What the Google Cloud outage taught us about fintech risk

TL;DR: A power event in Google Cloud's europe-west2 London region on 20 October 2026 took down app access at Revolut, Lloyds, Halifax and HSBC for several hours. Customer money and data were unaffected, and services were restored the same morning. The real takeaway for a Student Investor isn't the outage itself — it's that a company's share price depends on infrastructure you can't see on its balance sheet.

What actually happened

Just after 6am UK time, Google's own status pages flagged a problem in the europe-west2-b availability zone — a specific physical data centre cluster in London. Google attributed it to a power event. Because that one facility hosts backend services for a cluster of UK financial apps, the failure didn't stay contained to Google's own products. Within minutes, customers of the digital bank Revolut, and of high-street names Lloyds and Halifax, started reporting they couldn't log in, move money, or withdraw cash. HSBC also acknowledged difficulties. The courier firm DPD and a UK court video-hearing platform were hit too, which tells you how far a single cloud region's footprint actually reaches.

Revolut said in a customer email that it was "currently experiencing a service disruption affecting multiple services" and stressed that "all our customers' money and data are safe." A Lloyds spokesperson later said the underlying issue had been resolved and services were returning to normal. By mid-morning, most apps were functioning again.

Why one power failure can take down four banks at once

Here's the part worth sitting with. Revolut, Lloyds and HSBC are fierce competitors. On paper, if one of them has a bad day, that's its own problem — not the others'. But modern banks and fintech apps don't build and run every piece of their own technology. Many rent computing power, storage and networking from a small number of huge cloud providers — Google, Amazon and Microsoft are the three biggest. It's efficient: nobody has to build their own data centres from scratch. The trade-off is that if the provider has a bad day, every customer renting from that provider can have a bad day too, at the same time, for the same reason.

That's a different kind of risk from the ones you usually meet first in investing — a bad set of results, a profit warning, a shift in interest rates (see what interest rates do to shares). It's called operational risk, and this particular flavour of it — many companies depending on the same small set of outside vendors — is sometimes called concentration risk. You can't see it in a quarterly earnings report. You only really notice it on the one morning it goes wrong.

Does an outage like this actually move share prices?

Usually, not by much, and not for long — if the company recovers quickly and no money is actually lost, as happened here. A few hours of app downtime, fixed the same day with a clear "your money is safe" statement, tends to be a reputational story more than a financial one. Where it gets more serious is if an outage repeats, drags on for days, or coincides with an actual data or funds problem. That's when it starts showing up in the numbers that matter to investors: customer complaints, regulatory scrutiny, and in the worst cases, customers switching providers. The lesson isn't "panic at every outage headline" — it's closer to what we cover in risk and reward: every extra layer a business depends on is one more thing that can go wrong, and part of judging a company is judging how well it's built for the bad days, not just the good ones.

The diversification lesson hiding inside this story

New investors learn early that spreading money across different companies and sectors reduces risk — that's the whole idea behind diversification. This outage shows a less obvious version of the same principle, aimed at companies rather than portfolios: a business that leans on just one outside supplier for something critical is less diversified than it looks, even if its own balance sheet is healthy. Banks have started responding to exactly this risk by spreading workloads across more than one cloud region, or more than one provider, so a single power failure can't take the whole service down. It costs more to run two systems instead of one — but it buys resilience.

That resilience doesn't happen by accident. For banks and fintech firms, preventing a cascading failure like 20 October's usually comes down to decisions made years earlier, at the point the product was actually built: how data is replicated across regions, how the app fails over when one piece of infrastructure goes dark, and how quickly engineers can tell customers what's actually going on. Teams offering dedicated SaaS development services design that kind of multi-region redundancy into financial products from day one, rather than bolting it on after the first bad headline.

How to read the next cloud-outage headline

1. Check whether it's an access problem or a money problem. "I couldn't log in for two hours" is very different from "funds are missing." Read the company's own statement carefully — and treat a vague one as a bigger red flag than a specific one.
2. Note how fast it was fixed. Same-day recovery with a clear explanation is routine in modern tech operations. Days of silence or repeat outages are a worse signal.
3. Ask who else was affected. If several unrelated companies went down at the same time, the story is almost always about shared infrastructure, not about any one company's own mistakes.
4. Don't trade on the headline alone. One morning of downtime, on its own, rarely tells you much about a company's long-term prospects. It's one more data point to fold into the bigger picture, the way we describe in how the news actually moves a share price.

The takeaway

A power failure in one London data centre on 20 October 2026 was enough to lock millions of people out of their banking apps for a morning — a vivid reminder that the companies you follow in the Challenge don't operate in isolation. They rent critical infrastructure from a handful of giant providers, and that convenience comes with a shared-risk price tag. No money was lost this time, and the apps were back within hours. The useful habit to take from it isn't fear of the next outage — it's a sharper question to ask about any company you research: what does this business depend on that it doesn't control, and how well has it planned for the day that dependency fails?

This article is educational and is not financial advice. Details of the 20 October 2026 outage are as reported by The Irish Times and ABC News (Associated Press), with technical detail from The Register.