By Delles Simon · Marketing Coordinator · The Link Source
When Microsoft Goes Down: A Business Continuity Plan for Systems You Don't Control

When Exchange Online or Teams outages hit worldwide, email stops and business processes grind to a halt. Here is how Houston businesses protect operations when third-party cloud vendors go down.
Table of Contents
What Happened
Problems began at about midday on Thursday, November 12, 2026, in the US. Users in more than 70 countries and territories reported outages. More than 7.4 million reports were logged worldwide, peaking at nearly 20,000 a minute, more than 1,500 times normal levels.
Users reported that Microsoft 365 services were interrupted, including Outlook, Teams and the AI assistant Copilot. Microsoft said the outage hit features tied to Exchange Online, its cloud-based email and scheduling hub. Apps not connected to it kept working. Many Outlook users said they couldn't send or receive emails or had trouble loading their inboxes.
The outage affected governments, courts, healthcare services, businesses, airlines and banks in many places. Some hospital staff switched to paper records. In the US, staffers at the House of Representatives told Politico they had lost a day of work. Lost productivity is estimated at about $7 billion.
Cause
The outage surfaced after a network configuration update that Microsoft rolled out overnight. After the change, a number of servers stopped sending heartbeat signals to the load balancers, which marked those servers as unhealthy and took them offline.
As a result, all incoming traffic was redirected to the small number of servers that were still online. These servers were not sized to handle the full volume of requests. They quickly became overloaded, reached 100% CPU utilization and began returning HTTP 500 errors. This is why users experienced disruptions in Outlook and Teams.
Microsoft said the outage was "not related to cyber" or artificial intelligence agents. It said the incident "was introduced by an issue during an earlier network capacity upgrade that surfaced" following a recent network change. A process that establishes secure connections became overwhelmed, and servers stopped handling requests. Microsoft began rolling back the network configuration change.
How Microsoft Fixed the Issue
Microsoft's first attempts to restart the affected servers did not fix the problem. Each time servers came back online, peak traffic hit them while other servers kept dropping offline. This created a cycle of overload and failure.
To break this cycle, Microsoft bypassed the standard load-balancing rules and throttled traffic. Instead of sending all traffic at once, it brought servers back online one at a time and routed a limited, controlled amount of traffic to each. This approach was more stable but took time. During recovery, users saw intermittent behavior: some messages went through while others were delayed or failed. On Saturday, Microsoft said services had recovered across Teams, Outlook and its other products.
Why a Vendor Outage Becomes Your Problem
Email sits behind more business processes than most continuity plans account for. Approvals, customer replies, vendor invoices, scheduling, password resets and system alerts all move through it. When Outlook stops, those processes stop with it.
Security and customer contact have the same exposure. Defender XDR was among the affected services, so your security team's monitoring tools can go down at the same time as the systems they monitor. Dynamics 365 Contact Center was also affected, which can cut off the channels customers use to reach you.
How to Keep Your Business Running When a Vendor Goes Down
Not running everything in the cloud helped. Organizations that host Microsoft services on their own servers were somewhat spared. Fully moving back on-premises isn't realistic for most companies, but it shows the value of not relying on one cloud service for every critical function.
Pre-planned workarounds also made a difference. The disruption was mostly limited to email and scheduling, so teams that had other tools ready kept working. Where no plan existed, people improvised.
The outage couldn't have been prevented from the customer side, but businesses could have reduced the impact. You can't shorten a vendor's recovery time, but you can decide how your business operates while it recovers. These standard continuity steps can help:
- Map dependencies. List every vendor platform your operations rely on, then trace which business processes run through each one. Include the layers people forget: identity and sign-in, email, calendar, voice, contact center and security tooling. This map shows where a single provider's outage reaches the most work.
- Set downtime tolerance for each function. A two-hour gap in internal chat and a two-hour gap in customer contact carry different costs. Decide how long each function can be disrupted before it affects revenue, compliance or customers, and build your fallbacks in that order.
- Set up a backup communication channel. Agree in advance on one non-Microsoft fallback, such as a phone tree, text group or another chat tool. Keep a staff contact list with phone numbers outside Outlook, and name who sends updates and who approves them.
- Keep critical information available offline. Store client contacts, the next few days' meeting schedules and active project files somewhere you can open without signing in to Microsoft 365. Synced local copies or a printed or exported list works.
- Protect customer contact. Decide in advance how customers reach you when a core channel fails. That can include alternate call routing, updated greetings and a secondary channel for urgent requests. Have a short message ready for your website or social media explaining how to reach you by phone. During this outage, many organizations told customers to call instead of email.
- Cover order-taking and payments. Plan how you'll take orders, process payments and send invoices if email goes down. One sheriff's office had to move bond payments to in-person only because Outlook email was down.
- Write down manual workarounds for key processes. Approvals, password resets and system alerts often run through email. For each one, note who can approve by phone, how to reset access without email and where alerts go if the inbox is down. A one-line workaround written in advance saves hours of guesswork during an outage.
- Know where to get updates. Bookmark the Microsoft 365 status page and Microsoft's status account on X so you can track the vendor's progress.
- Review vendor agreements and concentration risk. Service credits in a vendor agreement generally do not cover lost productivity or missed customer commitments. Review your SLAs, ask what a vendor's incident communication commitments look like, and note where one provider supports too many critical functions at once. Weigh that risk against the steep switching costs of replacing deeply embedded tools.
- Test the plan on a schedule. A plan that has never been exercised will have gaps. Run a 30-minute tabletop exercise at least once a year using a realistic scenario, such as a multi-day email outage. Update the plan after every test and major technology change.
How The Link Source Can Help
The Link Source is a local Houston-based managed IT and business continuity provider. Because The Link Source is vendor-neutral, you get recommendations that are not tied to a single provider. Our team can help you map which communications, contact center, cloud and security platforms your operations depend on, identify where one provider creates concentrated risk, and design fallback options that fit your business. Where you need ongoing support, our managed services teams can help you run and maintain that design.
Contact The Link Source to start a conversation about continuity planning for your core platforms, resilience, and cloud strategy.
Article FAQ
What caused the November 2026 Microsoft 365 outage?
The outage was triggered by a network configuration update rolled out by Microsoft that caused servers to stop sending heartbeat signals to load balancers. The remaining online servers were overloaded with traffic, reaching 100% CPU utilization and returning HTTP 500 errors across Exchange Online, Teams, and Outlook.
Can small businesses prevent third-party cloud outages?
No customer can prevent a cloud provider from experiencing an outage. However, businesses can dramatically reduce the impact by establishing secondary communication channels, offline contact records, and tested manual workarounds for critical workflows like approvals and invoicing.
What is concentration risk in cloud services?
Concentration risk occurs when an organization relies on a single vendor for identity, email, chat, security monitoring, and file storage. If that provider goes down, all interdependent systems fail at once.
How often should a business test its continuity plan?
At least once a year through a 30-minute tabletop exercise testing a realistic outage scenario, plus whenever major infrastructure or platform changes occur.
