Updated, 16 November 2026, 18:36 GMT. Microsoft told business admins at 04:17 GMT on Saturday 14 November that the incident had ended at 16:10 GMT on 13 November, and gave a preliminary cause. We’ve added that, the start and end times Microsoft gave, and a section on preparing for the next outage.
Microsoft says the outage that disrupted Outlook and other Microsoft 365 email services for businesses ran from 16:20 GMT on Thursday 12 November to 16:10 GMT on Friday 13 November.
If you were expecting email, or sent invoices by email, on 12 or 13 November, check whether anything bounced.
In its final notice to business admins, Microsoft gave a preliminary cause: a latent bug in the systems that set up connections to Exchange Online, the service behind Microsoft 365 email. It says traffic shifts during standard network maintenance and upgrades exposed the bug, and that the incident was “a technical issue not related to cyber or AI / agents”.
That notice went out in the Microsoft 365 admin centre, which only business admins can see, and the public status page dropped the business incident without a final post. Cloud Admin Hub, which republishes Microsoft 365 service health posts, carries the notice in full.
The BBC, in a report updated on Saturday morning, said Microsoft had said its services were fully restored. Microsoft gives the same end time, 16:10 GMT on Friday, for personal and family accounts.
Should you switch to Gmail?
Alex De Sousa, Digital Royalty’s founder, says no: “Gmail occasionally has downtime too, swapping providers isn’t a fix, being prepared to work around outages is the fix.”
Google’s own Workspace status history shows Gmail problems this year. Its incident report for 8 April 2026 says Gmail users “may have experienced delays/failures when sending or receiving emails” for 8 hours 10 minutes, mostly on mail to outside domains.
In the early hours of 31 May, UK time, Gmail, Drive, Calendar, Meet and other Workspace services were disrupted for about 75 minutes, Google says. Neither was as long as this Microsoft outage, which lasted almost 24 hours for business customers by Microsoft’s own start and end times.
Still unknown
Microsoft hasn’t said publicly what the latent bug was or what it will change to stop it happening again. Its post-incident report may cover that.
It also hasn’t said how many organisations were affected, or whether any mail was lost.
Changes
Update, 16 November 2026, 18:36 GMT: Added Microsoft’s final notice to business admins (04:17 GMT, 14 November: incident over, start 16:20 GMT on 12 November and end 16:10 GMT on 13 November, preliminary root cause a latent bug) and its post-incident report notice, as republished by Cloud Admin Hub, plus the BBC’s report that Microsoft said services were fully restored. Changed the headline, rewrote the opening and the advice for the incident being over, and added a section on preparing for the next outage. This update is late: Microsoft’s final notice came early on 14 November and the BBC reported Microsoft’s all-clear that morning, but we were checking only Microsoft’s public status page. Also replaced BleepingComputer’s 18:26 GMT start with Microsoft’s 16:20 GMT, and added Microsoft’s 02:17 GMT admin-centre update, the BBC’s 17:54 GMT first-acknowledgement time and Matt Saunders’s comment.
Update, 14 November 2026, 05:49 GMT: Microsoft had removed the business incident from its public status page by 05:47 GMT, with no final public update in our checks; added that, pointed business admins to Service health, and changed the headline.
Update, 14 November 2026, 04:08 GMT: Added Microsoft’s 03:51 GMT consumer update (service restored for personal and family accounts, email backlog processed, final update, with start and end times), noted the business update due by 02:30 GMT still hadn’t appeared, and changed the headline.
Update, 14 November 2026, 02:39 GMT: Added Microsoft’s 01:51 GMT consumer update (sustained recovery; not considered remediated until the email backlog is processed; next update by 04:00 GMT), noted that its business update due by 02:30 GMT hadn’t appeared, and changed the headline and the outage’s running length.
Update, 13 November 2026, 22:03 GMT: Added Microsoft’s 22:00 GMT business update (sustained recovery, most affected features addressed, email backlog still being processed before it calls the incident resolved; next update by 02:30 GMT Saturday).
Update, 13 November 2026, 20:08 GMT: Added Microsoft’s 20:02 GMT business update and 20:03 GMT consumer update (still clearing queued mail and rebalancing traffic; no estimate for full recovery; next updates by 00:00 GMT on Saturday 14 November).
