Industry trackers logging Microsoft’s uptime have counted five significant Microsoft 365 and Azure outages in the last six months, enough that IT teams are now openly discussing it as a pattern rather than bad luck. Each one has taken down some combination of Outlook, Teams, SharePoint and Entra sign-in for hours at a time, and for the huge number of UK SMEs that run their entire operation through Microsoft 365, that means email stops, calls drop, files become unreachable and staff can’t even log in to check what’s wrong.
This matters more for small businesses than for big ones, not less. A large enterprise has a dedicated IT team, a status page to watch and often a second communication channel already in place. A ten-person consultancy usually has none of that. When Microsoft goes down, the whole business goes quiet at the same time, and there’s rarely a plan B sitting ready.
Why this keeps happening
Microsoft has attributed recent incidents to a mix of configuration changes, DNS issues and capacity problems as it pushes new AI features like Copilot deeper into the same infrastructure that runs core email and file storage. The practical point for SME owners isn’t the root cause, it’s that the trend line is up, not down, and a single cloud tenant is now a single point of failure for a growing list of outages, not a one-off event to shrug off.
It’s also worth saying plainly: this isn’t really a Microsoft-bashing story. Google Workspace and AWS have had their own high-profile outages this year too. The lesson isn’t “switch providers,” it’s that every business now depends on a small number of hyperscale cloud platforms for everything at once, and none of them are immune. Betting your whole operation on one provider having a perfect uptime record is no longer a safe assumption, whoever that provider is.
What to actually do about it
You don’t need to leave Microsoft, and most businesses shouldn’t. But relying on Microsoft 365 with no fallback at all is the mistake worth fixing this month.
Get a second channel that doesn’t depend on Microsoft. A free Signal or WhatsApp group for your core team, agreed in advance, means people can still coordinate when Outlook and Teams are both down. It costs nothing and takes ten minutes to set up.
Know where your critical files actually live. If contracts, client data or financial records only exist in SharePoint or OneDrive with no separate backup, an extended outage or an account lockout leaves you stuck. A simple, automated export to a second location is cheap insurance.
Write down your sign-in fallback. If Entra ID (Microsoft’s login system) is the thing that’s down, staff can’t get into anything, including tools that aren’t Microsoft’s but use “Sign in with Microsoft.” Knowing which of your tools has an alternative login path, and testing it once, beats discovering the problem mid-outage.
Check what your status page actually tells customers. If client-facing systems depend on Microsoft infrastructure, having a one-line holding message ready to fire off, rather than scrambling to explain a silent phone line, protects the relationship more than people expect.
None of this needs to be elaborate. This is exactly the kind of practical resilience work firms like KeepSafe specialise in helping SMEs think through before an incident, not during one, because the businesses that cope well with an outage are almost always the ones that spent twenty minutes planning for it beforehand.
The takeaway: five outages in six months is a pattern, not a fluke. You don’t need to abandon Microsoft 365, you need one working fallback for communication and one for file access, tested before you need it, not during your next downtime.
