can continue offline or on other tools?
- Where does the plan live, so it's reachable when Microsoft 365 isn't?
5. List your single points of failure
Gartner's advice to CIOs includes auditing how much you depend on one vendor for areas like email and authentication, owning independent backups, and stress-testing AI workload dependencies before you scale them. Start simple: list each critical business process and the vendor it depends on. Anything with only one name next to it is a single point of failure.
None of this means leaving Microsoft 365. It means knowing what happens when it has a bad day.
What a Microsoft 365 continuity plan needs
A plan isn't complete until each of these is in place:
- Independent backups: a copy of mailboxes, OneDrive, SharePoint and Teams stored outside Microsoft's cloud.
- Verified catch-up: a check after every outage that backup jobs finished.
- Tested restores: a known restore time for a mailbox and a folder.
- A backup channel: a way to reach people that doesn't run through Microsoft 365.
- A written plan: who decides, who communicates, and what work continues.
- A dependency list: which business processes rely on which vendor.
Band of Coders helps mid-market teams put these pieces in place. We provide Keepit licenses for Microsoft 365 backup, set them up properly and support them over time. Keepit stores backups in its own vendor-independent cloud, separate from Microsoft, and covers Exchange Online, OneDrive, SharePoint, Teams and Groups. Our engineers can help you test restores, close backup gaps and write a practical outage plan. And if you want a wider look at vendor concentration risk, a fractional CTO can map your dependencies and help you decide what to change.
What to watch next
- Microsoft's post-incident report. It should explain the root cause and what Microsoft is changing. Read it with your outage plan open.
- Your service credit deadline. Microsoft 365 commits to 99.9% monthly uptime for business plans, and credits aren't applied automatically. You have to file a claim in the admin center, and reports put the deadline for this outage at December 31, 2026.
- AI tools built on Microsoft 365. As you add AI tools that connect to your Microsoft 365 data, stress-test what happens to them during an outage before you scale them.
- Your next renewal. If you renew Microsoft 365 or your backup tool in 2027, confirm your backup scope covers mailboxes, OneDrive, SharePoint and Teams.
Want to know where you stand?
We offer a 30-minute continuity check. We'll go through your Microsoft 365 backups, whether they caught up after the outage, and your plan B, and tell you where the gaps are.
Would your team know what to do if Microsoft 365 went down tomorrow? Book a continuity check.
Frequently asked questions (FAQs)
Frequently asked questions about Microsoft 365 outages
What caused the November 2026 Microsoft 365 outage?


How long did the outage last?


Did the outage cause data loss?


Do I need third-party backup for Microsoft 365?


Can my business get a credit from Microsoft for the outage?


Related posts

Microsoft 365 outage: what it showed about backup and business continuity

Ransomware recovery for SaaS data: why protection isn't complete until recovery is proven
.png)


