The Microsoft 365 outage that cut many businesses off from Outlook, some Teams features and the Copilot assistant from Thursday 12 November 2026 was easing on Friday 13 November. Microsoft points to overloaded servers and rules out a cyberattack. For a Swiss SME, the episode raises a practical question: what do you do when your only cloud provider goes down?
Which services were affected?
The heart of the outage lies in Exchange Online, Microsoft's email and calendar service on which business Outlook mailboxes rely. According to Microsoft's status page, quoted by ICTjournal, other services that depend on Exchange Online were dragged down with it.
- Outlook: mailboxes inaccessible from the web, the desktop application or mobile, and emails delivered late.
- Microsoft Teams: chats and online meetings worked, but not calendar integration or appointment scheduling.
- Microsoft 365 Copilot: the artificial intelligence (AI) assistant sometimes stopped responding or could no longer access emails, calendars and files.
- Administration and security: SharePoint, the admin portal and Microsoft Defender XDR were disrupted. Link checking by Safe Links in Defender for Office 365 could be unavailable, meaning some links scanned by this service might not open.
Microsoft also listed Planner, To Do and Intune among the potentially affected services, according to the American channel ABC7. The Microsoft 365 connectors used by OpenAI and Anthropic, two AI developers, were also disrupted, report Inside IT and ICTjournal.
The outage was felt in Switzerland. Several hundred reports were logged on the allestörungen.ch platform, according to SRF, the German-language Swiss public broadcaster, quoted by ICTjournal. Elsewhere, New Zealand reported that email reception at its embassies and consulates was affected, according to the AWP news agency, as relayed by Allnews.
Timeline of the Microsoft 365 outage
- Thursday 12 November, around 15:45 (Paris time, identical to Swiss time): Teams falters first, according to Journal du Coin. Exchange Online follows at 18:54, initially in the Americas. Timings differ: Al Jazeera places the start of the Outlook problems at around 16:00 GMT, i.e. 17:00 in Switzerland.
- Within an hour: Downdetector, an outage-tracking site, records reports in 74 countries and territories. Around 19:00, it counts nearly 130,000 reports for Outlook, according to Journal du Coin.
- Thursday evening: shortly after 23:00 GMT, Microsoft observes a "limited recovery" by redirecting traffic away from the affected servers, according to Allnews. It later acknowledges that its changes "are not having the expected effect".
- Night of 12 to 13 November: at 2:12, Microsoft begins rolling back a network configuration change that coincides with the start of the incident, reports Journal du Coin.
- Friday 13 November, morning: around 7:00 GMT, Downdetector still counted at least 43,000 reports for Outlook in 72 countries, according to Al Jazeera. Microsoft reports initial progress.
- Friday, 9:17 New York time: Microsoft announces email queues "significantly reduced and close to normal levels", according to ABC7.
What cause does Microsoft give?
Microsoft's statements. The company first explained on the social network X that part of its infrastructure was under "high load" in several regions, probably causing the connection errors. It then specified, according to Al Jazeera, that servers had unexpectedly become unavailable and that this load did not stem from an increase in traffic.
On Thursday evening, Microsoft described the mechanism, quoted by ABC7: on the affected servers, a process responsible for establishing secure connections is overwhelmed, and the servers stop processing requests. According to Journal du Coin, the rest of the fleet absorbed the traffic from the failed machines before giving way in turn, and restarted servers went down again as soon as the first requests arrived.
No attack, according to Microsoft. "There is no indication that this is the result of malicious or cyber activity," Charles Lamanna, executive vice-president at Microsoft, wrote on X, quoted by Al Jazeera. Microsoft has not published a root cause and has not said whether the rolled-back network change was to blame. It promises a detailed report once the incident is resolved.
Our analysis. Nothing in the published information justifies treating this outage as a security incident. Rather, it is a reminder that a technical failure at a single provider is enough to cut off the email of thousands of organisations at the same time.
Where does the recovery stand?
Microsoft proceeded in stages. "Thanks to our latest traffic optimisation measures, we are seeing positive results for service stability and increased capacity," it stated on X, quoted by Inside IT. These measures were to be extended to the other affected regions.
