Mid-August 2026 was a rough week to be “all in on the cloud.” Australian users hit a cluster of incidents across Microsoft Azure, Google services (including Drive and a Melbourne-region disruption), and AWS — several of them inside seven days.
You cannot stop Microsoft or Google having a bad afternoon. You can stop that afternoon becoming a full-day standstill for your business. This is a practical resilience checklist for Australian businesses — Melbourne offices, regional sites, and remote teams — who live in Microsoft 365, cloud accounting, and shared drives.
Quick Navigation
A Bad Week, Not a Rare Fluke
Between 14 and 20 August 2026, status trackers logged issues with the Azure Portal in Australian regions, Google Drive and related services, a lengthy disruption in Google’s newer Melbourne cloud region, and a partial AWS incident. None of that means you should abandon the cloud. It means “the cloud” is still someone else’s computers — and those computers fail.
The SMB version of the problem
Enterprise teams talk about multi-region failover. A ten-person office in Melbourne, or a regional team anywhere in Australia, talks about: can we still email customers, take payments, and open yesterday’s files? Design for that, not for a data-centre architecture diagram.
Know What You Cannot Operate Without
Write the list down. Most businesses discover their critical path during the outage: email, identity (Microsoft 365 login), the website, Xero/MYOB, the CRM, phones, and the shared drive. If three of those sit behind one login, one outage knocks over the lot.
Usually critical
- •Microsoft 365 email and Teams
- •Accounting and payment systems
- •Customer files and the CRM
- •Domain, DNS, and the website
Usually forgotten
- •Who holds the admin logins (and MFA device)
- •How you contact staff if email is down
- •Whether backups actually restore
- •Old app integrations nobody remembers
The First 30 Minutes of an Outage
Panic-Googling and random password resets make it worse. Use a short, boring sequence.
- 1. Check the real status page — Microsoft 365 Service Health, Azure Status, Google Workspace Status — from a bookmark you already have. Do not trust an unsolicited “we’re down, click here” SMS. Outages are catnip for phishers.
- 2. Tell the team on a channel that still works — a phone tree, a WhatsApp group, SMS. Agree this in advance so you are not trying to email people that email is down.
- 3. Switch to the fallback — offline copies of price lists, a mobile hotspot if the office internet is the problem, a secondary mailbox only if you actually set one up.
- 4. Call your IT provider if identity, DNS, or backups are involved. Distinguishing “Microsoft is down” from “our tenant is misconfigured” is their job.
Backups Are Not the Same as “It’s in OneDrive”
Microsoft 365 retention is not a backup you control. If a sync error, ransomware, or a deleted tenant-wide folder takes files out, you need a copy that lives somewhere else and a restore you have actually tested. An untested backup is a rumour.
What
Email, SharePoint/OneDrive, finance files, and the website — not just the laptop Desktop folder.
Who
Named admin accounts, MFA, and a second person who can restore if the owner is away.
When
A restore test at least once a quarter. Pick one file. Get it back. Time it.
Do This Before the Next Wobble
- ✓List critical apps and who owns the admin login for each.
- ✓Turn on MFA everywhere that holds customer or finance data.
- ✓Confirm Microsoft 365 (or Google Workspace) is actually being backed up, not just synced.
- ✓Bookmark vendor status pages. Agree an out-of-band staff contact method.
- ✓Know whether your internet has a failover (second link or 4G/5G backup) if the office line dies.
Make the Next Outage Boring
We’ll map your critical cloud apps, check backups actually restore, and put a simple outage plan in place so the office is not guessing.