When Microsoft 365 Goes Down: A One-Page Continuity Checklist
In mid-November 2026, a widely reported outage disrupted Exchange Online, Teams, Copilot, SharePoint and OneDrive for many organisations at once. The provider said it was not a cyber attack. For most businesses, what mattered was not the cause but the hours without email and chat. Plans that assume your own servers are the failure point miss this scenario completely.
The short version
- A shared cloud outage stops many functions together and is outside your control, so the work is done before it happens.
- Three things decide how bad a day is: a second way to reach staff, a way to reach customers, and a list of what must keep running.
- Practise it. A drill without email will show gaps no document review finds.
The checklist by time
| When | Action | Owner |
|---|---|---|
| First 30 minutes | Confirm it is a provider outage using the provider's status page and a second source, and rule out your own network and account | IT |
| First 30 minutes | Tell staff through the backup channel, not the failed one, and give one point of contact | Crisis lead |
| First 2 hours | Switch priority teams to workarounds: phone lists, personal-device messaging agreed in advance, offline copies of key procedures | Team leads |
| First 2 hours | Send a holding message to key customers and partners by phone or an alternative channel | Communications |
| First day | Decide which activities pause and which continue manually, and record each decision with the time | Crisis lead |
| After restoration | Check for delayed or lost messages, restart paused work in priority order, and note what broke | IT and team leads |
| Within 2 weeks | Hold a review and update the continuity plan, contact lists and vendor questions | Continuity coordinator |
What to prepare before the next one
- A dependency list. Which services does each critical process rely on, and who owns each one?
- A backup way to talk. Agree a channel that does not depend on the same provider and test that everyone can join it.
- Offline copies. Contact lists, key procedures and emergency access instructions stored where an outage cannot reach them.
- Vendor questions. Ask what the service level covers, how quickly you will be told, and what exclusions apply.
Frequently asked questions
Is a provider outage covered by a business continuity plan?
Only if the plan names it. Many plans cover fire, flood and loss of premises, and treat IT as a single block. A provider outage needs its own scenario, with the services mapped, the workarounds agreed and the communication route tested, because the usual tools are exactly what has failed.
What does a service level agreement actually give us?
Usually a commitment to a level of availability and a service credit if the provider misses it. It rarely compensates for lost business and often has exclusions. Read what counts as downtime, how it is measured and how claims are made, then plan on the assumption that credits will not cover your losses.
Should we use two providers for email?
Sometimes, but not always. A second provider adds cost and complexity, and it only helps if staff and customers can actually use it during an outage. For many organisations, an agreed backup communication channel and offline contact lists give most of the benefit at far lower cost.
How often should we test this?
At least once a year, and after any major change in the tools you rely on. A short drill that takes away email and chat for a morning will show whether people know the backup channel, whether contact lists are current and whether customers can still reach you.