Is Your Business Ready for the Next Cloud Outage? The Communications Commissioner Explains

Header Image

From emergency accounts to the 3-2-1-1 backup rule, Pieris explains how firms can keep working when the cloud fails.

When Outlook, Exchange Online, Teams and Copilot were disrupted between 12 and 14 November 2026, organisations in more than 70 countries lost access to their everyday tools. For the Commissioner of Electronic Communications and Postal Regulation, Marios Pieri, the lesson is that relying on a single service provider is an operational risk, however large and well known that provider may be.

To help businesses prepare, the commissioner sets out nine practical prevention and preparedness measures. They cover everything from mapping hidden dependencies to testing backups and monitoring services independently.

What the November outage revealed

What did the Microsoft 365 disruption show organisations?

The outage affected organisations in more than 70 countries. It showed that depending on one service provider is an operational risk, however large and well known that provider is.

Mapping critical services and dependencies

Where should a business begin?

By recording its critical services and dependencies. Businesses should identify which functions depend on the cloud, including “hidden” dependencies such as authentication or third-party applications that send email.

They should then assess the impact of outages of different lengths, for example 4, 8, 24 and 48 hours. For each function, they should set a maximum acceptable downtime (RTO) and a maximum acceptable data loss (RPO).

Alternative communication channels

How should staff stay in touch if the main platforms go down?

Businesses should keep communication channels that do not depend on the same provider. These include SMS, mobile telephony, or a second messaging or videoconferencing platform. The channels should cover staff as well as partners and strategic associates.

An up-to-date contact list kept outside the cloud, in both local and printed form, should be available for emergencies.

Emergency access and authentication

What happens if employees cannot log in?

Businesses should create emergency accounts that are exempt from the usual access policies. These accounts should be protected with physical security keys and stored securely. Local administrator accounts should also be kept for critical systems.

Operating in a limited mode

Can a business keep working without its cloud services?

Operations should be designed so that core functions continue when external services are unavailable. That means offline working, local temporary storage with later synchronisation, and documented manual procedures.

Independent and local backups

What does a sound backup strategy look like?

Businesses should apply the 3-2-1-1 rule: three copies, on two different media, one stored off-site and one that is immutable or kept offline.

A local copy, for example on a NAS or a local server, allows immediate recovery without a cloud connection. Local synchronisation of critical folders should also be enabled on workstations.

Email, OneDrive, SharePoint and Teams data should be backed up with a third-party tool. Under the shared responsibility model, the provider ensures that the service is available, not that the customer’s own data can be recovered.

Businesses should also keep copies of their infrastructure, network and application settings, including DNS zones, firewalls and tenant policies. Backups should be encrypted, and their recovery should be tested regularly.

Application resilience

What about applications connected to the cloud?

Applications that connect to the cloud must handle outages without losing data or transactions.

Reducing dependence on a single provider

Should businesses spread their services across providers?

For critical functions such as email, businesses should use a continuity service or an alternative provider. Where justified, they should adopt multiple cloud providers or hybrid infrastructure. They should favour open standards so that switching is easier, and they should review the terms of their service level agreements (SLAs).

Response plans and regular testing

How often should preparedness be tested?

Businesses should draw up a response plan setting out roles, activation criteria and ready-made update messages. The plan should also be available in printed form.

Business continuity exercises should take place at least once a year, with a review after every incident.

Independent monitoring

How can a business tell that an outage is under way?

It should use monitoring tools from a different provider. These should run synthetic checks, such as login, authentication and sending email, and send alerts by SMS.

Businesses should not rely only on the provider’s status page, which is often updated late.