For almost 24 hours, Outlook, Teams, and other Microsoft 365 services were acting up in 74 countries. Hospitals, government offices, and companies had to improvise.
This Thursday, November 12, companies unable to communicate with their customers, embassies cut off from their email, and health services forced to go back to paper records were hit by a worldwide Microsoft outage. Originating from the American group’s infrastructure, the failure affected 74 countries and territories. Nearly 24 hours after the first disruptions, services still weren’t fully restored. The incident highlights just how much both public and private organizations rely daily on digital tools from a limited number of providers.Several Microsoft 365 apps were affected. The suite has more than 430 million users, according to the data shared. Outlook, used for emails and calendars, Teams, which allows for meetings and internal messaging, SharePoint and OneDrive, meant for document storage and sharing, were disrupted, as well as Copilot, Microsoft’s AI assistant.
Microsoft recorded the incident on Thursday, November 12 at 1:26 PM (Haiti time). The issues quickly spread to other regions. In the United States, Downdetector, a site that tracks outages reported by users, counted up to 138,058 reports about Outlook. This number reflects the reports received by the platform, not the actual number of people affected.Economic activities and public services blocked
In companies, the consequences were immediate. Without access to their email, employees could no longer receive orders, respond to clients, or send documents. Files stored online became hard to reach, and without Teams, some teams had to give up their usual meetings.
Restoring the connections didn’t fix everything. Emails delayed during the outage arrived all at once, forcing employees to deal with them on top of the day’s messages. For an organization that relies on Outlook, just a few hours of interruption are enough to postpone appointments, contract signings, or business operations.The outage also affected companies that don’t directly use Microsoft 365 but rely on the group’s infrastructure to run some of their services. OpenAI and Anthropic reported disruptions related to their connections with Microsoft. There were also issues reported with Zoom, Calendly, and Notion. When a central provider runs into a failure, it’s not just its direct clients who feel the impact.
The disruptions also affected public services. In New Zealand, authorities said that the email systems of their embassies, consulates, and foreign representations were impacted. Immigration services had trouble processing visa applications. In Japan, the Ministry of Foreign Affairs reported email problems and asked people with urgent matters to prioritize phone calls.In the healthcare sector, the situation was even trickier. In the UK, several NHS departments, the public health system, could no longer properly transmit information about patients. Operators for the 111 medical advice line and 999 emergency line couldn’t exchange some data as usual. Pending files piled up, raising concerns about delays in patient follow-ups. However, life-saving emergency interventions seemed to be unaffected.
Some establishments had to go back to handling paper files and favor phone calls. In Lincolnshire, psychiatric services asked doctors to call instead of emailing for urgent cases. In France, a medical analysis lab couldn’t access or print certain documents anymore, forcing some patients to leave. A higher education institution also ended up without its schedules and room plans.Transportation has also been affected. In India, airlines warned their customers that some requests might not reach them. In the aviation and logistics sectors, disruptions were reported in coordinating operations and accessing schedules. The stock exchange operator Euronext asked listed companies to favor phone communications for some matters, while noting that its own operations were continuing as usual.
Behind software that won’t open, sometimes an entire chain of activity comes to a halt: an order doesn’t arrive, a delivery is delayed, or a patient’s file gets stuck between two departments.Unstable servers behind the outage
The outage comes from a failure in Microsoft’s infrastructure. The applications companies use run on remote servers grouped in data centers. This is what’s called the cloud, or cloud computing. Users access their software and documents over the Internet, without everything necessarily being stored on their computers.
These services rely on each other. Outlook runs on Exchange Online, the infrastructure that handles business email. Teams also uses some of its services for calendars and meeting organization. When a central component becomes unavailable, several apps can simultaneously lose some features.According to Microsoft, some servers became unstable, forcing the still-operational equipment to take on extra load. A process responsible for managing secure connections also got overwhelmed. The technical teams restarted servers, changed some configurations, and rerouted traffic to gradually restore services.
The company ruled out a cyberattack and assured that artificial intelligence was not behind the incident. On Friday, November 13, it apologized to its customers and partners while announcing a significant recovery of services. Connections gradually stabilized, even though some users continued to experience occasional interruptions.The outage also created conditions favorable for scams. Taking advantage of the confusion, cybercriminals can pretend to be Microsoft technicians and offer to unlock an account in order to get a password or other confidential information. Cybersecurity organizations have warned users about these practices. Even during an outage, you should never give your login details to someone who spontaneously contacts you claiming to work for Microsoft.Risks of digital dependence
The incident reignites the debate about dependence on major tech platforms. Microsoft isn’t the only player in this market. Amazon Web Services and Google Cloud also host a significant portion of companies’ applications and data. According to Synergy Research Group, the three companies together held 61% of the global cloud infrastructure services market in the third quarter of 2026.
A large part of economic activity therefore depends on a limited number of providers. When one company experiences a failure, the disruptions can go beyond its direct customers and affect businesses that indirectly use its infrastructure.Speaking this Friday on the public service radio France Inter, lawyer Bastien Le Querrec added a nuance to the debate on digital sovereignty. For him, the issue is less about the nationality of the provider and more about the concentration of services in the hands of a few big companies. So, replacing one tech giant with another wouldn’t necessarily eliminate this vulnerability.
For Haitian businesses, banks, government offices, media outlets, and schools, this outage feels like a warning. There’s no precise data to determine how many organizations in the country were directly affected.The risk still remains real, though. A company that receives all its orders through Outlook could suddenly find itself cut off from its clients overnight. A media outlet that keeps all its working documents solely online could temporarily lose access to its archives. An administration that has moved most of its activities to the cloud must be able to keep running when its provider has an outage.
Preparing for this kind of incident involves concrete steps: having backups that are independent of the platforms used daily, planning alternative ways to communicate when email becomes inaccessible, and setting up offline procedures for priority tasks. Using multiple providers can also help reduce certain dependencies, but this comes with extra costs and makes technical management more complicated.