On Tuesday, October 20, 2026, millions of Britons discovered just how much of everyday life runs through a handful of American tech companies. A power failure at one of Google's cloud data centre campuses in London knocked key services offline for hours — and the fallout reached deep into the UK's financial system, hitting digital bank Revolut, Lloyds, HSBC and others, alongside delivery firms, property portals, parts of the NHS and even the court system. The incident, trending per news coverage across the UK and beyond, is now shaping up as the first major real-world test of Britain's brand-new regime for overseeing the tech giants that banks depend on.
What happened on Tuesday morning
Google Cloud's status page shows the problems began at 0608 GMT — just after 7 a.m. local time in London — at its europe-west2 region, the company's name for its London-based cluster of data centres and availability zones that serves customers across Europe. The company warned of "elevated errors and intermittent network failures" across multiple services and apologised to those affected.
Google later blamed "a power event we had suffered" in part of the London area. By shortly after 1 p.m. local time the company said it was "seeing the recovery of services" across most of the region, and engineers said the power event had been "mitigated". Full service was restored later the same day, Google confirmed, after an hours-long disruption.
Banks bore the brunt
Revolut, the digital-only bank used by tens of millions of customers, was the hardest hit. Its status page showed card payments, top-ups by bank card, ATM withdrawals, currency exchange and stock trading as partially unavailable. A person familiar with the matter told Reuters that Revolut's outage was related to the Google Cloud disruption. The bank said it had been "working to fix the issue since it was identified this morning" and stressed that "all our customers' money and data are safe".
Other banks felt the shockwave too. Lloyds said its app was online and customers could log in normally, though some online services were misbehaving; the bank added that it was aware of similar issues hitting other organisations. Downdetector, the crowdsourced outage tracker, showed users reporting problems at HSBC, Barclays and Lloyds — although it was not immediately clear whether the Google Cloud outage was behind every one of those reports. Halifax was also named among affected firms.
The disruption spread well beyond banking. Parcel firm DPD, property website Rightmove, Royal Mail (which stressed that letter and parcel deliveries continued as normal), and NHS England services all saw problems reported. Even HM Courts & Tribunals Service reported issues with the cloud video platform used for some remote hearings, though it was unclear whether that was connected to the Google incident. Separately, ING Germany said customers had trouble logging into its online banking and app — an issue it attributed to a technical fault, with no confirmed link to London.
Why one power failure can knock out so many different companies
The outage is a vivid illustration of what computer scientists call the dependency effect of modern cloud ecosystems. Thousands of companies — including most major banks — rent their computing infrastructure from the same three or four providers: Amazon Web Services, Google Cloud and Microsoft Azure, plus a handful of others. When one region of one provider stumbles, the products and services built on top of it can all fail at once, in what looks like a cascade of unrelated outages.
Dr Amro Al-Said Ahmad, a lecturer in computer science at Keele University, told The Times the incident "highlights the dependency effect of modern cloud ecosystems", adding: "When a core cloud region experiences failures, the impact extends far beyond the directly affected services … products that depend on those services can all experience cascading failures."
Banks are among the heaviest cloud users in the economy. Core banking, mobile apps, card authorisation, fraud checks and trading platforms increasingly run on rented cloud capacity rather than in the banks' own data centres. That makes a cloud region failure an instant banking problem — exactly what customers saw on Tuesday.
Regulators were already worried — and now they are watching
The timing is uncomfortable for everyone involved, because the UK had only just put these providers under direct financial oversight. On July 13, 2026, HM Treasury designated four firms as the country's first "Critical Third Parties": Amazon Web Services EMEA SARL, Google Cloud EMEA Limited, Microsoft Ireland Operations Ltd and Oracle Corporation UK Limited. The move means the Bank of England, the Prudential Regulation Authority and the Financial Conduct Authority jointly supervise the resilience of the critical services those firms provide to the UK financial sector, under a framework established by the Financial Services and Markets Act 2023.
The reason for the regime is precisely Tuesday's scenario. The Treasury's designation order noted that, because so many firms rely on these services, "disruption or failure could affect multiple firms or markets at the same time, potentially impacting UK financial stability and services used by millions of consumers and businesses". More than 65% of UK firms rely on a small group of cloud providers for core operations, according to the Bank of England, and the top three providers accounted for 73% of all named providers in a 2024 survey by the Bank and the FCA.
FCA chief executive Nikhil Rathi has warned that "when the same providers serve thousands of firms, a single failure can reverberate across the financial system" — a sentence Tuesday's outage turned into a live demonstration. The FCA told media it was "aware of an incident affecting services provided by Google and are engaging with them and affected firms to understand the impact".
The episode is also drawing attention to the limits of the new regime. Analysts note that UK law gives the regulators no power to fine a Critical Third Party directly; sanctions run instead to restricting or prohibiting services under the Act, and regulated firms remain fully accountable for their own third-party risk. That means the hard questions land partly on the banks themselves: how much geographic redundancy do they buy, do they run across multiple cloud providers, and how fast can they fail over when one region goes dark?
A pattern of wake-up calls
Tuesday's outage sits in a growing catalogue of incidents that have pushed cloud concentration up the policy agenda. EU regulators last year designated Google Cloud among 19 critical third-party computing providers that the bloc's finance industry depends on, and a January 2026 UK Treasury Committee report called for major AI and cloud providers to be brought under the oversight regime by the end of 2026. Recent attacks that breached payment systems at around 20 Indian financial institutions showed the sector can also be hit deliberately; Tuesday showed it can fail by accident, from something as mundane as a power event in a London building.
There are parallels with other single-point-of-failure shocks. The Starlink outage that left Ukrainian forces exposed earlier this month prompted US senators to demand an investigation into over-dependence on one company's satellites — the same concentration logic, in orbit rather than in a data centre.
What customers should know
For ordinary customers, Tuesday's lesson is practical rather than alarming. Revolut and the affected banks all confirmed that customer money and data remained safe — the outage affected access to services, not the integrity of accounts. If a bank app misbehaves during such an event, checking the bank's status page or social media channels is more useful than repeatedly retrying a payment that may already be queued.
The bigger question is structural, and it will not be settled this week: whether banks, cloud providers and regulators can build enough redundancy into a system where, as one commentator put it, efficiency must not come at the expense of resilience. With cloud firms now under formal oversight, the next outage will be judged not just on how fast services recover, but on whether the new regime makes the system meaningfully safer — a test that has now, unintentionally, begun. Governments are meanwhile pouring money into domestic chip capacity, the physical foundation everything in the cloud ultimately runs on.
Sources: The Times, CNN, The Wall Street Journal, Reuters, Dow Jones Newswires/Morningstar, dpa-AFX via TradingView, Bank of England, AP News, Dr Amro Al-Said Ahmad (Keele University) via The Times, UK Financial Conduct Authority.