Outlook or Microsoft 365 Down? How to Keep Business Email Running With a Backup Mailbox

When Outlook will not open and Teams will not load, the first job is to work out whether the problem is on your computer or at Microsoft. The second is to keep the business talking to customers while mail is delayed. The third, once things are calm again, is to make sure one provider going down cannot stop you completely next time.

This guide was written during the global Microsoft 365 outage that began on Thursday, November 12, 2026. As of about 8 a.m. Eastern on Friday, November 13, Microsoft had not declared the incident over. Its public service health page still listed a service degradation, saying users may be unable to connect to multiple Microsoft 365 services. The rest of this guide covers how to confirm an outage, what to do right now, and how to set up a backup mailbox and a simple fallback plan for the next incident. It is a how-to, not a post-mortem of Microsoft's systems.

Quick answer

Check Microsoft's official status page at status.cloud.microsoft and the @MSFT365Status account before you assume the fault is yours. If Microsoft is down, use the phone or another channel for anything urgent, try Outlook on the web, the desktop app and mobile once each, and do not keep resending the same message. Microsoft said it is working through email queues and has extended the delivery timeout to 48 hours so delayed mail can still get through. For next time, keep an independent backup mailbox, an exported contact list, and a short written fallback plan.

At a glance

Question	What to do
Is it really an outage?	Check status.cloud.microsoft, the Microsoft 365 admin center (if you have access), @MSFT365Status, then a tracker such as Downdetector.
What if only my PC is failing?	Try another network, another device, and Outlook on the web. If those work, the problem is local.
Urgent messages right now	Use the phone, SMS, or another chat tool. Tell people your Microsoft mailbox may be delayed.
Should I resend mail?	No. Queued messages may still be delivered. Resending often creates duplicates once service returns.
Best next-time fix	A mailbox at an independent provider, a current contact export, and a one-page fallback plan.

What this outage looked like

On Thursday, November 12, 2026, shortly before 1 p.m. Eastern, users began reporting that they could not reach Microsoft 365 services including Outlook, Teams, Copilot and OneDrive. CNBC and The Verge both put the start around that time. Microsoft later said a number of servers had become "unexpectedly unavailable," which put extra load on the systems that were still running, and that there was no evidence of malicious activity. Those updates were carried by the BBC, CNBC and The Verge.

As of about 8 a.m. Eastern on Friday, recovery was under way but incomplete. Around 6 a.m. ET, Microsoft said recovery actions had been completed across portions of its Asia Pacific infrastructure, with improvements in mailbox access and delayed email processing, the BBC and NBC News reported. In an update around 7 a.m. ET, carried by the BBC, CNBC and NBC News, the company said it was seeing "encouraging recovery progress in Europe," that North America was "improving," and that Japan remained "comparatively more stable," while some users might still see intermittent issues. Windows Central, quoting a Microsoft 365 Status update that morning, said the extended 48-hour email timeout window remained in place to help protect delayed messages. That is why the "do not keep resending" advice below matters. The Verge noted there was still no time given for a full restoration.

You do not need every detail of this incident to follow the rest of the guide. The same checks and fallback steps apply the next time Outlook or Microsoft 365 is unavailable, whether the cause is a global outage or a problem in your own tenant.

1. Check whether it is really an outage

Start with official sources, then use a public tracker as a second opinion. A status page can lag, and a tracker can spike on rumors, so use both.

1. Open Microsoft's public service health page: go to status.cloud.microsoft. This is the page Microsoft has been pointing people to when the admin center itself is hard to reach. Look for a service degradation or incident covering Exchange Online, Outlook, Teams or Microsoft 365.
2. If you are a Microsoft 365 admin, also open the Service health section of the Microsoft 365 admin center. During this incident Microsoft directed admins to issue MO1550821 (and earlier to EX1550543 for Exchange Online mailbox access).
3. Read the official status account: @MSFT365Status on X. Microsoft has been posting incident updates there throughout this outage, including the 48-hour mail timeout and regional recovery notes.
4. Cross-check a public tracker: Downdetector's Outlook page (and the Microsoft 365 or Teams pages, if you use those apps) shows whether user reports have jumped. The BBC reported that Outlook reports peaked at nearly 140,000 around 18:00 GMT (1 p.m. Eastern) on Thursday, from 74 countries. Treat those figures as a snapshot of reports, not a count of everyone affected. Downdetector also warns that scammers sometimes post fake support numbers in the comments. Do not call a number you find there.
5. Rule out a local problem: try another network (Wi-Fi to mobile data, or the other way round), another device, and a different browser. If Outlook works on your phone but not on one PC, the outage is probably not the whole story.

