Most business continuity planning is written for offices. Systems go down, people work around it for a few hours, and the backlog gets cleared the next day. Nobody is harmed and the cost is measured in lost productivity.
Care does not work like that. In residential aged care, home care and disability support, the systems that fail are the ones that tell staff who needs what, when, and where — and the people depending on that information cannot wait for the backlog to clear. That is why the continuity question for a care provider is not "how long until we are back" but "what happens to the people we support in the meantime".
The scenario worth planning for specifically is the roster failing at shift change.
Why shift change is the moment that matters
A rostering system outage at eleven in the morning is a problem. The same outage at six in the morning, as the day shift is trying to find out where they are going and the night shift is trying to hand over, is a different kind of problem entirely.
In home care the roster *is* the service. If workers cannot see their schedule, they do not know which client to visit, at what time, for what tasks, or whether anything changed overnight. Clients waiting for personal care, medication prompts or meal support do not know why nobody has arrived. Coordinators are fielding calls from both sides with no more information than anyone else. Every minute the system is down, decisions are being made from memory and guesswork about people who are relying on them being made correctly.
In residential care the same failure looks like a floor of staff without access to care plans, medication charts and clinical notes at the start of a shift — working from what they remember about residents rather than what is documented.
None of this is hypothetical. It is the ordinary consequence of an ordinary IT failure — a server fault, an expired certificate, a cloud provider incident, a ransomware attack — landing at the worst possible time, which over a long enough period it eventually will.
The questions most providers cannot answer
We ask care providers three things, and the honest answers are usually uncomfortable.
Where does the roster actually run, and what does it depend on? Many providers use a cloud rostering or care management platform and assume that means "it is the vendor's problem". Partly true. But the platform depends on your internet connection, your staff's devices, your identity system for logins, and sometimes an on-premise integration nobody remembers setting up. Any of those failing produces the same outcome as the platform itself failing.
When did anyone last restore from the backup? A backup that has never been restored from is a hope, not a control. Vendors back up their platforms; whether *you* can get *your* data back, in a form you can use, in a timeframe you could survive, is a question only a test answers.
What is the manual fallback, and does the night shift know it? If the system were unavailable right now, how would a coordinator tell forty workers where to go? Is there a printed or exported copy of tomorrow's roster anywhere? Does anyone on shift know where it is and that it exists? Usually the answer is that there was a process once, and nobody current has seen it.
What a real continuity design looks like for care
It is less about technology than about deciding, in advance, what happens.
Treat the rostering and care systems as safety-critical. That means a documented understanding of what they depend on, backups that are confirmed real and restore-tested, and a recovery time that has been agreed as acceptable — not discovered during the incident. "We would find out" is not a recovery time.
Keep an offline copy of the next 24 to 48 hours. A scheduled export of the upcoming roster — to a location reachable without the main system, and to something as simple as a printed sheet in the office — converts a total outage into a degraded day. It is cheap, it is unglamorous, and it is the single control that most directly protects the people receiving care.
Give every shift a way to reach every worker without the system. A current contact list that does not live only inside the platform that has just failed. The number of providers whose only copy of workers' phone numbers is in the rostering app is larger than it should be.
Plan for the platform being fine and everything around it failing. Internet redundancy for the office (a mobile broadband failover is enough). Staff devices that can reach the platform on mobile data. Logins that do not depend on an on-premise server being up.
Write down who decides what. When the system is down at shift change, who calls the outage, who switches to the fallback, who tells clients and families, and who is allowed to make care decisions from the offline copy. Assigning that in the moment is how mistakes happen.
Rehearse it once. A tabletop exercise — "it is 6am and the roster is down, go" — takes an hour and finds every gap in the plan while nobody is at risk.
The obligations sitting underneath this
Care providers are not just protecting operations. Health information is classified as sensitive information under the Privacy Act 1988, and the small business exemption does not apply to health service providers, so even a small provider carries the Act's obligations to protect it. The Aged Care Quality and Safety Commission assesses governance and information management rather than assuming it. NDIS providers carry the NDIS Practice Standards on top. A continuity plan that keeps care and medication records available *and* protected during an incident is part of meeting those obligations, not a separate IT nicety.
This is the foundation of how we approach IT for aged care providers, and the same thinking applies to the rostering-dependent world of NDIS and disability providers. The backup and continuity work is where it starts — not because backups are interesting, but because the alternative is a floor of staff and a list of clients depending on what someone can remember.
One question to answer this week
If the roster went down at six tomorrow morning, could the coordinator on duty tell every worker where to go without it?
If the answer is no, that is the highest-value thing to fix — and it is usually fixable in a day.
*Last updated: 17 September 2026.*