The data problem in care delivery

If you run an aged care, NDIS or support-at-home business, you are not short on data. Every shift logged, every progress note, every invoice, every incident report, every medication record and every NDIS claim is data. The problem is not that you lack it — it's that the data lives in silos: separate systems that don't talk to each other.

A typical mid-market provider might run a client management system (like Lumary, Brevity or ShiftCare), a rostering and timesheet tool, accounting software like Xero or MYOB, a separate platform for NDIS claiming, and email or a help desk for family and referrer enquiries. Each one holds a piece of the truth about a single client or a single worker, but no one system holds all of it.

The result is that answering a simple question — *"Is this client's care profitable, and are we delivering the hours their plan funds?"* — means logging into four tools, exporting spreadsheets, and stitching them together by hand. That's the data problem in plain terms: the information exists, but it isn't connected, so it can't easily be used.

Aged care and disability support are unusually data-heavy and unusually high-stakes. You're coordinating a mobile workforce, funded by complex schemes (NDIS price guides, Support at Home, Home Care Packages), and caring for vulnerable people where mistakes carry real consequences. That combination multiplies the number of systems and the cost of getting things wrong.

Consider funding utilisation. An NDIS participant's plan funds a set number of hours per category. If your delivery data sits in the rostering tool and the budget sits in the claiming system, no one notices a plan is being underused until it's nearly expired — or overused until you're delivering unfunded care for free. Both outcomes hurt the client and the business, and both are invisible when the data is split.

The same applies to compliance and quality. Incident reports, restrictive practice records and care-plan reviews are often scattered across notes, emails and PDFs. When the Aged Care Quality and Safety Commission or the NDIS Commission asks for evidence, staff spend days reassembling a story that the systems already contain but can't tell on their own.

There's also a workforce angle. Award interpretation, travel time, broken shifts and SCHADS pay rules mean your rostering data and your payroll data have to agree. When they don't, you get underpayments, overpayments and disputes — all because two systems hold slightly different versions of the same shift.

Fragmentation rarely shows up as a single big bill, which is why it's easy to ignore. Instead it leaks value quietly. Coordinators spend hours each week copying numbers between systems. Finance closes the month late because reconciling claims against delivery is manual. Managers make decisions on gut feel because the report they need would take two days to build.

It also caps what you can do next. Everyone is talking about AI and automation — predicting which clients are at risk of a hospital admission, auto-drafting progress note summaries, flagging unusual claims. None of that works reliably on data trapped in separate tools. An AI model can only reason over what it can see, and right now most providers' information is something no single tool can see in full.

The hidden cost is opportunity: the questions you stop asking because the answer is too hard to get. Over time, a fragmented business simply operates with less self-awareness than its data would otherwise allow.

The fix is conceptually simple, even if the plumbing isn't. You bring the data from your scattered tools into one place — a single, governed database — where a client record, their funding, their delivered hours, their notes and the cost of their care all sit together and stay current. This is often called a unified data foundation, and it's the layer that analytics, AI and automation are built on top of. (A platform like Intellova exists to do exactly this unifying step, so your team doesn't have to.)

It's worth being clear about the order of operations. You don't start with a fancy dashboard or an AI tool; you start by connecting the data. A dashboard built on disconnected exports is just a prettier spreadsheet, and an AI feature bolted onto one system only ever sees that system's slice. Get the foundation right, and the same questions that used to take two days start taking two minutes — and questions you couldn't ask before become answerable.

Practical takeaway: Before this week ends, write down the five systems where your most important information lives, and pick one question that needs data from at least two of them — for example, *"Which clients are underusing their funded hours this quarter?"* If answering it means exporting and manually merging spreadsheets, you've just found your data problem in concrete form. That single question is the starting point for everything the rest of this course will teach you to do.