A power event at
Google Cloud’s London data centre on 20 October 2026 disrupted banking apps, delivery services and public services across the UK, after a power event hit one of the cloud provider’s London data centres shortly after 7:00am UK time
.
Revolut customers were unable to make card payments, withdraw cash or sign into the app, while Lloyds, Halifax, HSBC, Royal Mail and NHS England were among the other organisations affected
. The incident, which
lasted 7 hours and 52 minutes
, came just weeks after
Azure logged two separate outages that knocked gateway services offline across 18 regions and left Azure OpenAI customers in Sweden staring at failed API calls for nearly six hours between September 29 and October 1, 2026
. For Kent SMEs that have moved critical workloads to the cloud, these back-to-back failures expose a question many haven’t answered: what happens to your business when a single cloud provider goes dark?
Google Cloud London Outage 20 October 2026: What Actually Failed
Google Cloud confirmed the root cause in a status update, stating plainly that a power event had caused the failure
in the
europe-west2-b zone inside Google’s London region
.
Google’s status page reported elevated errors and intermittent network failures across multiple services in the wider europe-west2 region
, affecting:
- BigQuery Data Transfer Service, Cloud Filestore, Cloud Run, Google App Engine, Google Cloud NetApp Volumes, Google Kubernetes Engine, and Managed Service for Apache Kafka, all within the europe-west2 region
- Banking services including Revolut, which reported problems with card payments and cash withdrawals, and Lloyds and HSBC customers who also experienced difficulties
- HM Courts and Tribunals Service remote hearing platform, delivery and retail services including DPD, Royal Mail and Rightmove, and NHS England services
Guy Baybutt, of Veg Box Company, said: “Disruptions like this can have serious financial consequences. It was incredibly frustrating that a technical failure completely outside our control brought so much of our business to a standstill, affecting not just us but also the customers who rely on our deliveries”
.
The outage led to financial losses for small businesses unable to access or fulfil orders
.
Google restored full service availability by midday on 20 October 2026
.
Azure September 2026 Outages Hit 18 Regions in 40 Hours
The Google Cloud incident wasn’t isolated.
Between September 29 and October 1, 2026, Azure logged two separate outages that knocked gateway services offline across 18 regions and left Azure OpenAI customers in Sweden staring at failed API calls for nearly six hours
.
Microsoft advised users of its ExpressRoute Gateway, VPN Gateway, and Azure VMware service that they may experience degraded or interrupted network connectivity. This looks like a self-inflicted wound because Microsoft’s status page says it has “paused the infrastructure servicing activity associated with the onset of this event”
.
Secondary reporting based on Microsoft’s own incident notices put the affected footprint at 18 regions, spanning West US, West US 3, North Europe, West Europe, France Central, UK West, UK South, Switzerland North, Southeast Asia, East Asia, Japan West, Korea Central, South Africa North, UAE North, Mexico Central, Germany North, South India, and Jio India Central
.
The issue began when a platform-triggered gateway update resumed across multiple regions, increasing request volume to the regional gateway control plane service, thereby affecting its ability to process management operations reliably
.
What Cloud Concentration Risk Means for Kent SMEs
The power failure at a London data centre on Tuesday 20 October 2026 disrupted UK banking, business and public services, raising concerns about Britain’s dependence on a handful of major cloud providers and the damage an outage can do to local businesses and organisations
.
The disruption is already being cited as one of the clearest examples yet of how concentrated a single cloud provider’s regional footprint has become across UK finance and public services
.
A single power fault in one London building took banking apps, courier services and court video links offline for most of a working morning. The Google Cloud outage of 20 October 2026 is now the UK’s clearest recent case study in cloud concentration risk
.
For Kent SMEs, the practical impact is immediate. If your Microsoft 365 email, CRM, accounting software, e-commerce platform or backup storage all depend on a single cloud provider in a single region, you have a single point of failure. When that provider experiences an outage—whether from a power event like Google Cloud’s London incident, a maintenance error like Azure’s gateway failure, or infrastructure failure—you’re offline until the provider restores service. You don’t control the timeline, and you can’t accelerate recovery.
Azure and Microsoft 365 Dependencies for Kent Businesses
Many Kent SMEs run their entire IT stack on Microsoft platforms. That includes Microsoft 365 for email and collaboration, Azure Active Directory for identity, Azure-hosted virtual machines for line-of-business applications, and third-party SaaS products that themselves depend on Azure infrastructure. When
the Azure event started at 20:30 UTC on 30 September 2026
and affected UK South and UK West regions, businesses in those availability zones lost access to VPN gateways, ExpressRoute connections and virtualised VMware workloads.
If your remote workers depend on Azure VPN Gateway to access on-premises file servers, or your website runs on Azure App Service in UK South, you were offline. The same applies to cloud backup solutions hosted exclusively in a single Azure region. We covered related authentication issues in our post on Windows 11 October 2026 NTLMv1 blocks, but the outage problem is architectural, not protocol-level.
What Kent SMEs Must Do About Cloud Concentration Risk Now
1. Map Your Cloud Dependencies and Single Points of Failure
Start with an inventory. List every cloud service your business depends on to operate—email, CRM, accounting, backups, website hosting, payment processing, remote access—and identify which cloud provider hosts each one. Then identify which region hosts that service. If multiple critical services sit in a single provider’s single region, you have concentration risk. If your primary and backup systems both depend on the same provider (for example, Microsoft 365 email plus Azure-hosted backup), you have correlated risk: both fail together during a provider-level outage.
