When systems go down: putting operational resilience to the test

A technology outage may begin with losing access to a key piece of software. But it soon becomes a test of operational resilience: what can the business still do, and how will people coordinate without their usual tools?

November’s Microsoft 365 outage brought those questions into focus. Disruption affecting Outlook and Teams calendars  left organisations struggling to communicate. The technical failure quickly became a business problem: how could people serve customers, coordinate work and make decisions without the platforms they usually rely on? 

No organisation can prevent every outage. But it can avoid discovering, in the middle of one, that nobody has agreed which activities matter most or how to keep them running. 

Preparing through business continuity planning 

That agreement is the starting point for minimum viable business planning. Rather than asking each department to defend its own priorities, leaders need to consider what the organisation needs to continue or resume operations in a worst-case scenario. Just as importantly, they need to agree what can temporarily stop. 

Minimum viable business planning starts with the organisation, not individual departments. If everything became unavailable, what would we need to restore first, and what could wait?

Matt Lomax Associate Director - Data and Digital Advisory

Those priorities depend on people, applications, communications, data and suppliers, as well as increasingly connected equipment. Cyberattacks affecting supermarket refrigeration systems in October and early November spoiled stock across Europe, demonstrating how a cyber incident can disrupt physical operations. Mapping these connections helps organisations identify where a single failure could threaten an essential service. 

A Google Cloud disruption in London on 20 October highlighted another dependency. A power failure at one facility interrupted workloads and put pressure on other zones as services moved across. For customers, the practical lesson is that using cloud infrastructure does not, by itself, guarantee that recovery arrangements will work as expected. 

Contracts do not guarantee availability or transfer accountability. Recovery commitments, incident notification and exit provisions should sit alongside practical alternatives for responding to incidents. Continuity arrangements also need to keep pace as technology and supplier dependencies change. 

Even a well-considered plan remains an assumption until it is tested in a representative scenario. Business continuity testing uses realistic exercises to show whether contacts are accessible, alternative arrangements work and services can recover when a key person or familiar tool is unavailable. 

Bringing leaders and operational teams into those exercises can expose gaps that a document review misses, including what happens when several services need to be restored at the same time. 

The test is not whether you have a continuity plan. It is whether people can use it when a key person, system or communication channel is unavailable.

Matt Lomax Associate Director - Data and Digital Advisory

For boards and executives, the findings can show whether recovery priorities and acceptable levels of disruption are realistic, and where responsibilities or resources need to change. 

Acting on those findings often requires more than updating a policy. For many organisations, agreeing which services can be suspended, testing what is genuinely critical and putting effective recovery arrangements in place can involve significant rework, investment and several months of sustained effort. Plans then need to be revisited as technology, suppliers and threats change, while employees need to understand when to stop an activity rather than improvise an unsafe workaround 

When disruption arrives 

Preparation will not remove uncertainty when an incident begins. A technology failure may call for recovery, while a suspected cyber incident may require services to be isolated first. Cyber resilience means being prepared for both: leaders can face difficult decisions about stopping a service before they fully understand what has happened, while also considering how it can later be restored safely. 

Established incident procedures bring IT, operations, risk, communications and relevant leaders together around those decisions. The immediate goal is to establish what is known, understand the business impact and assess the available options, rather than wait for certainty about every detail.

In the first hour, you may not have all the answers. What matters is understanding what you know, the impact on the business and the options available to you

Matt Lomax Associate Director - Data and Digital Advisory

This is where earlier decisions about the minimum viable business earn their value. Instead of debating departmental priorities during an incident, the response team can direct resources towards agreed critical services and pause less urgent work. 

But those decisions still depend on people being able to communicate. If email and instant messaging apps are unavailable at the same time, accessible contact details and approved independent channels, such as telephone or SMS, become essential. Everyone involved also needs to understand how to use those channels securely. 

Employees, customers and suppliers need a clear account of what is affected, what alternatives are available and when they will receive another update. Clear communication helps maintain trust while the response continues. 

Restoring access is not the end of recovery. Approvals and transaction records created during manual work need to support subsequent reconciliation, alongside checks that critical systems and integrations are operating correctly. 

Operational resilience is a business priority 

Ultimately, technology resilience is not just about restoring systems. It is demonstrated by the choices an organisation can make when its usual ways of working disappear. 

Knowing what matters most, understanding the dependencies around it and practising how to protect or restore it gives leaders a firmer basis for making those choices when disruption arrives. 

 

Get in touch 

To discuss your organisation’s approach to operational resilience, from defining your minimum viable business to testing your response to disruption, get in touch with our technology and cyber security experts. 

Contact us

Key contacts