The Microsoft 365 Outage: Write Down What Stopped
For more than a day last week, a large part of the working world could not send an email. Microsoft says the cause was a network change, not an attack. For a small business the useful question is narrower: what, exactly, stopped at yours?
What happened
Reports of problems with Microsoft 365 began climbing at about 17:25 GMT on Thursday 12 November, according to Ookla figures cited by Reuters. The disruption ran for more than 24 hours. On Saturday 14 November Microsoft said it had addressed the underlying issue and restored service to the remaining customers and regions.
Microsoft said the incident hit features tied to Exchange Online, its cloud email and scheduling service, and listed the services that could be affected.
| Service named by Microsoft | What it does for a small business |
|---|---|
| Exchange Online | Email and calendars |
| Microsoft Teams | Chat, calls and meetings |
| SharePoint Online | Shared files and intranet pages |
| Microsoft 365 Copilot | The AI assistant across those apps |
| Planner and To Do | Task lists |
| Intune | Managing laptops and phones |
| Defender XDR and Defender for Office 365 | Security monitoring and email protection |
Downdetector logged reports from 74 countries and territories within an hour, the Associated Press reported. Government agencies, courts, airlines and trading firms were among those affected. Reuters noted that organizations running Microsoft services on their own servers were somewhat spared, and called it one of the longest outages in the company's history.
What Microsoft says caused it
On Friday Microsoft apologised and described the cause in a statement: a technical issue, introduced during an earlier network capacity upgrade, that surfaced following a recent network change. Earlier it had said that a process on the affected servers responsible for establishing secure connections became overwhelmed, causing the servers to stop processing requests.
It said the incident was not related to a cyberattack or to AI agents, and that it restored services gradually to keep the recovery stable. A full post-incident report is promised once its review is complete, so treat the explanation so far as preliminary.
Why the AI assistant went down too
Copilot is on the list, which matters if you have been adding AI to your daily work. An assistant that reads your email, calendar and files lives on the same platform as the email, calendar and files. When the platform stops, the assistant stops with it.
That does not make it a bad tool. It means AI built into a suite does not give you a second way of working. Your team lost the inbox and the thing that summarises the inbox in the same minute, which is the dependency question in what stops when your AI provider goes down.
The security tools are the less obvious casualty. If device management and email protection are named as affected, it is fair to ask your IT provider what you could and could not see during those hours.
Write down what stopped, this week
The most useful thing you can do costs nothing and has a short shelf life. While people still remember, ask each team what they could not do on Thursday and Friday. In a month the answer will be nothing much, and the list will be lost.
Expect the list to be longer than email. Things that quietly ride on the inbox include invoices going out, approvals that arrive as messages, alerts from other systems, password resets and sign-in codes sent by email, and customer enquiries from your website form.
For each item, ask one question: could this wait a day? Where the answer is no, it needs a fallback that does not sit on the same platform.
| If this cannot wait | A fallback that is independent |
|---|---|
| Reaching your own staff | A phone list or a group chat on a different service |
| Reaching key customers | Their phone numbers, saved somewhere outside the mail system |
| Taking orders or enquiries | A web form that stores entries as well as emailing them |
| Signing in to other systems | Sign-in codes from an app, not from email |
| Telling customers you are affected | A status note on your website or social accounts |
Keep the plan itself somewhere you can reach when the suite is down. A continuity document stored only in SharePoint may not have been readable on Thursday.
What to check now that it is back
Look for mail that arrived late or twice. Microsoft said it was working through a backlog of messages built up during the incident. Orders, invoices and replies may have landed out of sequence, and a customer who got no answer on Thursday may still be waiting.
Ask your IT provider to confirm that security monitoring and device management are reporting normally again. Then read the post-incident report when it is published, and check what your agreement says about service credits for downtime, since terms vary by plan.
Finally, put the dependency where it will be seen. One platform carrying email, files, meetings, security and AI is a reasonable choice for a small business and it is a concentrated risk. It belongs on your risk register with an owner, and in your continuity plan with the fallbacks you just wrote down.
Frequently Asked Questions
What happened in the Microsoft 365 outage?
Starting on Thursday 12 November 2026, a global disruption affected Microsoft 365 services for more than 24 hours. Microsoft said it hit features tied to Exchange Online, and listed Teams, SharePoint Online, Microsoft 365 Copilot, Planner, To Do, Intune and Defender services as potentially affected. Service was reported fully restored on Saturday 14 November.
What caused it?
Microsoft said it was a technical issue introduced during an earlier network capacity upgrade that surfaced after a recent network change. It also described a process responsible for establishing secure connections becoming overwhelmed on affected servers. A full post-incident report has been promised, so this explanation is preliminary.
Was it a cyberattack or caused by AI?
Microsoft says no to both. Its statement described the incident as a technical issue not related to cyber or AI agents, and said it was not aware of any malicious activity.
Why did Copilot stop working?
Because it runs on the same platform as the email, calendars and files it works with. An AI assistant built into a suite depends on that suite being available, so it does not provide an alternative way of working when the suite is down.
What should a small business do after the outage?
Ask each team what they could not do while it lasted and write the answers down. For anything that could not wait a day, set up a fallback on a different service, such as a phone list, sign-in codes from an app, or a web form that stores entries. Then check for late or duplicated email and confirm security tools are reporting normally.
Are we owed compensation for the downtime?
It depends on your plan and agreement. Service credit terms vary, so check what your own agreement says or ask the partner you bought through. Read the post-incident report when Microsoft publishes it.
Find the single points of failure
We map which of your processes depend on one platform and set up plain fallbacks for the ones that cannot wait a day.
Related Articles
Legitimate Interest: The Consent Exception in Bill C-36
Records Management: What AI Does to Everything You Kept
Suno Lawsuits: Guardrails Are Not Your Legal Defence
Ajan leads the ChatGPT.ca team: 200+ custom GPT builds and automation projects for 50+ businesses across 20+ industries. Based in Markham, Ontario. PIPEDA-compliant solutions.