Microsoft is down. How do you reach your people?

Hands holding a phone above an open laptop during a Microsoft 365 outage

4 minute read

On Thursday (US time), a fault in Microsoft’s cloud took Outlook, Teams, SharePoint and other Microsoft 365 services offline for organisations in more than 70 countries. Microsoft has said servers becoming unavailable caused the outage and that it has seen no sign of a cyber attack. Access has been coming back in stages, including in Australia.

For most organisations, the outage exposed one problem before any other. The tools they use to tell staff what is happening were the tools that had stopped working.

Why a Microsoft 365 outage hits continuity plans hard

Many business continuity plans assume email and Teams will be there. The plan sits in SharePoint, the response team meets in Teams, and staff updates go out by email. When Microsoft 365 goes down, all three fail at once. If staff sign in to other systems through Microsoft Entra ID, those can fail too.

That is a concentration risk. It belongs in your business impact analysis (BIA) as a dependency, with a workaround recorded against every critical process that relies on it.

The first hour of a Microsoft outage

  1. Confirm the scope. Check the Microsoft 365 service health page and an independent tracker such as Downdetector. Work out whether the fault is yours or Microsoft’s before you send anything.
  2. Assess it as a business continuity event. For most organisations a Microsoft outage is a disruption to manage under the business continuity plan, and the crisis team does not need to be stood up. Bring your business continuity lead and the owners of affected processes together on a channel that does not depend on Microsoft, such as SMS, a phone bridge or a mobile app hosted elsewhere.
  3. Check it against your recovery targets. Compare how long the outage has run, and how long it is likely to run, against the recovery time objective (RTO) and maximum tolerable period of disruption (MTPD) for each process that depends on Microsoft. Escalate to the crisis team only if the outage looks set to run well past those limits, or if it starts to affect customers, safety or your regulatory obligations in ways the plan does not cover.
  4. Tell staff early. A short message saying what is affected, what to use instead and when the next update will come stops people resetting passwords and flooding the service desk.
  5. Switch critical processes to their workarounds. Your BIA should already name the workaround for each process that relies on email, Teams or SharePoint.
  6. Start the log. Record decisions and times from the first minute. You will need them for the post-incident review and, in regulated sectors such as financial services under CPS 230, for your regulator.
  7. Contact customers and suppliers who expect to hear from you by email. Use phone, SMS or your website.

Mass communication when email is down

A mass notification tool only helps in this kind of outage if it runs independently of Microsoft. Check three things before you need it:

  • Where it is hosted. A tool that runs on Microsoft Azure or depends on Microsoft 365 to send messages may go down with everything else.
  • How people sign in when Entra ID is unavailable. Single sign-on is convenient until the identity provider is the thing that has failed. You need a break-glass login.
  • Which channels it uses. Email alerts will not reach Outlook users during an outage. SMS, push notifications and voice calls need to be ready, and staff mobile numbers need to be current.

How Battleground Live handles a Microsoft outage

Battleground Live is hosted on AWS, separate from Microsoft’s cloud. Touchbase, our mass communication module, sends SMS, email and push messages to staff and collects their replies, so the response team can see who has received the message and who needs help. Each client has a dedicated SMS number, so staff recognise the message when it arrives.

If your Microsoft identity provider is down, users can still get in through break-glass access using a magic link or password. Plans, checklists, roles and contact details are on the Battleground mobile app, so the response team can open them without SharePoint or Teams.

Questions for your next continuity review

  • Does your BIA list Microsoft 365 as a dependency for each critical process, with a recorded workaround?
  • Where is your business continuity plan stored, and can your response team open it when Microsoft is offline?
  • Have you exercised a Microsoft or cloud provider outage, including the point at which it would escalate to the crisis team?
  • Are staff mobile numbers current, and has anyone tested sending to them?
  • Which customers and suppliers rely on your email to hear from you?

If any of these are hard to answer, we can help. Talk to us about a Microsoft outage exercise for your business continuity team, or book a Touchbase demo.

Frequently asked questions

Confirm the outage on the Microsoft 365 service health page, bring your business continuity lead and affected process owners together on a channel that does not depend on Microsoft, check the expected duration against each process’s RTO and MTPD, tell staff what to use instead, and switch critical processes to the workarounds in your business continuity plan.

Use a mass notification system hosted outside Microsoft that sends SMS, push notifications or voice calls, with a sign-in method that works when Microsoft Entra ID is unavailable.

Yes. Record Microsoft 365 as a dependency in your business impact analysis, with a workaround for each critical process that relies on email, Teams or SharePoint, and exercise the scenario, including when it would escalate to the crisis team.

CEO and Founder | Battleground

Craig Goldberg

Share this article with your network

More articles