Microsoft has fully restored Outlook, Teams and other Microsoft 365 services after a major outage that began on 12 November and lasted for more than a day.
Scale and nature of the disruptions
Monitoring service Downdetector recorded more than 7.4 million reports of disruptions, and problems were observed in 74 countries and territories. The number of complaints reached almost 20,000 per minute - more than 1,500 times the usual level.
Users reported that they could not open Outlook mailboxes, send or receive messages, use calendars or hold online meetings in Teams; there were also disruptions in Microsoft 365, Copilot, SharePoint and other corporate tools.

Cause of the outage and recovery process
The incident was linked to technical problems in the server infrastructure, notably Exchange Online. The outage occurred after a prior upgrade to network capacity and a subsequent change in network configuration, which caused some servers to become unreachable during automated checks.
Deployment of updates was temporarily halted, some traffic was redirected away from the affected infrastructure and servers were restarted. Recovery proceeded gradually because servers had to be returned to service and accumulated email queues processed; cyberattacks and artificial intelligence were ruled out as causes of the incident.
Impact on businesses and public institutions
The disruptions affected workflows in public institutions, healthcare, the financial sector, aviation, logistics and corporate reporting. In the United Kingdom, police forces experienced problems with receiving online crime reports, and parts of the National Health Service (NHS) reverted to paper records and faced delays with electronic prescriptions to pharmacies.
Financial firms reported disruptions to asset management and transactions, and some organisations used backup communication channels. Airlines and logistics companies recorded issues with customer email support and updates, and some companies postponed publication of financial results or reports.

Estimates of financial losses
The total economic damage has not been established. Mario De Chicco, an analyst at Morningstar DBRS, estimated possible insured losses at $300−500 million, noting this is a projection of insurance payouts specifically, not a full estimate of all economic losses.
For context, the 2024 global CrowdStrike outage was estimated at $400 million to $1.5 billion in insured losses. The company also notes that more than 430 million people use Microsoft 365 services and the number of paid commercial subscriptions exceeds 450 million; as of 14 November the services had been fully restored in all regions.
Issues with Claude and Anthropic
Artificial intelligence tools connected to corporate services were also affected. Anthropic reported outages in the Microsoft 365 Connector and Claude for Outlook, which could have impacted mail, calendar and search functions.
Users on 13 November recorded authentication errors and degraded performance in Claude.ai, Claude Code and other Anthropic products. Anthropic’s official status page indicated that authentication and general service issues were resolved on 13 November, and core Claude functions were operating as of 14 November, although there were ongoing reports of an elevated number of errors during payments and plan changes.