Update, 13 November 2026, 18:03 GMT: Added Microsoft’s 17:57 GMT business update, including its first account of what triggered the incident (an issue from an earlier network capacity upgrade that surfaced after a recent network change), that it has addressed the underlying issue and is still restoring service to remaining customers and regions.
Update, 13 November 2026, 16:37 GMT: Added Microsoft’s 16:24 GMT consumer update (signs of sustained recovery, still listed as a service degradation; root cause still under investigation).
Update, 13 November 2026, 15:58 GMT: Added Microsoft’s 15:56 GMT business update (addressing residual impact; next update by 18:00 GMT).
Update, 13 November 2026, 14:58 GMT: Added Microsoft’s 14:56 GMT business update (sustained improvements), its 14:10 GMT consumer update, and Fasthosts’ notice to its Microsoft 365 customers.
Update, 13 November 2026, 14:02 GMT: Added Microsoft’s 12:58 and 14:00 GMT business updates (email queues approaching normal levels) and IONOS’s notice to its Microsoft 365 customers.
Update, 13 November 2026, 12:14 GMT: Added Microsoft’s 12:00 GMT business update (recovery improving in Europe and North America) and its 11:55 GMT consumer update.
Update, 13 November 2026, 11:45 GMT: Added Xero’s warning that emails sent from Xero to Microsoft inboxes may bounce.
Update, 13 November 2026, 11:29 GMT: Added Microsoft’s 11:04 GMT update, the detail of which business services are and aren’t affected, and a section on whether to switch to Gmail, with Alex De Sousa’s view.
Correction, 13 November 2026, 11:29 GMT: The first version of this article said:
- that Teams was down. That held for personal accounts, where Microsoft listed Teams as affected until it declared the consumer incident over on 14 November, but for business accounts Microsoft says Teams calls, meetings and chats are mostly unaffected and its calendar and scheduling features are not.
- that the outage started just before 1pm US Eastern time. That was CNBC’s time for when connection problems began; Microsoft dates the incident itself from 13:26 Eastern (18:26 GMT) (since superseded; see the 16 November update).
- that Microsoft was reverting a network configuration change that lined up with the start of the problems. We could not find that in the reports we cited. BleepingComputer reported that Microsoft was checking whether reverting some recent configuration updates could help recovery, and Microsoft hasn’t named a cause (since superseded; see the 16 November update).
- that SharePoint files might not open in the browser. Microsoft says opening files stored in SharePoint still works.
What Microsoft said was and wasn’t working
Microsoft’s business updates from 11:04 GMT on Friday onwards listed the same affected services:
- Users might have been unable to open their mailbox in Outlook on the web, Outlook desktop or on phones, and email delivery might have been delayed.
- In Teams, most of the core features, including calls, meetings, group chats and channel messages, were not affected. Calendar, scheduling, meeting invites and edits were.
- Opening files stored in SharePoint still worked. Sharing files with people outside your organisation, and SharePoint features that depend on email, might fail.
- Admins might have been unable to load some admin centre pages.
- Copilot, Planner, To Do, Intune, Purview, Power Apps sign-in, Universal Print and some Windows 365 setups were also listed.
For personal and family accounts, Microsoft’s consumer status said at 03:51 GMT on Saturday that services had “fully recovered” and the backlog of email messages had been processed, “thus this issue is remediated”. It listed the incident as service restored and said that was its final update.
It gives the start of the consumer incident as 10:05 GMT on Thursday 12 November and the end as 16:10 GMT on Friday 13 November. Its 16:24 GMT update had named Outlook.com, OneDrive, OneNote, Microsoft To Do, Teams and Microsoft 365 for the web as affected.
What Microsoft says went wrong
In the admin-centre notice, Microsoft says the latent bug caused “a rapidly increasing number of requests to queue for processing”. As the queue grew, it says, the affected servers “exhausted their available processing capacity and could no longer accept new connections”, and people couldn’t reach the services.
Microsoft called this a preliminary root cause and said its analysis was still going on. It promised a preliminary post-incident report within two business days and a final one within five. Cloud Admin Hub shows a notice at 04:46 GMT on Monday 16 November that a post-incident report has been published. We haven’t seen it.
