Most small businesses now run on the cloud, whether they think of it that way or not. Email, accounting, scheduling, card payments, patient records and phone systems often live on servers owned by someone else, in data centers you will never see. That is usually a good thing: those providers keep systems running far more reliably than a server in a back closet ever could. But "usually" is not "always."
Two weeks ago, a power problem inside one Google Cloud data center zone in London disrupted banks, delivery companies and retailers for most of a working day. The outage happened overseas, but the lesson applies to every office in New York: when a service you depend on goes dark, you need a plan for the hours until it comes back.
What happened on October 20
According to The Register, customer downtime began at 06:08 UTC (just after 2 a.m. in New York) when a "power event" knocked out europe-west2-b, one of three availability zones in Google's London region. An availability zone is a separate section of a cloud region with its own power and networking, designed so that a failure in one should not take down the others.
The impact still spread widely. The Register listed Lloyds Bank, HSBC, Royal Mail, Rightmove, DPD, BT, Screwfix and Revolut among the companies affected, and an academic it cited estimated that 38 Google Cloud services were down. Google said at 14:03 UTC that most customers had service back, with a very small number still affected. For many businesses, that meant roughly eight hours of disruption.
The Irish News reported that Revolut customers had trouble with card payments, transfers, cash withdrawals and logging in, and that more than 1,500 people reported problems on the outage-tracking site Downdetector that morning. A budgeting app called Emma also went offline and blamed the Google Cloud problem. Google apologized and said it would publish an analysis after its investigation.
Why this matters even if you have never heard of Google Cloud
Very few small businesses rent servers from Google, Amazon or Microsoft directly. Instead, they use apps that are built on top of those platforms: a point-of-sale system, a practice management program, an online booking tool, a payroll service. When the platform underneath has trouble, every app built on it can stumble at once, and you may not even know which platform your vendors use.
That is the real takeaway. Most of the people affected on October 20 were not Google customers. They were customers of their bank, their delivery company or their budgeting app. Your clients and patients are in the same position with you: if your systems stop, they experience it as your outage, not someone else's.
Step 1: Make a list of what you depend on
Sit down with whoever knows your systems best and write down every online service the business uses in a normal week. For each one, note:
- What it does for you, such as payments, email, scheduling, records or phones.
- How long you could manage without it. An hour? A day? Not at all?
- Who the vendor is and how to reach their support team.
- Where its status page is. Most providers publish one. Bookmark them all in one folder so you can check quickly instead of guessing.
The services you cannot live without for more than an hour are where your planning effort should go first.
Step 2: Decide how you will keep working
For each critical service, agree in advance on a manual workaround, so staff are not inventing one in the middle of a busy morning:
- Payments. Know whether your card terminal has an offline mode, and what your processor's rules are for using it. Decide whether you will take checks, send invoices later or pause certain sales.
- Appointments and orders. Print or export tomorrow's schedule at the end of each day. A paper order pad and a printed price list cost almost nothing and can save a whole afternoon.
- Contacts. Keep an up-to-date list of staff, key customers and vendors outside your email system, so you can still reach people if email is down.
- Phones. If your phone system is cloud-based, know how to forward the main number to a mobile phone, and test it before you need it.
Step 3: Keep your own copy of important data
Cloud providers protect their systems, but your contract usually does not promise that your data will be reachable every minute. Keeping a separate copy of the files and records you rely on most means an outage at one provider does not lock you out of everything.
- Use your apps' export features to save regular copies of customer lists, invoices and schedules.
- Back up cloud email and file storage, such as Microsoft 365 or Google Workspace, to a separate backup service rather than relying only on the provider's recycle bin.
- Test a restore from time to time. A backup you have never opened is a guess, not a plan.
Step 4: Do not let your own connection be the weak link
Sometimes the cloud is fine and the problem is the line into your office. A second internet connection from a different provider, or a cellular backup that switches on automatically, keeps your cloud tools reachable when your main line fails. For offices that depend on card payments or cloud phones, it is often one of the least expensive ways to cut downtime.
Step 5: Write it down and practice once
Ready.gov, the federal emergency preparedness site, lays out business continuity planning in six steps: prepare, set objectives, identify risks, develop strategies, assign teams and tasks, and test the plan. You do not need a thick binder to follow that advice. A two-page document that says who decides an outage is happening, which workarounds to use, and how you will tell customers is enough to start.
Then run a short drill. Pick a quiet afternoon, pretend your main business app is down, and see how far staff get with the workaround. You will find gaps in thirty minutes that would otherwise surface in the middle of a real outage.
Questions to ask your software vendors
- Which cloud platform and region do you run on, and do you have a backup region?
- How will you tell us about an outage, and where is your status page?
- Can we export our data whenever we want, and in what format?
- What does your service agreement promise about uptime, and what happens if you miss it?
The answers will tell you how much of the planning the vendor has already done, and how much is left to you.
How Rise Technologies can help
Rise Technologies has helped businesses across New York, Long Island and the Tri-State area with IT for more than 10 years. Our certified engineers can review the cloud services your business relies on through an IT assessment, set up data backup and recovery that keeps copies of your important data outside any single provider, and design cloud and hosting solutions that fit how your business actually works. When something does go wrong, our remote technicians can start on an issue within minutes.
Cloud outages are rare, but they do happen, and a little planning makes the difference between a stressful morning and a lost day. Call Rise Technologies at (516) 545-0065 or request a free quote to talk through a cloud outage plan for your business.
