Business cases exist for good reason. They protect public money, surface hidden assumptions, force agencies to say what success looks like, and create the record that accountability requires. In government, that discipline is not bureaucratic overhead. It is an
obligation.
But the way the government currently applies business-case logic to digital programs often produces the opposite of its intended effect. Platforms arrive late, run over budget, and need the vendor involved to change something as small as a field label. This happens not despite the business case process, but partly because of its structure.
The issue is not rigour. It is that we apply the same level of rigour to every decision, regardless of how consequential that decision is.
A better way to think about what you are investing in
An approach gaining traction in the Digital Twin and geospatial space is to separate your investment into two layers and apply proportionate scrutiny to each.
The infrastructure layer is the shared foundation. Parcel data, flood data, demographic datasets, joined on common identifiers and exposed through open interfaces. It is stable, central, and designed for use by many agencies over many years.
The application layer sits above it. These are discrete, lightweight, agency-owned tools that draw from the infrastructure. A housing feasibility calculator. A transport optimiser. A health infrastructure planning tool. They are specific, short-lived, and replaceable.
The problem is not that the government writes business cases. It is that the government writes the same kind of business case for both layers, then folds them into a single integrated platform contract that moves at the speed of the slower, more complex thing. That is always the infrastructure.
What a better model looks like
The infrastructure business case should be more rigorous, not less. This is the decision that compounds across agencies and budget cycles. It should address data sovereignty and security classification, interface governance and interoperability standards, long-term custodianship, vendor lock-in risk, and total cost of ownership shared across all the agencies that benefit. This is exactly where central agency scrutiny belongs, because getting it wrong is expensive and slow to reverse.
The application business case should be different in form. Rather than a long return-on- investment forecast for a tool that might take weeks to build, agencies need a light assurance framework. Does this tool draw only from approved infrastructure? Does it meet security requirements? Is the agency the data controller? Can it be switched off cleanly? Those questions can be answered in days, and the authority to act on them should sit with the agency inside a pre-agreed envelope, much like minor capital works budgets already work.
Pre-approved application types reduce duplication without reducing accountability. Central agencies can endorse a set of common tool types, such as feasibility tools, scenario models, and dashboards, that agencies can commission without a fresh business case each time, provided they draw from approved infrastructure. Accountability does not disappear. It shifts from forecasting upfront to reporting afterwards on what was built, how it was used, and what decision it informed.
Panel contracts replace repeated procurement without bypassing competition. A single competitive process sets up a pre-qualified panel, and individual builds are called off against it. New Zealand’s All-of-Government contracts and Australia’s Digital Marketplace already provide this model. The real question is whether agencies are actively structuring their Digital Twin programs to use them.
The accountability argument
The case for change is not that government should be less careful with public money. It is that the current model mistakes the process for protection. A four-hundred-page business case for an integrated platform does not reduce risk. It documents that risk thoroughly before it materialises.
Separating the layers lets the government do what good business cases were always meant to do. Concentrate scrutiny on the consequential, hard-to-reverse decisions, and apply faster, lighter governance where the cost of being wrong is small, and the ability to correct course is high.
That is not a lower standard. It is a more honest one.
GC3 Digital works with government agencies on digital strategy, procurement design, and technology governance. Get in touch.
