When the Cloud Goes Down: What Every Business Owner Should Know

Your website is working perfectly. Your customers can browse your products, book appointments, and place orders without a hitch. And then, at 11:20 on a Tuesday morning, it just stops.

Not because of anything you did. Not because of a bug in your code. Because a company you've probably never heard of made a change to one of their internal databases, and that change knocked your site offline along with millions of others.

This is the reality of running a business online in 2025 and 2026. And it happened to some very large companies, very recently.

The Outages That Actually Happened

Over the past few months, three major cloud infrastructure providers went down in ways that affected businesses all over the world.

Amazon Web Services — October 20, 2025

AWS is the backbone of the internet. Netflix, Airbnb, Slack, Spotify — most of the apps and websites you use every day run on Amazon's servers. On October 20, a rare timing conflict between two automated systems caused a critical DNS record to be accidentally deleted. A DNS record is essentially the address book entry that tells the internet where to find a service. Without it, nothing works.

The outage lasted over 15 hours. 113 different services were affected. More than 11 million outage reports flooded in globally. Brands like Delta, Uber, Starbucks, and Amazon's own services like Alexa and Prime Video went dark. The estimated economic impact ran into the hundreds of billions of dollars.

Fifteen hours. For a bug that was essentially a typo in an automated script.

Cloudflare — November 18, 2025

Cloudflare is the company that sits in front of a huge portion of the internet, handling security, performance, and delivery for millions of websites. When Cloudflare has problems, the web notices.

On November 18, a database permissions change caused an internal configuration file to double in size. That bloated file crashed a core part of Cloudflare's system when it was distributed across their network. Their CDN, their Workers platform, their security tools — all of it went down. The outage lasted nearly four hours and was described as their worst since 2019.

Supabase — February 12, 2026

Supabase is a popular backend platform used by developers to store data and handle user authentication. Last week, an engineer at Supabase deployed a new internal monitoring service. That deployment accidentally triggered a cloud security setting that blocked all internet traffic to an entire region.

Databases, login systems, APIs — all offline for nearly four hours. Not because of a hacker. Not because of a hardware failure. Because of one internal deployment that enabled the wrong setting.

Why This Actually Matters to Your Business

Here's what these outages have in common: they weren't caused by bad actors. They weren't caused by obviously negligent companies. They were caused by the everyday reality of running complex software at massive scale. Smart engineers at well-funded companies made changes that had unexpected consequences.

If your business depends on your website to take orders, book clients, or generate leads, an outage isn't just an inconvenience. Every hour offline is revenue you're not making. Customers who land on a broken site don't wait around — they go to a competitor. And if it happens more than once, they stop trusting that you'll be there when they need you.

The cost isn't just the downtime itself. It's the customers who quietly never come back.

The Uncomfortable Truth About "The Cloud"

Here's the thing: there's no such thing as a perfectly reliable website. Every website depends on layers of infrastructure — hosting providers, database services, content delivery networks, authentication platforms — and any one of those layers can fail.

Even companies that do everything right can be taken down by a vendor they rely on. Your developer could build you a flawless website, host it on the best infrastructure available, and it could still go offline because Cloudflare or AWS had a bad Tuesday.

That doesn't mean there's nothing you can do. It means knowing the risks and making smart decisions about them.

What You Can Actually Do

You're not going to prevent AWS from having outages. But you can make smart choices that limit how much those outages hurt your business.

1. Know what you're running on — and who you can call

Most business owners don't know what infrastructure their website runs on. That's fine until something breaks. Ask your developer or agency to give you a simple breakdown: where is the site hosted, what services does it depend on, and who do you contact if something goes wrong? Having that list before an incident is worth a lot more than scrambling for it after.

2. Subscribe to your provider's status pages

Every major infrastructure provider has a public status page where they post outage notifications. Sign up for alerts from your hosting provider, your payment processor, and any other critical service. When something breaks, you'll know within minutes whether it's your site or their infrastructure — and you won't spend hours troubleshooting a problem you can't fix.

3. Have a plan for when things go down

What do you do when your site is offline? Do you have a way to communicate with your customers? A temporary message on social media, an email to your list, a phone number people can call? A short, honest "we're experiencing technical difficulties and will be back shortly" message is infinitely better than silence. Customers are more forgiving than most businesses expect, as long as you're upfront with them.

4. Don't put everything in one basket

If your entire business — bookings, payments, customer data, contact forms — runs through a single platform and that platform goes down, you're completely stuck. Talk to your developer about which parts of your business are most critical and whether there are ways to add some redundancy. It doesn't have to be complicated or expensive. Sometimes it's as simple as having a backup contact form that goes straight to email.

5. Back up your data — and actually test the backups

Outages usually don't destroy data, but it happens. If the platform that holds your customer records, your orders, or your content goes down permanently, you want to have a recent copy somewhere else. Ask whoever manages your website when backups were last taken and — more importantly — whether they've ever actually tried to restore from one.

How We Handle It

When we build websites for clients, we spend time thinking about failure modes — not because we expect things to go wrong, but because the cost of not thinking about it is too high.

That means choosing infrastructure with strong uptime track records, setting up monitoring so we know about problems before clients do, and making sure clients have visibility into what their site depends on. It also means having honest conversations about risk: some businesses need every possible protection against downtime, and some are fine with standard hosting. The right answer depends on your business.

If you're not sure what your website is sitting on, or you've never had a conversation with your developer about what happens when something breaks, that's a good conversation to start. We have it regularly with our clients, and it's one of the more valuable things we do.

Reach out if you'd like to talk through it.