When Microsoft 365 Goes Down: Why Your Business Needs an Outage Plan and a Tested Backup of Its Cloud Data

Microsoft had an outage the day I recorded this, and email and Teams went with it for hours. It doesn't happen often, but it happens, and most companies have never decided what their employees do when it does. Write the plan, practice it, and keep a backup of your cloud data that you've actually restored.

Keith Parker
2026-11-16
8 minute read
When Microsoft 365 Goes Down: Why Your Business Needs an Outage Plan and a Tested Backup of Its Cloud Data

Christmas came early for some folks today. Microsoft is having an outage.

Email and Teams are out. When I recorded this, we were going on about four hours, so probably most of the day.

Think about what that does to your office. Nobody can send a quote. Nobody can answer the client who's waiting on a reply. The meeting at two o'clock was a Teams link, and the Teams link doesn't work. People start texting each other asking if their email is down too, and somebody restarts their laptop three times because they assume it's them.

Then they wait.

I'd like to point out that we should have a contingency plan. Not for the whole world ending. For the much more likely day when the one service your whole company runs on stops working for an afternoon.

Outages are rare, but they're not a surprise

It doesn't happen often. In my experience, once or twice a year is the average. But we should have a plan in place for when it does happen.

Microsoft says the same thing in its own terms of service. The Microsoft Services Agreement, which covers its consumer services, says plainly that "all online services suffer occasional disruptions and outages," and that Microsoft isn't liable for what you lose as a result. It goes on to recommend that you regularly back up the content and data you store there.

That's the company running the service telling you to plan for it going down. I'd take them at their word.

The cloud did a lot of good things for small businesses. You don't have a server in a closet anymore, you don't patch Exchange on a Saturday, and your people can work from anywhere. What it didn't do is make outages go away. It moved them. When your email lived on a server in your office, an outage was your problem to fix. Now it's Microsoft's problem to fix, and your problem to live through.

What an outage plan actually covers

We should have an incident response plan that calls for, hey, what's going to happen when email goes out?

What are your employees going to do?

That's the whole question, and most companies have never answered it. A good plan for a Microsoft 365 outage fits on a page or two. It doesn't need to be a binder. It needs to answer the questions people will be asking each other in the first hour.

Who finds out first, and how do they confirm it's Microsoft and not us? Microsoft publishes a public service status page for exactly this, for when you can't sign in to the admin center to check service health yourself. Somebody at your company should know where it is before the day they need it.

How do we reach each other? If your phone list lives in Outlook and Outlook is down, you don't have a phone list. Keep one somewhere else. A printed copy in a drawer is fine.

How do we reach clients? Decide ahead of time who calls the clients with something urgent in flight, and from what phone.

What keeps working? Work out which jobs can carry on without email and which ones stop cold, so people aren't sitting idle waiting for an all-clear.

Who says it's over? Somebody should own the call that the service is back and everyone can pick up where they left off.

Have a plan, then practice it

We should have a plan, and we should practice it.

A plan nobody has read is close to no plan at all. The first time anyone opens it shouldn't be in the middle of the outage, because that's when you find the phone number that's wrong or the person it names who left last year.

Practicing doesn't mean pulling the plug on your own email. Sit your leadership down for an hour, say "email and Teams have been down since nine this morning," and walk through what happens next. You'll find the gaps fast. NIST's contingency planning guide for information systems is built on the same idea: a plan gets tested, people get trained on it, and it gets updated with what you learn. Ready.gov publishes free business continuity materials that include a test exercise planner, if you want a structure to follow.

Back up what you keep in the cloud

We should also have a backup of all of our content that is stored at Microsoft or Google or wherever it's stored, right?

Most outages are a bad afternoon. The email comes back and nothing is lost. But what if data is lost and we can't recover? A deleted mailbox, a SharePoint library somebody wiped, a ransomware infection that synced its damage up to OneDrive. Microsoft keeps the service running. Keeping a copy of your data that you control is on you.

For our clients, we back everything up to a third-party service. It's part of their compliance requirements. Even if you don't have compliance requirements, I always recommend it.

The rule of thumb CISA gives for backups is the 3-2-1 rule: keep three copies of any important file, one primary and two backups, on two different types of media, with one copy offsite. For a business that lives in Microsoft 365 or Google Workspace, the cloud copy is your primary. The backup should sit somewhere else, with a different company, so one outage or one mistake can't take out both.

A backup you haven't restored is a guess

I always recommend we have a backup and we have a routine plan of testing the backup, because you don't know if it works until you actually go to restore it.

