What one outage means for business continuity

All insights
Sage Insights Business continuityOperational resilienceThird-party risk

What one outage means for business continuity

On 13 November, a global Microsoft 365 outage removed access to MS Outlook, Teams and other services for the majority of the day. Most organisations could only wait. It was an unplanned test of business continuity.

When one provider runs email, chat, files and security, one fault can stop them all.

Briana van Tilburg

Co-founder and Managing Director

Monica Mihai

Co-founder and Specialist Consultant

3 minute readNovember 2026

We have both led our share of IT upgrades and business transformations across airlines, banks, healthcare, insurance and government.

Experience tells us that in any change, something will go wrong at some point. Good planning and testing lower the risk, but they cannot rule out incidents entirely.

Last week showed that even the world's largest tech companies are not immune.

Microsoft 365 began failing around the world at about 3.15am AEDT on Friday 13 November. Outlook, Teams, SharePoint and Copilot were all hit. So were security tools such as Defender and Purview.

Microsoft traced the fault to an earlier network upgrade. A later change brought it to the surface. Servers were overwhelmed setting up secure connections. There was no sign of a cyber attack.

Recovery was slow. Restored servers could fail again once traffic came back. Full service took about a day to return.

Who felt it

In Australia, ASIO and the Australian Federal Police were affected. Service NSW closed some regional centres. Courts, health services and schools were disrupted.

Some small businesses reported losses of up to $30,000 in a day. Energy Queensland enacted its business continuity plans. The Australian Government activated an agreement with Microsoft to prioritise Australian interests.

Why it hit so hard

Forrester says consolidating productivity, identity, security and cloud under one vendor "creates systemic fragility". For many organisations, one provider runs email, chat, files, sign-in and security tools. When the main channel failed, the backup channel failed with it. Some people were locked out because their sign-in codes were sent by email.

Staff also lost the tools they use to raise problems. Some turned to personal email and apps, which can move sensitive data outside company controls. Scammers moved fast too. Microsoft warned of fake outage notices that used the real incident number.

Key issues for organisations

Resilience is not about how fast a provider recovers. It is about how well you keep operating while you wait.

Many continuity plans focus on recovering systems. But business continuity is also about considering how you can run your business operations, without your usual platforms, and for how long.

For banks, insurers and super funds, this is now a prudential rule. APRA's CPS 230 requires them to keep critical operations within tolerance levels through severe disruptions. They must test their continuity plans every year, including for disruption at a key provider.

CPS 230 treats core technology providers as material service providers. Entities must manage concentration risk, and the providers their providers rely on. A disruption outside tolerance must be reported to APRA within 24 hours.

What comes next

Microsoft has promised a full report. Run your own review now, while it is fresh. Record what stopped, for how long, and which workarounds held.

"Everyone needs a Plan B and it can't be hypothetical. It must be rehearsed and tested," said Richard Buckland of UNSW.

Priorities for leaders

  1. Know your critical business processes.

    Look past daily work to monthly, quarterly and annual cycles. Make that calendar visible to everyone.

  2. Map where your dependence sits.

    Trace each critical operation to the platforms behind it, from sign-in to security tools.

  3. Decide what must keep running.

    Set the minimum service you need, and how long you can go without it.

  4. Build real alternatives.

    Set up a backup channel outside your main provider, such as SMS or a phone tree. A backup on the same platform is not a backup.

  5. Rehearse a full provider failure.

    Test a scenario where one provider is down for a day. Include leaders, IT and key suppliers.

  6. Make it a board issue.

    Give the board clear ownership of concentration risk. A supplier's service level is not a continuity plan.

If your organisation needs help to be prepared for the next ‘systems down’ event, we can help. Get in touch for a free and confidential chat.

General information only.

Sources. ABC News, 13 November 2026. CIO Dive, 13 November 2026. iTnews, 14 November 2026. Forrester, 13 November 2026. Control Risks, 13 November 2026. APRA, Prudential Standard CPS 230 and Prudential Practice Guide CPG 230. Microsoft, July 2024. Optus network outage, 8 November 2023.

More insights

View all insights

Privacy and AI

What 10 December means for automated decisions

From 10 December 2026, privacy policies must describe certain automated decision-making. The hard part is finding where those decisions already happen.

