Episode Details

Ever wondered what happens when just one M365 service goes down, but it drags the others with it? You're not alone. Today we're unpacking the tangled reality of M365 outages—and why your existing playbook might be missing the hidden dependencies that leave you scrambling. Think Exchange going dark is your only problem? Wait until SharePoint and Teams start failing, too. If you want to stop firefighting and start predicting, let’s walk through how real-world incident response demands more than ‘turn it off and back on again’.Why M365 Outages Are Never Just One ThingIf you’ve ever watched a Teams outage and thought, “At least Exchange and SharePoint are safe,” you’re definitely not alone. But the reality isn’t so generous. It starts out as a handful of complaints—maybe someone can’t join a meeting or sends a message and it spins forever. Fifteen minutes later, email sends slow down, OneDrive starts timing out, and calendar sync is suddenly out of whack. By noon, you’re walking past conference rooms full of confused users, because meeting chats are down, shared files are missing, and even your incident comms are stalling out. This is Microsoft 365 at its most stubborn: a platform that hides just how tangled it really is—until the dominoes start to fall.Let me run you through what this looks like in the wild. Imagine kicking off your Monday with an odd Teams problem. Not a full outage—just calls that drop and a few people who can’t log in. Most admins would start with Teams diagnostics, maybe check the Microsoft 365 admin center for an alert or two. But before you can even sort the first round of trouble tickets, someone from HR calls—Outlook can’t send outside emails. This isn’t a coincidence. The connection you might not see is Azure Active Directory authentication. Even if Teams and Exchange Online themselves are showing ‘healthy’ in the portal, without authentication, nobody’s getting in. SharePoint starts to lock people out, group files become unreachable, and by noon, half your org is stuck in a credentials loop while your status dashboard stays stubbornly green. It doesn’t take much: a permissions service that hiccups, a regional failover gone wrong, or an update that trips a dependency under the hood.August 2023 gave us a real taste of this ripple effect. That month, Microsoft confirmed a major authentication outage that—on paper—started with a glitch in Azure AD. The first alerts flagged Teams login issues, but within twenty minutes, reports flooded in about mail flow outages on Exchange and SharePoint document access flatlining. Even Microsoft’s own support status page choked for a while, leaving admins to hunt for updates on Twitter and Reddit. Nobody could confirm if it was a cyberattack or just a bad code push. In these moments, it becomes obvious that Microsoft 365 doesn’t break the way single applications do—it breaks like a city-wide traffic jam. One red light on a busy avenue, and suddenly cars are backed up for miles across unconnected neighborhoods.That’s the catch: invisible links are everywhere. You can have Teams and SharePoint provisioned perfectly, but the minute a shared identity provider stutters, everything locks up. And here’s the twist—when a service is ‘up,’ it doesn’t always mean it’s usable. You might see the SharePoint site load, but try syncing files or using any Power Platform integration and watch the error messages pile up. Sometimes, services remain online just long enough to confuse users, who can open apps but can’t save or share anything critical. It’s like getting into the office building only to find the elevators and conference rooms all badge-locked.Let’s talk about playbooks, since this is where most response plans fall flat. Most orgs have runbooks or OneNote pages that treat each service as an island. They’ll have a Teams page, an Exchange checklist, and maybe a few notes jammed under ‘SharePoint issues.’ That model worked in the old on-premises days, when an Exchange failure meant y