How-to · Microsoft 365 admin

Handle a Microsoft 365 outage and claim SLA credits

At the end you will know whether the problem is Microsoft's or yours, have a record of the incident that a service credit claim needs, and have the claim ready to submit within Microsoft's deadline.

Updated 8 steps4 official sources
Before you start

What do you need before you handle a Microsoft 365 outage and claim SLA credits?

Applies to: Any Microsoft 365 tenant during or after an Exchange Online, Teams, SharePoint or other Microsoft 365 service incident

Admin roles you need

  • Service Support admin or Helpdesk admin role, to view Service health
  • If you buy through a Cloud Solution Provider (CSP): the contact details of your CSP partner, who submits credit claims for you

Licences

None needed.

Web portals you will use (nothing to install)

Microsoft 365 admin centerHealth > Service health shows incidents for your tenanthttps://admin.microsoft.comOpen

Reference documentation

Microsoft 365 service status pagePublic status, for when you can't sign in to the admin centerhttps://status.cloud.microsoftRead

Also have ready

  • A shared note or ticket where you log times, symptoms and users affected as the incident runs
Step by step

8 steps to handle a Microsoft 365 outage and claim SLA credits

  1. 1

    Confirm it is a Microsoft incident

    Where: Microsoft 365 admin center > Health > Service health

    Look for an active incident or advisory for the service your users report, for example Exchange Online. Note its ID: two letters for the service and a number, such as EX1550543. If you can't sign in to the admin center, check the public service status page and Microsoft's @MSFT365Status account on X.

    You should see: You know whether Microsoft has a matching incident, and its ID.
  2. 2

    Report it if it is not listed

    Where: Microsoft 365 admin center > Health > Service health > Report an issue

    If users are affected and Service health shows nothing, select Report an issue and fill in the short form. Microsoft compares it with reports from other organisations and, if the problem is on its side, adds an incident or advisory. Your report appears on the Reported issues tab.

    You should see: Your report shows on the Reported issues tab with a status.
  3. 3

    Turn on email updates for the incident

    Where: Service health > select the incident > Manage notifications for this issue

    Add up to two email addresses, ideally one that does not depend on the affected service, so updates still reach you. For all future incidents, use Customize > Email on the Service health page and choose the services to be told about.

    You should see: Updates for this incident arrive by email as Microsoft posts them.
  4. 4

    Record the impact as it happens

    Where: Your incident log or ticket

    Write down the incident ID, the time your users were first affected and when service returned, how many users and which locations were affected, and what they could not do. Keep screenshots of Service health. A credit claim needs this evidence, and it is much harder to rebuild afterwards.

    You should see: A dated log of start, end, users and locations affected, linked to the incident ID.
  5. 5

    Keep mail safe, and don't make it worse

    Where: Your users and your mail settings

    Tell users not to resend mail. Sending mail servers queue messages and retry; the SMTP standard suggests retrying for at least four to five days, so most mail arrives late rather than lost. Don't change MX records, connectors or licences to "fix" a Microsoft incident: it makes recovery harder to read. If a migration cutover is planned, postpone it until Service health is clear.

    You should see: Users know mail is delayed, and no changes were made during the incident.
  6. 6

    Read the post-incident report

    Where: Service health > History > the incident

    When Microsoft has finished its review, the incident status changes to Post-incident report published. The report gives the root cause and the steps Microsoft is taking. Service health keeps the history for 30 days, so download or save it.

    You should see: You have the report saved with your incident log.
  7. 7

    Check whether a service credit applies

    Where: Microsoft's Service Level Agreement for Online Services

    Credits are not applied automatically: you must claim them. Microsoft's SLA for Online Services sets the uptime target for each service and how downtime is counted. Microsoft says advisories are not usually eligible, so check that your event was an incident. Compare your recorded downtime with the SLA for that service.

    You should see: You know whether the downtime falls under the SLA, and for which service.
  8. 8

    Submit the claim before the deadline

    Where: Through your CSP partner (Partner Center), or Microsoft support if you buy direct

    Bought through a CSP partner: only the partner can submit the claim, in Partner Center. For Microsoft 365, the deadline is one month from the end of the billing month in which the incident happened. Send your partner the tenant ID, the incident ID, the times and duration, users and locations affected, and an email asking for the credit sent from an address on the affected tenant's domain, not a personal address.

    Bought direct from Microsoft: follow the claim process in the version of the SLA for Online Services that applies to your agreement.

    You should see: The claim is with Microsoft, or your partner, inside the deadline.
From experience

Tips for handling a Microsoft 365 outage and claiming SLA credits

Plan your notifications before the next outage

Set Service health email notifications now, to an address outside Microsoft 365 if you can. During an Exchange Online outage, an alert sent to an Exchange Online mailbox may never reach you.

Make "vendor service health is clear" a go / no-go rule

Before any migration cutover, check Service health that morning and again before you change DNS. A cutover during a vendor incident makes it hard to tell a migration fault from a service fault.

Save the evidence while it is still there

Service health only keeps 30 days of history. Save the incident page and the post-incident report with your log, so a late claim or an audit question can still be answered.

Check it worked

How do you confirm it worked after handling a Microsoft 365 outage and claiming SLA credits?

  • Service health shows the incident as Service restored or Post-incident report published.

  • Your incident log has the incident ID, start and end times, users and locations affected.

  • If you claimed a credit, you have the claim or partner ticket reference and the date submitted.

Undo

Does any of this change your tenant?

Rollback:

Nothing in these steps changes your tenant. If you postponed a cutover, reschedule it once Service health has been clear for a few days.

Troubleshooting

What errors can you see when handling a Microsoft 365 outage and claiming SLA credits?

Can't sign in to the admin center during the outage

Cause: The sign-in or admin services may be affected too.

Fix: Use the public service status page and @MSFT365Status on X until the admin center is back.

Partner says the claim can't be processed

Cause: Claims need specific evidence; customer name, ticket number and timestamps alone are not enough.

Fix: Send the tenant ID, incident ID, impact details and an email from the affected tenant's domain.

The incident is no longer in Service health

Cause: Service health keeps history for 30 days.

Fix: Save incident details and the post-incident report while they are still listed.

FAQ

Questions about handling a Microsoft 365 outage and claiming SLA credits

Are Microsoft 365 SLA credits automatic?

No. You, or your CSP partner, have to submit a claim with evidence of the downtime, within the deadline.

How long do I have to claim?

Through a CSP partner, claims for Microsoft 365 must be submitted within one month from the end of the billing month in which the incident happened. Direct customers follow the SLA for Online Services that applies to their agreement.

Will email sent during an Exchange Online outage be lost?

Usually not. Sending servers queue and retry, and the SMTP standard suggests retrying for at least four to five days, so most mail arrives late.

Where do I find the incident ID?

In the Microsoft 365 admin center under Health > Service health. It is two letters for the service followed by a number, such as EX1550543 for Exchange Online.

Sources

Official sources for handling a Microsoft 365 outage and claiming SLA credits

Vendors change menus and requirements. Confirm on the source before you rely on a detail. Last checked .

Get one practical migration tip a day

Follow Cloud-Migrations.au on LinkedIn or X for deadlines, error fixes, free templates and outage briefs as they happen.