3 minute read Read article
All insights
Sage Insights Business continuityOperational resilienceThird-party risk

What one outage means for business continuity

On 13 November, a global Microsoft 365 outage removed access to MS Outlook, Teams and other services for the majority of the day. Most organisations could only wait. It was an unplanned test of business continuity.

When one provider runs email, chat, files and security, one fault can stop them all.

Briana van Tilburg

Co-founder and Managing Director

Monica Mihai

Co-founder and Specialist Consultant

3 minute readNovember 2026

We have both led our share of IT upgrades and business transformations across airlines, banks, healthcare, insurance and government.

Experience tells us that in any change, something will go wrong at some point. Good planning and testing lower the risk, but they cannot rule out incidents entirely.

Last week showed that even the world's largest tech companies are not immune.

Microsoft 365 began failing around the world at about 3.15am AEDT on Friday 13 November. Outlook, Teams, SharePoint and Copilot were all hit. So were security tools such as Defender and Purview.

Microsoft traced the fault to an earlier network upgrade. A later change brought it to the surface. Servers were overwhelmed setting up secure connections. There was no sign of a cyber attack.

Recovery was slow. Restored servers could fail again once traffic came back. Full service took about a day to return.

Who felt it

In Australia, ASIO and the Australian Federal Police were affected. Service NSW closed some regional centres. Courts, health services and schools were disrupted.

Some small businesses reported losses of up to $30,000 in a day. Energy Queensland enacted its business continuity plans. The Australian Government activated an agreement with Microsoft to prioritise Australian interests.

Why it hit so hard

Forrester says consolidating productivity, identity, security and cloud under one vendor "creates systemic fragility". For many organisations, one provider runs email, chat, files, sign-in and security tools. When the main channel failed, the backup channel failed with it. Some people were locked out because their sign-in codes were sent by email.

Staff also lost the tools they use to raise problems. Some turned to personal email and apps, which can move sensitive data outside company controls. Scammers moved fast too. Microsoft warned of fake outage notices that used the real incident number.

Key issues for organisations

Resilience is not about how fast a provider recovers. It is about how well you keep operating while you wait.

Many continuity plans focus on recovering systems. But business continuity is also about considering how you can run your business operations, without your usual platforms, and for how long.

For banks, insurers and super funds, this is now a prudential rule. APRA's CPS 230 requires them to keep critical operations within tolerance levels through severe disruptions. They must test their continuity plans every year, including for disruption at a key provider.

CPS 230 treats core technology providers as material service providers. Entities must manage concentration risk, and the providers their providers rely on. A disruption outside tolerance must be reported to APRA within 24 hours.

What comes next

Microsoft has promised a full report. Run your own review now, while it is fresh. Record what stopped, for how long, and which workarounds held.

"Everyone needs a Plan B and it can't be hypothetical. It must be rehearsed and tested," said Richard Buckland of UNSW.

Priorities for leaders

  1. Know your critical business processes.

    Look past daily work to monthly, quarterly and annual cycles. Make that calendar visible to everyone.

  2. Map where your dependence sits.

    Trace each critical operation to the platforms behind it, from sign-in to security tools.

  3. Decide what must keep running.

    Set the minimum service you need, and how long you can go without it.

  4. Build real alternatives.

    Set up a backup channel outside your main provider, such as SMS or a phone tree. A backup on the same platform is not a backup.

  5. Rehearse a full provider failure.

    Test a scenario where one provider is down for a day. Include leaders, IT and key suppliers.

  6. Make it a board issue.

    Give the board clear ownership of concentration risk. A supplier's service level is not a continuity plan.

If your organisation needs help to be prepared for the next ‘systems down’ event, we can help. Get in touch for a free and confidential chat.

General information only.

Sources. ABC News, 13 November 2026. CIO Dive, 13 November 2026. iTnews, 14 November 2026. Forrester, 13 November 2026. Control Risks, 13 November 2026. APRA, Prudential Standard CPS 230 and Prudential Practice Guide CPG 230. Microsoft, July 2024. Optus network outage, 8 November 2023.

More insights

Privacy and AI

What 10 December means for automated decisions

From 10 December 2026, privacy policies must describe certain automated decision-making. The hard part is finding where those decisions already happen.

3 minute read Read article
View all insights
Previous
Previous

What 10 December means for automated decisions