This is the part that gets skipped. A backup that reports success every night tells you a copy ran. It doesn't tell you that you can find the mailbox you need, bring it back, and open it. The only way to know that is to do it.

Put it on the calendar. Pick a mailbox and a folder from SharePoint or OneDrive, restore them to a test location, open the files, and check that they're current. Write down how long it took, because that's roughly how long you'd be waiting on the bad day. Then do it again next quarter.

How to put this in place this month

1. Write the one-page plan. Answer the five questions above: how you confirm it, how you reach each other, how you reach clients, what keeps working, and who calls the all-clear. Keep a printed copy.

2. Get your contacts out of Outlook. Employees, key clients and key vendors, with phone numbers, somewhere that doesn't depend on Microsoft being up.

3. Bookmark the status page. Make sure the people who'd field the first "is email down?" message know where to check.

4. Run a one-hour drill. Leadership in a room, the scenario on the table, every unanswered question written down. Give each one an owner and a date.

5. Back up your Microsoft 365 or Google data with a third party. Email, OneDrive, SharePoint, Teams files. If you already do, confirm it covers everything you've added since it was set up.

6. Schedule a test restore. Then schedule the next one.

Where to start

So my takeaways are: there was a Microsoft outage today. Have a contingency plan. Email will go out again. And also have a backup plan, and a schedule of testing those backups. All very important to do.

Start with these, in this order:

  1. Print your company phone list and put it somewhere everyone knows.
  2. Ask whoever handles your IT when they last restored something from your cloud backup, and what it backs up.
  3. Book an hour with your leadership team to talk through a day without email.
  4. Turn what comes out of that hour into the one-page plan, and put a copy in the drawer with the phone list.

Hopefully it comes up soon.


One question to answer this week: if email and Teams went down tomorrow morning, what would your employees do? If nobody knows, book 15 minutes and we'll help you write the plan and test your backup, or run our free external scan to see what an attacker can already find about your company.

Keith Parker is the CEO of Methodology IT, a managed IT services provider in Los Angeles that has supported healthcare practices, nonprofits, and growing businesses across Southern California since 1997.

Sources: Microsoft Services Agreement, published July 30, 2026 · Microsoft Learn: How to check Microsoft 365 service health · Microsoft service health status · NIST SP 800-34 Rev. 1, Contingency Planning Guide for Federal Information Systems · Ready.gov: Business Continuity Planning · CISA: Data Backup Options

Ready when you are

Ready to make IT work?

No pressure, no sales pitch. A senior tech will walk your environment with you and leave you with a report, whether you hire us or not.

Keith Parker
By Keith Parker · Founder & CEO, Methodology IT · UCLA Computer Engineering
Keith Parker
About the author
Keith Parker
Founder & CEO, Methodology IT · Los Angeles

Keith Parker founded Methodology IT in 1997 and has run IT and security for Los Angeles medical practices, law firms, nonprofits and entertainment companies ever since. He holds a degree in computer engineering from UCLA and serves as a CIS instructor at Los Angeles Southwest College (LASC). After years of coaching youth softball and baseball, he now teaches a CIS class and runs a technology and entrepreneurship club at a Los Angeles high school serving underrepresented communities. He also builds the AI tools his team and clients use every day, writes a weekly newsletter on using AI safely at work, and was featured in MSP Success magazine for helping businesses through the 2025 SoCal wildfires.

Keith Parker
By Keith Parker · Founder & CEO, Methodology IT · UCLA Computer Engineering
Keith Parker
About the author
Keith Parker
Founder & CEO, Methodology IT · Los Angeles

Keith Parker founded Methodology IT in 1997 and has run IT and security for Los Angeles medical practices, law firms, nonprofits and entertainment companies ever since. He holds a degree in computer engineering from UCLA and serves as a CIS instructor at Los Angeles Southwest College (LASC). After years of coaching youth softball and baseball, he now teaches a CIS class and runs a technology and entrepreneurship club at a Los Angeles high school serving underrepresented communities. He also builds the AI tools his team and clients use every day, writes a weekly newsletter on using AI safely at work, and was featured in MSP Success magazine for helping businesses through the 2025 SoCal wildfires.

  • Physical Security
  • IT Compliance
  • Custom Software
  • IT Consulting
  • Backup & Disaster Recovery
  • Incident Response
  • Websites & AI Agents
  • ·Burbank·Glendale·San Fernando Valley·Pasadena·Santa Clarita·Hollywood·Inglewood·South Bay·West LA
    CRN AI Innovators 2026