Microsoft 365 went down again this week — Outlook, Teams, SharePoint and several other services all affected at once, traced back to an authentication problem in Microsoft’s own infrastructure. If that sentence feels familiar, it should: this is the fifth major incident in six months, not a one-off bad day. For most small businesses, Microsoft 365 isn’t just email anymore — it’s the phone system, the meeting room and the internal chat all in one place. When it goes down, a lot goes down with it.
This is a pattern, not a one-off
Since late last year, Microsoft’s cloud services have had repeated incidents — a global DNS failure, more than one regional Azure outage, disruption to Copilot, and a nine-hour Microsoft 365 outage in the spring, on top of this week’s authentication failure. There are structural reasons this keeps happening: Azure is absorbing an enormous amount of new AI workload and infrastructure growth, its services are tightly interdependent so one component failing can cascade into several others, and relying on a single hyperscaler for practically everything means a lot of businesses go down for the same reason at the same time. None of that is likely to reverse soon — if anything, the dependency is only growing.
What it actually costs when it happens
Industry estimates put the average cost of Microsoft 365 downtime at somewhere around £5,600 an hour for a typical UK small business, once you account for lost productivity and the knock-on effect on customers. Over a nine-hour outage — not an unrealistic length, based on this year’s incidents — that adds up to somewhere between £20,000 and £45,000 for a business of around twenty staff. And yet nearly half of UK small businesses have no documented plan for what to do when Microsoft 365 goes down. Most just wait it out.
Backup isn’t the same as “it’s in the cloud”
A common misunderstanding is assuming Microsoft is already backing everything up for you. It isn’t, not in the way that actually protects you — Microsoft’s retention settings are built to prevent accidental data loss for a limited window, not to recover you from a prolonged outage or give you an independent copy of your data. A proper third-party backup for your Microsoft 365 data is one of the least expensive, highest-impact things a small business can put in place — typically only a few pounds per user per month — and it’s the single control that makes the biggest difference when something goes wrong.
A few things that actually help
- An independent way to communicate when Teams is down. If your only way of reaching colleagues or customers lives inside Microsoft 365, an outage cuts off your ability to even tell people what’s happening.
- A written outage playbook, rehearsed occasionally. Knowing who does what in the first twenty minutes of an outage matters more than any technical fix, and it costs nothing beyond the time to write it down and run through it.
- Proper MFA and conditional access. Separate from outage resilience, but it reduces how often you’re dealing with a security incident on top of everything else — the two problems compound each other.
- Service health alerts routed somewhere your team will actually see them. Knowing within minutes that it’s a Microsoft-wide issue, rather than assuming it’s something on your end, saves a lot of wasted troubleshooting time.
None of this requires moving off Microsoft 365 or building an expensive multi-cloud setup — for the vast majority of small businesses, that would be a disproportionate response to the actual risk. It just requires treating “what happens when it goes down” as a question worth answering before it happens, not during.
Not sure how exposed your business actually is if Microsoft 365 goes down tomorrow? Get in touch with Valetech Solutions for a proportionate look at your backup, continuity and cloud setup.