2. Implement Multi-Region Resilience Where Business-Critical
For workloads you control—such as Azure virtual machines, web applications or databases—configure multi-region replication or failover. Azure offers availability zones and paired regions specifically to guard against single-zone or single-region failures. Google Cloud and AWS offer equivalent architectures. This doesn’t eliminate all risk, but it reduces exposure to localised power events or infrastructure failures in a single data centre or availability zone.
Multi-region failover adds cost and complexity, so prioritise based on business impact. If losing access to your CRM for eight hours costs you thousands in lost sales or damages client relationships, architect for resilience. If a service can tolerate downtime, accept the risk and document the decision.
3. Diversify Critical Services Across Providers
Where practical, avoid single-provider dependency for your most critical services. If your email runs on Microsoft 365 and your backup runs on Azure, consider a secondary backup provider on AWS or a dedicated backup service with its own infrastructure. If your e-commerce platform runs on Google Cloud, ensure your payment gateway and customer communication tools don’t share the same infrastructure. This won’t prevent every outage, but it reduces the chance that a single provider failure takes down your entire operation simultaneously.
Diversification has trade-offs. Managing multiple providers increases administrative overhead and can raise costs. But for businesses where eight hours of downtime triggers contract penalties, lost revenue or reputational damage—like the financial losses reported by
small businesses unable to access or fulfil orders during the Google Cloud outage
—the investment is justified.
4. Test Failover and Document Recovery Procedures
Resilient architecture only works if you know how to activate it under pressure. Test your failover procedures at least quarterly. If you’ve configured Azure Traffic Manager to redirect traffic from UK South to North Europe during an outage, verify it works. If you maintain a secondary email relay or backup restore pathway, confirm your team knows how to trigger it and how long restoration takes.
Document every step in a runbook that non-specialists can follow. During an outage, the person who configured your infrastructure may be unavailable. Your operations team needs clear, tested instructions they can execute without guesswork. We’ve discussed similar business continuity planning in our article on Q3 2026 ransomware defences, and the same discipline applies to cloud resilience.
5. Review SLAs and Understand What You’re Actually Guaranteed
Cloud providers publish service level agreements that promise specific uptime percentages—typically 99.9% or 99.95% for enterprise tiers. Read the fine print. Those SLAs usually cover the provider’s infrastructure, not your application’s availability, and they often exclude downtime caused by factors outside the provider’s control or maintenance windows. Compensation for SLA breaches is typically limited to service credits, not reimbursement for your lost revenue, contract penalties or reputational damage.
If your business requires guaranteed uptime beyond what a single provider’s SLA offers, you need redundancy at the architecture level. That means multi-region failover, provider diversification, or hybrid infrastructure that keeps critical services on-premises with cloud used for secondary workloads.
6. Implement Local or Hybrid Backup for Business-Critical Data
Even if your primary systems run entirely in the cloud, maintain a local or geographically separate backup of your most critical data.
The UK Prime Minister’s office confirmed it was in contact with Google Cloud as services were restored
, which underscores the scale and national impact of these outages. If a cloud provider experiences a prolonged or catastrophic failure, you need the ability to restore core business operations independently.
For most Kent SMEs, this means at least a nightly export of customer records, financial data and essential documents to local storage or a secondary cloud provider. We covered cloud backup adoption in our post on 68% of UK SMEs now using cloud backup in 2026, but the October outages show that cloud-to-cloud or cloud-plus-local hybrid strategies offer stronger resilience than single-provider cloud-only approaches.
Practical Next Steps for Saltwood and Kent SMEs
Cloud outages are not theoretical.
On 8 October, twelve days before the outage, the Bank-backed CMORG group finished SIMEX26, a sector-wide exercise on a global cloud services disruption. It involved the 38 largest and most systemically important banks and market infrastructure operators. The Bank said it expects firms to act on the lessons
. If the UK’s largest financial institutions are rehearsing cloud failure scenarios, smaller businesses should be doing the same.
Start this week:
- Audit which cloud providers and regions host your email, backups, CRM, accounting, remote access and website
- Identify where multiple critical services share a single provider or region
- Calculate the cost of eight hours’ downtime for each service—lost sales, contract penalties, staff productivity, reputational impact
- Prioritise resilience investment based on business impact, starting with revenue-critical and compliance-sensitive workloads
- Document and test failover procedures for any multi-region or multi-provider configurations you implement
- Review your DR plan to ensure it addresses provider-level outages, not just ransomware or hardware failure
The October 2026 outages demonstrated that even the largest cloud providers experience significant regional failures. Your business continuity plan must account for that reality. If you haven’t stress-tested your cloud dependencies or documented how your business continues operating when your primary provider goes offline, you’re carrying unquantified risk.
Meridian Micro Limited works with SMEs across Kent and the South East to design resilient cloud and hybrid IT architectures that balance cost, complexity and business continuity requirements. If the Google Cloud and Azure outages have exposed gaps in your resilience planning, or if you’re unsure whether your current cloud setup can withstand a provider-level failure, call us on 01303 883111 to discuss your specific infrastructure and business continuity needs. We’ll help you identify single points of failure, evaluate multi-region or multi-provider options, and implement tested failover procedures that keep your business operating when cloud providers can’t.