According to ICTjournal, a fix to a component responsible for authenticating and routing requests stabilised some of the servers on Friday morning. Microsoft was then rolling it out to other parts of its infrastructure. In its update on Friday at 9:17, New York time, the company was still warning that some users could experience intermittent access problems, according to ABC7. It had given no date for full recovery.
Does a Swiss SME depend too much on a single cloud provider?
The question concerns almost the entire economy: SMEs account for 99.7% of the country's companies, notes the Confederation's SME portal (kmu.admin.ch). This portal notes that the web giants offer turnkey solutions that are often cheaper than local partners. The latter, on the other hand, host critical data in Switzerland and are often more responsive when needed.
The Federal Office for Cybersecurity (FOCS) lists clear responsibilities for the cloud and outsourced IT among its organisational measures, as we detailed in our article on the ICT minimum standard, the Confederation's reference framework. It is precisely on organisation that SMEs lag behind. According to the 2026 SME Cybersecurity study, emergency plans and supplier checks remain far less widespread than technical measures.
The issue goes beyond Microsoft. On 20 October 2026, an outage in Google Cloud's London region reportedly blocked card payments and the app of the neobank Revolut, according to a source quoted by the Reuters news agency; Revolut had not confirmed this link. According to Ashwin Pal, partner at the firm RSM Australia, quoted by Journal du Coin, nearly 4 million businesses depend on Microsoft software, and each hour of downtime would cost professional services firms more than 100,000 dollars.
What a business continuity plan provides for an email outage
A business continuity plan is not only for cyberattacks. The same sections as a cyberattack response plan apply to an outage like this one: who decides, who to notify, what to tell customers. "What is the minimum service to maintain in the event of a global outage?" sums up Walter Pasquarelli, an associate researcher at the University of Cambridge, quoted by Al Jazeera.
| Need | What the outage showed | What the plan can set out |
|---|---|---|
| Backup channel | Teams chats and calls largely kept working | The internal channel to use if email goes down, and the phone numbers of key people |
| Fallback email | Desktop Outlook could still display emails already downloaded, according to Al Jazeera | A contact address outside the main infrastructure, known to customers and partners |
| Informing customers | Emails arrived late, with no announced recovery time | A template message stating what still works and how to reach the company |
| Backups | Mailboxes are among the critical data, according to kmu.admin.ch | Online and offline copies, and regular testing of their restoration |
| Inventory | Copilot, SharePoint and link filtering were disrupted along with Exchange Online | The list of services that rely on the same provider |
The last two rows reflect published recommendations. The SME portal advises combining online and offline backups. The 2026 study finds that only one SME in four regularly tests restoring its backups. The first three rows are our own analysis, to be adapted to the size of each company.
One practical point concerns security codes received by email. Journal du Coin notes that they could arrive hours late and advises against repeatedly requesting them to be resent, which risks triggering the platforms' anti-fraud protections. For the rest, the basic IT security measures and an SME cybersecurity test provide an initial assessment. The SME portal also cites the Cyber Safe label for identifying recognised providers.
What we don't know yet
- The root cause: Microsoft has not said whether the rolled-back network change caused the outage. Its detailed report is expected after resolution.
- The number of users affected, which Microsoft has not confirmed, nor the extent of the disruption in Swiss public administrations.
Sources
- ICTjournal: "Microsoft confronté à des pannes d'Outlook et de Teams" (13 November 2026)
- Inside IT: "Microsoft kämpft gegen weltweite Störungen" (13 November 2026)
- Allnews (AWP): "Microsoft peine à réparer la panne mondiale d'Outlook" (13 November 2026)
- Journal du Coin: "Panne Microsoft 365 : Outlook à l'arrêt depuis jeudi soir" (13 November 2026)
- Al Jazeera: "Microsoft hit by global outages: What happened and why?" (13 November 2026)
- ABC7: live coverage of the Microsoft outage (13 November 2026)
- The Confederation's SME portal: "Cybersécurité"