2. What to do right now

If official status confirms a Microsoft problem, stop chasing settings on your own machine and switch to a short, calm checklist.

Use another channel for anything urgent

Call, text, or use a chat tool that does not run on Microsoft 365. Schools, police forces and other organizations in the UK were already asking the public to phone rather than email during this outage, the BBC reported. Tell your own customers and colleagues the same thing: if the matter is time-sensitive, do not wait for Outlook to come back.

If you have a second email address that does not sit on Microsoft 365, use that for urgent correspondence and say so in the message, so people know why it is coming from a different address.

Try Outlook on the web, the desktop app, and mobile once each

During this incident Microsoft warned that people might be unable to connect to Outlook on the web or on the desktop, and to other services that depend on email. CNBC, in a Friday morning update, said that warning was still in place even as recovery improved in some regions. Trying each path once is still worthwhile, because an outage is often uneven:

• Outlook on the web (sign in by typing the usual address yourself, such as outlook.office.com or your organization's sign-in page). If the web app loads, you can work from a browser even when the desktop app is stuck.
• Outlook for Windows or Mac. If you use Cached Exchange Mode, the desktop app may still show mail it already downloaded, even when it cannot reach the server. That can be enough to look up a phone number or a recent thread. New messages will not arrive, and anything you send may sit in the Outbox until the service recovers.
• The Outlook mobile app, or mail on your phone. Some people get a different result on cellular data than on office Wi-Fi.

If none of them work, do not spend the morning reinstalling the app or resetting your password. That will not bring Microsoft's servers back, and a password reset can lock you out of other devices once service returns.

Do not keep resending

If a message is sitting in the Outbox or you are not sure it left, leave it there. Microsoft said it was prioritizing the reduction of email queues and backlogs, and that it had extended the email timeout to 48 hours "to help protect delayed messages as delivery catches up." In plain terms: mail that was accepted or queued during the outage should keep trying for up to two days. Hitting Send again usually means the same message arrives twice when the queues clear.

If something is still missing after that window, then ask the sender to try again, or resend once.

Watch for phishing

Outages are a common moment for fake "restore your mailbox" or "re-authenticate now" messages. Microsoft does not need you to click a link in an unexpected email to bring Exchange Online back. Use the official status page and sign-in addresses you already know. Do not call support numbers from Downdetector comments or from ads that appear while you search for the outage.

3. Set up resilience for next time

The painful part of a long outage is rarely the missing inbox itself. It is not having another way to reach customers, a solicitor, a school, or a supplier. You can fix that without abandoning Microsoft 365, if that is still the system your team lives in.

Keep a backup mailbox on an independent provider

The simplest fallback is a second mailbox that does not run on Microsoft's infrastructure. It can be a published backup address (for example, a different domain, or a clearly labeled backup contact on your website and email signatures), or a mailbox your staff can send from when Outlook is down. The important part is independence: if the backup sits inside the same Microsoft 365 tenant, it will go down with the rest.

One independent option is Thexyz, a business email host based in Canada. Mailboxes are on your own domain, plans start at $2.95 a month for 10 GB, and the service is built to work with Outlook, Apple Mail and Gmail apps, according to Thexyz's email hosting page. A fully managed migration from providers including Microsoft 365 is included, and Thexyz says the copy can run in the background before you change any DNS. There is a 5-day money-back guarantee rather than a free trial, and no long contract, according to its provider comparison page. You can also use another independent host. The point is to have a working mailbox that is not on the same platform as your primary mail.

Do not try to "fail over" your whole company domain to a new provider in the middle of an outage unless you already have that plan rehearsed. Changing MX records under pressure can split delivery, and the change can take hours to spread across the internet.

What a backup MX record is (and what it is not)

Email routing is controlled by MX records in your domain's DNS. Each MX record has a priority number. Sending servers try the lowest number first (the primary). If that server cannot be reached, they should try the next one. A backup MX is a second mail server, usually at a different provider, that can accept mail for your domain when the primary is down, hold it, and pass it on when the primary returns.

That idea is sound, but it is easy to get wrong:

• Microsoft 365 generally expects to be the primary mail exchanger for a domain that uses Exchange Online. Adding a backup MX in front of or beside Microsoft 365 is a specialized setup, not a simple toggle. A bad configuration can delay mail, confuse spam filters, or cause some messages to be rejected.
• Not every sending server fails over to a backup MX cleanly, so it is not a guarantee that every message will be accepted during an outage.
• Changing MX records during a live incident is usually a poor emergency move, because of DNS caching and the risk of mail being delivered to two places at once.

