Telecom BusinessMicrosoft 365Service OutageCloud ResilienceEnterprise Communications

Microsoft 365 outage drew 7.7 million reports across 90+ countries

Microsoft 365's November outage drew 7.7 million reports, exposing how a shared gateway defect can halt email, meetings and public services without breaking broadband or mobile access.

4 min read

Luke Kehoe: Microsoft Outage Exposes Concentration Risks in Digital InfrastructureTelecom Business
Luke Kehoe: Microsoft Outage Exposes Concentration Risks in Digital InfrastructureAI-generated

Why it matters

  • More than 7.7 million reports were logged in 60 hours, including 3.9 million from the United States.
  • U.S. reports peaked at about 14,000 per minute at 1:25 p.m. ET.
  • Microsoft cited a July software defect while nearly 14% of gateway capacity was out of service; more than 80% of those servers later stopped responding.
  • More than 90 countries and territories logged at least 1,000 outage reports.
  • Microsoft 365 has more than 450 million paid commercial seats and is designated a critical technology provider in European financial sectors.

The story

By Elena Vasquez

Microsoft 365's November 12-13 outage generated more than 7.7 million Downdetector reports about Outlook, Teams and other Microsoft services in its first 60 hours. About 3.9 million, or half, came from the United States.

Downdetector, an Ookla platform that collects user reports about service issues worldwide, recorded its largest event since the AWS outage of October 2025. U.S. reports began at about 12:30 p.m. ET on Thursday and peaked near 14,000 per minute at 1:25 p.m. ET.

Counting every service Downdetector tracks, Thursday was the busiest day on its U.S. site in the past year, ahead of January 14, when Verizon suffered a network outage. The U.S. House of Representatives said its users were affected, as did city governments including Austin's.

The disruption reached public institutions as well as businesses. The House and Austin's city government reported effects on their users. Microsoft put customer impact at almost 24 hours, long enough for people in each region to encounter the failure as they began work.

What caused the Microsoft 365 failure?

Microsoft's preliminary review, dated November 15, traced the outage to a "software defect introduced in July" in networking gateway servers that route traffic into Microsoft 365. The defect surfaced when Teams traffic moved onto a gateway shared with Exchange Online, Microsoft's email service, while nearly 14% of that gateway's capacity had already been taken out of service.

At the worst point, more than 80% of those servers stopped responding. Microsoft found no sign of a cyberattack, while its final review is due November 20.

How did the outage move across the working day?

The outage followed the working day around the world. Downdetector recorded its regional peaks at:

  • Around 9 a.m. local time on Friday in Japan and the UK.
  • 10:15 a.m. in India.
  • Between 8:30 a.m. and 9:15 a.m. across most of Europe.

Worldwide, reports stayed at one-fifth or more of their busiest hour for about 20 hours, roughly twice as long as any other large outage in the previous year. Steam's Christmas Eve 2025 failure came closest, at about nine hours. More than 90 countries and territories logged at least 1,000 outage reports.

Institutional effects extended beyond lost email and delayed meetings. The consequences included:

  • Some UK hospitals rescheduling procedures.
  • Japan's Digital Agency being unable to receive email.
  • Immigration New Zealand pausing visa processing.

Did broadband and mobile access networks fail?

Those gateways sit inside Microsoft's own platform, beyond the point where broadband and mobile networks hand traffic over. The wider internet held up: Ookla's round-trip latency measurements to major content delivery networks and Speedtest servers did not move through the worst of the outage.

That separation points to a failure inside the Microsoft 365 platform rather than a broad failure of broadband or mobile access networks. The knock-on effect appeared elsewhere, as Gmail outage reports rose more than 40 times above normal, probably because bounced emails led senders to assume their own systems were at fault.

What does the concentration risk mean?

Microsoft 365 has more than 450 million paid commercial seats, and many organizations use it for email, meetings and documents. That concentration turns a gateway failure into a simultaneous interruption for many workplaces, rather than separate incidents they can absorb.

EU financial regulators and the UK Treasury have named Microsoft a "critical technology provider" to their financial sectors. The same dependency reaches beyond banks into legislatures, hospitals and city halls, making the outage a test of concentration risk as well as software resilience.

The analysis comes from Luke Kehoe, who leads Ookla's research and thought leadership in Europe and has collaborated with mobile operators, telecom vendors and government agencies. Kehoe, an electronic engineering alumnus of University College Dublin, has contributed to publications on 5G, IoT, open RAN and edge computing, including work with the Telecom Infra Project and the World Economic Forum.

Kehoe's central point is that shared dependencies concentrate risk: when many organizations rely on the same platform, one fault can stop work across all of them at once. Microsoft's final review, due November 20, is the next opportunity to verify its preliminary account of the defect, capacity loss and shared gateway.

Also reported

Original: ookla.com

Share this article:

« PreviousNext »

More from Elena Vasquez

Elena Vasquez

Show full bio

Market editor covering companies, deals and strategy in Carrier business, 5G and 6G, fiber and broadband at Telecom Gazette.

568 articles