A power-related incident in Google Cloud's London infrastructure disrupted Revolut and a range of other UK services on October 20, leaving some customers unable to use online banking and businesses unable to process orders. Google Cloud said later that day that the disruption had been resolved.
The provider traced the problem to a third-party facility supporting its europe-west2 region. During the incident, its status updates reported increased error rates and intermittent network failures across multiple services. Google subsequently narrowed the origin to a power event in one of the region's availability zones, known as europe-west2-b, and said it would publish a fuller analysis on its service-health dashboard.
Revolut was among the most visibly affected companies. More than 1,500 user reports were recorded by an outage-monitoring service shortly after 08:00 BST, according to the BBC. Customers described being unable to make payments, view incoming transfers or access money while away from home. Revolut said the interruption stemmed from the Google Cloud outage, that its services had returned to normal, and that customer funds and data remained safe throughout.
Other financial institutions also experienced trouble. Customers of Lloyds Bank, Halifax and Bank of Scotland reported problems with some online services. Lloyds Banking Group did not confirm a cause in the BBC report, but said the underlying issues appeared to have been resolved and that it expected its services to recover. HSBC was also identified among the services confirming disruption.
The impact extended beyond banking. Users flagged problems involving Royal Mail, parcel company DPD and e-bike operator Forest. HM Courts and Tribunals Service reported an issue with the video system used for remote hearings. A fresh-produce seller told the BBC that it could not download website orders or book parcels for collection, creating potential delays and lost sales involving perishable goods.
Google Cloud is one of a small group of providers that supply computing infrastructure to many organizations from shared regional facilities. This incident showed how a localized failure can become visible across unrelated consumer and public-sector services when they depend on the same underlying region. It also illustrated the difference between restoration and explanation: Google reported that systems were back, while its promised incident analysis was still needed to describe the failure and the safeguards involved.
For affected users, the immediate outage ended on October 20. The broader operational question for customers is whether critical services have practical alternatives when a cloud region or one of its component facilities becomes unavailable.
