Why a single tech outage can bring the modern workplace to a halt
Analysis: Digital transformation has made our workplaces more fragile and dependant on a range of connected services, tools and ecosystems Resilience is a funny thing. When Microsoft's services failed last week, a financial firm in Taiwan confirmed trades by fax, UK NHS staff fell back on paper records and insurer Aviva distributed alternative phone numbers and even used Bluetooth to transfer recordings. These organisations weren’t resisting the digital age, but trying to keep working when the technology they relied on stopped cooperating. The Microsoft outage began on Thursday 12 November, when changes to its network exposed a software fault that disrupted Exchange Online, the service supporting Outlook's email and calendar functions. Microsoft confirmed that the incident was a technical failure rather than a cyberattack or an AI-related problem. We need your consent to load this rte-player content We use rte-player to manage extra content that can set cookies on your device and collect data about your activity. Please review their details and accept them to load the content. Manage Preferences From RTÉ News, Microsoft says last week's global outage caused by network capacity upgrade So has digital transformation made workplaces more fragile? In one specific way, yes. We have swapped visible inefficiency for invisible dependency. Work once done by people now runs through chains of connected services, and few organisations have mapped where those chains break. Over the past decade, businesses have increasingly connected their everyday activities through different applications. An ordinary office might use Outlook for email, Teams for meetings, a separate platform for customer enquiries and another system for managing appointments. Although these appear to be independent tools, they frequently rely on one another to complete routine tasks. Consider what happens when a customer emails a business. Support software retrieves the message from Outlook, creates a ticket and assigns it. The employee books an appointment, which triggers a calendar invitation and an automated reminder. Several applications are involved, but almost nobody touches the process. When Exchange Online failed, those connected applications couldn’t complete their tasks even though the software itself still worked. Enquiries went unprocessed, appointments weren’t created and reminders never arrived. We need your consent to load this rte-player content We use rte-player to manage extra content that can set cookies on your device and collect data about your activity. Please review their details and accept them to load the content. Manage Preferences From RTÉ Radio 1's Drivetime, Elaine Burke from the For Tech's Sake podcast and Prof Alan Smeaton from DCU on Microsoft's offer of buyouts t o 60,000 employees to cut costs Customer-support company Pylon experienced this during the outage, reporting delays and failures in processing Outlook email and calendar events. Even after Microsoft began restoring its services, Pylon's engineers had to recover events missed during the disruption. Restoring the underlying service therefore does not necessarily resolve the consequences and organisations may still need to identify missed enquiries, complete interrupted tasks and reconcile actions employees carried out manually while systems were unavailable. Technical recovery and business recovery are not always the same thing, and two features explain this. The first is tight coupling : a fault spreads faster than anyone can intervene when one service feeds another automatically. The second is concentration: a handful of providers underpin much of the economy, so one fault becomes everyone’s fault, as with CrowdStrike in 2024 . In fairness, cloud providers often deliver reliability that most firms could never build themselves. The problem isn’t the cloud. It’s assuming it will never fail. The timing was particularly striking. Just one day before Microsoft's outage, the UK's Financial Conduct Authority published a review questioning whether financial firms could demonstrate their ability to withstand serious technological disruption, rather than simply pointing to planned improvements. The following day offered a real-world test. The Microsoft outage quickly became a showcase for organisations that had prepared for disruption and invested in resilience long before their systems failed. We need your consent to load this rte-player content We use rte-player to manage extra content that can set cookies on your device and collect data about your activity. Please review their details and accept them to load the content. Manage Preferences From RTÉ Radio 1's News at One, Tadhg Crowley from the Irish Medical Organisation on how health services are dealing with problems caused by the Microsoft outage Reuters reported that European stock-exchange operator Euronext directed listed companies to use the telephone, while maintaining procedures to verify callers' identities. Elsewhere, Aviva had rehearsed a major technology outage just two weeks earlier and was able to activate its contingency arrangements when the disruption arrived. These challenges are increasingly reflected in financial regulation. Under the EU's Digital Operational Resilience Act , covered financial firms must understand their technological dependencies , test their preparedness and manage risks associated with external providers. UK regulators also require certain financial firms to establish impact tolerances : limits on how much disruption their most important services can withstand before unacceptable harm occurs. These can be expressed through measures such as the duration of an outage, the number of customers affected or the volume of disrupted transactions. Work once done by people now runs through chains of connected services, and few organisations have mapped where those chains break. The principles reach far beyond finance. A retailer needs to know whether it can still take orders and a clinic needs to know how to reach patients when automated reminders fail. Both should rehearse. Netflix famously built Chaos Monkey , a tool that randomly shuts down parts of its own live systems so engineers learn how they fail before customers do. Most organisations needn’t go that far, but removing an essential service for a day and watching whether staff can keep working securely is a start. Agree alternative channels, decide who can authorise emergency procedures, and reconcile interrupted work afterwards. Digital transformation has made routine work faster and more efficient, but it can also make interruptions more costly. The next time an organisation replaces a manual process with automation or connects another system to its digital infrastructure, it should ask what happens when that connection fails. If nobody can complete the work without it, that efficiency gain has introduced a dependency the organisation may never have properly considered. Follow RTÉ Brainstorm on WhatsApp and Instagram for more stories and updates The views expressed here are those of the author and do not represent or reflect the views of RTÉ