When the Cloud Goes Dark: Why Every Business Needs a Plan B
The November 2026 Microsoft outage offers a reminder that even the world’s largest technology platforms can fail. For businesses that depend on digital systems, the important question is not whether an outage will happen, but whether work can continue when it does.
When the working day suddenly stops
On 12 November 2026, businesses around the world experienced a familiar but increasingly disruptive problem: the technology they relied on was suddenly unavailable.
A major Microsoft 365 outage affected access to services including Outlook, Exchange Online, Teams and other connected applications. Some users could not access their email, while others encountered problems with messaging, cloud-based services and business applications.
Microsoft traced the disruption to infrastructure problems that left servers unexpectedly unavailable, placing additional pressure on those still operating. Recovery efforts continued for hours.
Although the incident was widely reported internationally, the extent of its impact on South African businesses was not immediately clear. That uncertainty does not make the event any less relevant locally.
South African businesses rely on many of the same cloud platforms, software subscriptions and digital communication systems as their international counterparts. The question is what happens when one of those systems becomes unavailable.
Convenience has created dependency
Cloud computing has transformed the way businesses operate. Documents can be accessed from multiple locations. Employees can collaborate without sharing physical offices. Software updates happen automatically, and companies no longer need to maintain extensive computer infrastructure of their own.
For smaller businesses, these advantages can be particularly valuable. But convenience has introduced a different kind of dependency. A company may own its computers, machinery and premises, yet depend on external services to access customer correspondence, retrieve documents, authenticate software or manage its daily operations.
The equipment is still there. The employees are still there. The electricity may even be working. But a critical part of the business is somewhere else. And when that part becomes unavailable, the consequences can extend well beyond the IT department.
When the machines work but the business doesn’t
Consider a small manufacturing workshop. Its laser cutter, CNC router or printing equipment may be perfectly operational. There is no mechanical breakdown, no damaged component and no interruption to the electrical supply. Yet production can still be affected.
Perhaps the latest customer artwork is stored exclusively in a cloud folder. Perhaps the job instructions are attached to an inaccessible email. Perhaps the production schedule is maintained in an online application, or the software needed to prepare a job requires an internet connection to verify its licence.
None of these problems involves the manufacturing equipment itself. Nevertheless, each can interrupt the production process. This illustrates an important distinction between machine availability and operational availability. A machine can be ready to work while the business systems surrounding it are not.
As workshops become more digitally connected, understanding these dependencies becomes increasingly important.
Cloud storage is not necessarily a backup
One of the most common misunderstandings concerns cloud storage.
Services that synchronise files between computers and online accounts are extremely convenient. They can also provide useful protection against certain types of hardware failure. But synchronisation and backup are not the same thing.
If a file is accidentally deleted, corrupted or overwritten, those changes may be synchronised across connected devices. Version history and recovery features can help, but they depend on the service’s capabilities, retention settings and accessibility.
An independent backup serves a different purpose: preserving a recoverable copy of important information. For a manufacturing business, that information might include design files, machine settings, customer specifications, job records and software configurations.
A practical backup arrangement should consider not only whether the files exist, but whether they can be recovered when the normal systems are unavailable. An inaccessible backup is of little immediate use when a customer is waiting for an urgent order.
The overlooked problem of software access
Keeping copies of important files is only part of the solution. Some applications require online authentication, subscription verification or access to a remote server before they can operate. Others function offline but lose certain features when disconnected.
A business might therefore have all its design files safely stored on a local computer, yet still be unable to open or process them using its normal software. This is why business continuity planning should include software as well as data.
Which applications can operate without an internet connection? Which require periodic licence verification? Can existing jobs still be opened and completed? Are alternative workflows available? These are questions best answered during an ordinary working day, rather than during an outage.
What does a sensible Plan B look like?
Preparing for digital disruption does not necessarily require expensive infrastructure or an elaborate disaster recovery system. For a small business, the starting point is understanding which activities are essential and what they depend on.
A workshop might decide to maintain local copies of current production files, preserve independent backups of critical designs and configurations, and document how urgent jobs can continue when online systems are unavailable.
Alternative communication arrangements may also be useful. If business email becomes inaccessible, customers and suppliers should still have a practical way to make contact. Equally important is testing.
A backup that has never been restored is an assumption, not a demonstrated recovery capability. And an offline workflow that nobody knows how to use offers limited protection.
The objective is not to reproduce every cloud service locally. It is to identify the few functions that genuinely matter when normal operations are interrupted.
Reliability is not the same as availability
Major cloud platforms invest enormous resources in infrastructure reliability, security and redundancy.
For most small businesses, those platforms can provide capabilities that would be difficult or prohibitively expensive to reproduce independently. But even highly reliable services experience interruptions.
A service offering 99.9% availability, for example, could still experience approximately 8.8 hours of downtime over a year, depending on how availability is measured.
That may sound insignificant until those hours coincide with a production deadline, an important customer order or the end of a working day. The practical lesson is not that cloud services are unreliable.
It is that no single system should be assumed to be continuously available simply because it usually works.
A South African perspective
South African businesses already understand the importance of operational resilience. Years of electricity interruptions have encouraged many companies to invest in backup power, alternative energy supplies and equipment protection.
The underlying principle is familiar: when an essential service becomes unavailable, the business needs a way to continue. Digital infrastructure deserves similar consideration. An inverter may keep the computers running during a power failure, but it cannot restore an unavailable cloud application.
A generator may keep the machines operating, but it cannot retrieve a design file that nobody can access. Resilience increasingly requires thinking beyond the physical equipment.
The lesson from November 2026
The Microsoft outage will eventually be resolved, investigated and added to the long history of technology incidents.
Most users will return to their normal routines, and the convenience of cloud computing will quickly make the disruption seem like a distant inconvenience. But businesses can take something useful from the experience.
The next interruption may involve a cloud platform, an internet connection, a software licence server or a local network. Its cause may be entirely outside the company’s control. Its consequences need not be.
For businesses that depend on digital technology, the most useful question is a simple one:
If our essential online services stopped working tomorrow, what could we still do?
The best time to discover the answer is while everything is working.
Editorial sources: Reuters, 12 November 2026 · BleepingComputer, 12 November 2026 · MyBroadband, 13 November 2026
