The Financial Times reported on 13 November 2026 that Microsoft was working to restore Microsoft 365 services after a global outage affected Outlook, Teams, Copilot and other tools. Microsoft said it had no indication that the disruption was caused by malicious or cyber activity. The incident disrupted email, meetings, file access and services connected to Microsoft 365.
Downdetector, which tracks user reports, recorded a sharp rise in Microsoft 365 problems on Thursday, with reports peaking at nearly 140,000 around 18:00 GMT. Other reporting based on the same monitoring data put the disruption across 74 countries and territories. Microsoft 365 apps have more than 430 million users, according to Microsoft’s website.
Exchange Online was most affected
Microsoft said a number of servers became unexpectedly unavailable, which placed additional load on remaining systems. The company described the condition as elevated load across infrastructure in multiple regions. It also warned that restored services could become unresponsive again once traffic returned.
By early Friday, Microsoft said Exchange services were the most affected part of the platform. That meant Outlook mailbox access was a main problem, with effects on other products that depend on Exchange. Teams chats and online meetings were described as largely functional, while scheduling and creating chats could still be affected by Exchange issues.
The affected services included Outlook, Teams, SharePoint Online, OneDrive, Microsoft Copilot 365, Microsoft Planner and Exchange Online. Users also reported problems with files and calendar details. Copilot was affected where it depended on Microsoft 365 services.
Recovery work did not hold at first
Microsoft attempted several recovery actions during the incident. These included rolling back recent changes, limiting traffic to affected servers, shifting traffic to alternate routes, restarting servers and validating them before returning them to service.
The company later said some changes were not “taking as expected”. It said it was working on alternative workstreams to increase capacity in response to sustained traffic management imbalances. Microsoft also said it was scaling up virtual machines and querying machines in a bad state so recent network changes could sync.
Microsoft later narrowed the technical explanation. It said the process on affected servers responsible for establishing secure connections was becoming overwhelmed, causing those servers to stop processing requests. Microsoft had not published a full root-cause review while recovery work was still under way.
Connected services also felt the outage
The outage did not stop at Microsoft’s own applications. OpenAI said it was investigating an issue affecting its connections to Microsoft services, with some tools not working as expected. Anthropic said it had identified an issue with an upstream provider affecting its Microsoft 365 connector.
Public sector systems were also affected in some countries. Singapore authorities said some government systems were affected, while emergency services remained fully operational. Japan’s government said agencies using Outlook had alternative communication methods, including telephone, chat tools and text messages.
In Australia, some hospitals and health clinics reported clerical issues, but medical representatives said patient care was not affected. Some facilities returned to paper for prescriptions where authentication depended on access to Microsoft-supported email. Service NSW asked customers to delay visits to service centres where possible and use online transactions.
Continuity plans matter when SaaS fails
The incident shows the operational risk created when email, identity, meetings, document access and workflow tools sit behind the same provider path. For many organisations, Microsoft 365 is not only productivity software but also part of authentication, scheduling, case handling and external communication. When Exchange Online is impaired, the effect can reach systems that do not look like email at first glance.
Banks, universities and public bodies warned users about opportunistic scams during the outage. Their advice was to be wary of unexpected messages, links, app downloads and calls offering IT support. New Zealand’s Ministry of Business, Innovation and Employment also warned staff not to move confidential or customer details into personal email or unapproved tools just to keep work moving.
For administrators, the practical issue is not whether a cloud platform can fail. It is whether the organisation has tested ways to communicate, authenticate, approve work and serve customers while the main collaboration platform is unavailable. Alternative channels are useful only if staff know when to use them and what data may pass through them.
When Outlook, Teams, Exchange Online and Copilot fail together, the M365 + Copilot topic at ECS deals with the same tenant-level questions around service dependency, governance and adoption. Sessions and practitioner discussions give attendees room to compare rollout decisions, fallback communication patterns and the operational limits of Microsoft 365 services. Bring the outage scenarios that are hard to settle from a service health page and use the conference to test them against people who run these platforms in production.
This article was written by the ECS editorial team with the assistance of Amal Agent, run.events' AI assistant, which supported the research, drafting and proofreading.