What Microsoft said during the outage
In its last public business update, at 22:00 GMT on Friday, Microsoft said it continued “to observe sustained service recovery” and believed “the majority of those impacted scenarios have been addressed”. It still listed the incident as a service degradation.
It said it would keep monitoring for an extended period because the incident was so severe, and wouldn’t consider it resolved until the backlog of email built up during the outage had been processed.
It promised its next business update by 02:30 GMT on Saturday. None appeared on the public status page in our checks, though Cloud Admin Hub shows a 02:17 GMT update in the admin centre saying recovery was continuing. By 05:47 GMT the business incident was gone from the public page, which went back to Microsoft’s standard message saying the page is updated when problems stop admins reaching Service health in the Microsoft 365 admin centre.
At 20:02 GMT it had said it was “still draining queued messages and rebalancing traffic”, that sign-in and access might improve before mail delivery caught up, and: “We do not have an ETA for full recovery yet.”
At 17:57 GMT it had said it was seeing “broad recovery and stability” and had “addressed the underlying issue”, and was still restoring service to the remaining customers and regions. It apologised for the disruption and said it had restored services gradually, checking stability at each stage, which took longer. Earlier in the afternoon it had said email queues were “significantly reduced and approaching normal levels”.
Microsoft’s final notice gives the start as 16:20 GMT on Thursday 12 November. During the outage, BleepingComputer reported that Microsoft dated it from 18:26 GMT. Microsoft first acknowledged mailbox access problems in the Americas at 17:54 GMT on Thursday, the BBC says.
The problem was in Exchange Online. Parts of other Microsoft 365 services that talk to Exchange Online and email were affected, and the parts that don’t depend on it kept working, Microsoft said.
What Microsoft said about the cause during the outage
At 17:57 GMT on Friday Microsoft said for the first time what had triggered the incident. It said the incident “was introduced by an issue during an earlier network capacity upgrade that surfaced following a recent network change”, and that it was “a technical issue not related to cyber or AI / agents”. Its final notice on Saturday added the latent bug described above.
What it said before that:
- On Thursday it put the problem down to elevated load across several regions, caused by servers becoming “unexpectedly unavailable” and straining the rest, as reported by CNBC and BleepingComputer.
- It later said “a process on the affected servers responsible for establishing secure connections becomes overwhelmed”, ABC7 reported. BleepingComputer reported that it was checking whether reverting a small number of recent configuration updates could speed up recovery.
- Overnight, ABC7 reported, it first saw “positive results” from restarting affected infrastructure, then said partial recovery had not been sustained.
- On its status page on Friday it said it had “no evidence of malicious activity”.
- At 16:24 GMT on Friday its consumer status said its investigation into the underlying root cause was still going on.
What to check now
Microsoft extended its email timeout window to 48 hours to protect delayed messages while delivery caught up. Mail that still hasn’t arrived may have bounced back to whoever sent it, so if you’re expecting something from 12 or 13 November, ask the sender whether it bounced.
If you email invoices or quotes from Xero, check for bounces from those days. Xero said emails sent from Xero to Microsoft inboxes such as Outlook or Hotmail might bounce because of the outage.
On a business plan, your admin can read Microsoft’s record of the incident under Service health in the Microsoft 365 admin centre, as MO1550821. If you buy Microsoft 365 through IONOS or Fasthosts, both of which told customers during the outage that the fix had to come from Microsoft, ask them to send you Microsoft’s incident record.
Treat an email telling you to sign in again or “verify” your Microsoft account because of the outage as suspect.
Preparing for the next outage
None of this gets you into mail already stored in Outlook while it’s down. Matt Saunders of the consultancy Adaptavist told the BBC that “organisations must have a plan in place for how to work without such tools”.
How the outage unfolded
- The business incident starts, according to Microsoft’s final notice.
- Microsoft first acknowledges mailbox access problems in the Americas, the BBC says.
- The incident ends, by Microsoft’s later account, for business and personal accounts alike.
- Microsoft says for the first time what triggered the incident, and that it has addressed the underlying issue.
- Microsoft tells business admins the incident is over and names a latent bug as the preliminary cause.
- Cloud Admin Hub shows a notice that Microsoft’s post-incident report has been published.
Alex De Sousa is the founder of Digital Royalty, a software development company in Portsmouth that builds websites, web apps and AI tools for UK businesses.