For most small teams, a documented backup mailbox at an independent provider is more practical than trying to bolt a backup MX onto Microsoft 365 after the fact. If you do want a true backup MX, plan it in advance with someone who understands both your DNS and Microsoft 365's mail-flow requirements, and test it before you need it.

Export your contacts and keep the file somewhere else

If Outlook is unreachable, a contact list that only lives in Microsoft 365 is not much use. Export a copy on a regular schedule and store it outside the same account (an encrypted drive, a printed call sheet for key contacts, or a mailbox at your backup provider).

In Outlook on the web, open People, use the manage/export option, and save a CSV. In Outlook for Windows, you can export contacts from File > Open & Export > Import/Export. The exact labels vary by Outlook version, so if you cannot find the command, Microsoft's own help pages for exporting contacts are the place to check. Update the file when staff or key customers change, not only after an outage.

Write a one-page fallback plan

A plan only works if people can find it when email is down. Keep a short document (and a paper copy) that answers:

• Where to check official status (the Microsoft health page and @MSFT365Status).
• Which phone numbers, SMS threads or non-Microsoft chat tools the team should use.
• The backup email address, who can sign in to it, and who is allowed to send from it.
• How you will tell customers (website banner, voicemail greeting, social post).
• Who is allowed to change DNS, and a reminder not to change MX records in a panic.
• Where the exported contact list is stored.

Test it once: send a message from the backup mailbox, call the backup numbers, and confirm that at least two people can reach the contact list without signing in to Microsoft 365.

Frequently asked questions

Will my queued mail still arrive?

Microsoft said it had extended the email timeout to 48 hours to help protect delayed messages while it worked through queues. That is a strong reason not to resend. It is not a promise that every message will arrive. If something is still missing after service is clearly back, follow up once.

Should I switch off Microsoft 365 today?

Not as an emergency move, unless you already have another provider configured and tested. A mid-outage cutover can lose or split mail. Use other channels now, then decide later whether you want Microsoft 365 as your only mail system or as one system with a backup.

Is a personal Gmail or iCloud address a good backup?

It is better than having no second channel at all, and it is fine for a short emergency. For a business, a mailbox on your own domain at an independent host is cleaner: it looks professional, and you can move it later without changing the address. Do not start putting company mail into an unpaid personal account as a long-term habit.

Can I just add a backup MX and leave Microsoft as primary?

Sometimes, if it is designed carefully. It is not a weekend DIY fix for a Microsoft 365 tenant, and it is not something to configure while mail is already failing. Start with a separate backup mailbox and a written plan; treat backup MX as a later, planned project if you still need it.

Sources

1. BBC News live: Global Microsoft outage hits schools, doctors and businesses — timeline, Microsoft recovery comments, cause, 48-hour queues context, and UK impact.
2. BBC News: What we know about the global Microsoft outage — Downdetector figure of 140,000 Outlook reports from 74 countries on Thursday evening; Microsoft's "unexpectedly unavailable" servers explanation.
3. BBC News: Global Microsoft outage affecting tens of thousands of users — Outlook reports peaked at nearly 140,000 around 18:00 GMT Thursday, per Downdetector.
4. CNBC: Outlook outage continues as Microsoft flags 'encouraging recovery progress' — start shortly before 1 p.m. ET Thursday; 48-hour timeout; Friday morning regional recovery.
5. The Verge: Microsoft Outlook is still down — services affected; Microsoft status wording, including the 48-hour email timeout.
6. Windows Central: Microsoft 365, Outlook, Teams, OneDrive, and Copilot are still down — @MSFT365Status updates, incident MO1550821, 48-hour timeout still in place as of a Friday morning update around 7:33 a.m. ET.
7. NBC News live: Microsoft outage hits Outlook, Teams across the world — 6 a.m. and about 7 a.m. ET Friday recovery updates, live blog updated 7:57 a.m. ET.
8. Microsoft service health status and @MSFT365Status — official places to check during an incident.
9. Downdetector: Outlook status — public user-report tracker (not an official Microsoft page).
10. Thexyz, email hosting plans, email migrations, and email provider comparison — features and prices cited above, checked November 13, 2026.

Last checked about 8 a.m. Eastern, November 13, 2026. This is a practical guide written during an ongoing incident; Microsoft's status may have changed since it was last checked.