By André Snoxall

The Acacia IT project was intended to transform healthcare, not simply replace ageing systems, but failed to follow its master plan, maintain clinical involvement, test whether continued spending was delivering value, and mistook extensive governance structures for effective oversight, writes project consultant André Snoxall.

The problems with the Northern Territory’s Acacia digital health program are easy to describe as an IT failure.

That description misses the most important point.

Acacia was never simply about installing new computer software. It was intended to change how healthcare works across the Territory: giving clinicians better access to patient information; replacing ageing systems; reducing duplication and risk; improving the way patients move through hospitals and clinics; and ultimately creating a single, more connected health record.

By March 2026, $319.6 million had been spent against a stated program budget of $335 million. Yet only two of the six planned stages had been delivered, and much of the old technology remained in use. A Public Accounts Committee inquiry was announced in February last year, and its report, published at the end of August, concluded that the program had not delivered the benefits on which the original investment decision was based.

So what actually went wrong?

Think of Acacia as building a new city

Imagine that the government decided to build a new city.

It would need two quite different kinds of management.

First, somebody must make sure that each individual building is constructed properly: foundations, walls, plumbing, electrical work, budget, timetable and quality.

That is broadly what a project-management method such as Projects in Controlled Environments, known as PRINCE2, is designed to do.

But somebody also needs to make sure that all the individual buildings combine to create a functioning city.

Where do the roads go? How do the electricity, water and sewerage systems connect? Where are the schools and hospitals? How will people actually live and work there? What happens to the old town? And, most importantly, does the finished city achieve what government promised when it decided to build it?

That is the role of program management, for which a method such as Managing Successful Programs, or MSP, is designed.

Acacia needed both. But the Department of Corporate and Digital Development, which was implementing the project on behalf of NT Health, employed a program director with, it seems, little if any experience in working with clients to lead large-scale programs.

The evidence suggests that those in charge became increasingly focused on building the individual pieces of technology without maintaining sufficient control over how all those pieces were supposed to combine to transform healthcare.

That is the central problem.

First mistake: treating healthcare transformation as an IT project

DCDD is the Territory’s central technology agency. It has an important role. Government needs professional expertise in infrastructure, cybersecurity, systems integration, vendors, contracts and IT service management.

But DCDD does not run hospitals. NT Health does. And that distinction matters.

The government’s own evidence to the Public Accounts Committee recognised that DCDD could produce the IT system but that only NT Health truly understood the business it was intended to support.

Yet Cabinet placed responsibility for delivering Acacia, control of its funding and program sponsorship with the central IT department rather than NT Health. DCDD therefore controlled the money while NT Health was expected to receive the clinical and operational benefits.

The committee heard evidence that this arrangement created tension and misalignment between the two agencies.

That is rather like asking the company building a hospital to decide how medicine should be practised inside it.

The builder is essential. But the doctors, nurses, administrators and health executives must determine what the hospital needs to achieve.

For Acacia, the proper division should have been simple: NT Health owns the healthcare transformation, clinical change, adoption and benefits, while DCDD delivers and supports the technology required to make that transformation possible.

Technology should enable the change. It should not define the change.

Second mistake: losing the master plan

Large programs need an agreed master plan. The Core Clinical Systems Renewal Program, which is the unit of DCDD tasked with implementing Acacia, paid to have a plan developed and then, apparently, shelved it. It appears to have decided instead to rely on a timetable loosely based on wishful thinking that the plan had already demonstrated was not achievable.

A timetable tells you when activities are supposed to happen.

But a genuine program plan tells you what is being built; why it is being built; who is responsible; what each project must deliver; what it will cost; how the pieces depend on one another; what old systems will disappear; what the finished organisation will look like; and how success will be measured.

The evidence raises serious questions about whether Acacia maintained such an authoritative program baseline throughout its life and about the competence of the executives who were responsible for the people delivering the program and who even now remain in positions of authority.

Without a plan, a very basic management question becomes surprisingly difficult. It is hard to tell whether we are actually ahead, behind or even still heading towards the destination we originally approved.

A changing schedule cannot answer that question.

Third mistake: confusing ‘software installed’ with ‘problem solved’

The problem of confusing the software installed with the problem being solved may be the easiest mistake for a non-technical reader to understand.

Suppose government spends money introducing an online service intended to reduce the time it takes citizens to complete a task from two hours to ten minutes.

The IT department successfully installs the system. That is an output.

But if citizens cannot use it, staff create workarounds, transactions still take two hours, and nobody measures the improvement, government has not achieved the benefit for which it paid.

Acacia suffered from exactly this distinction.

The 2019 Benefits Realisation Strategy, which I helped develop, set out a mechanism for identifying benefits, establishing starting measurements, setting targets, assigning people responsible for achieving them, and measuring them throughout implementation.

But the Public Accounts Committee found that this approach was not properly put into practice. It specifically rejected the idea that measuring benefits later through an external assessment was an adequate substitute for actively managing them while the system was being designed and introduced.

That is crucial. You do not wait until a house is finished to discover whether anyone remembered to connect the water.

Benefits must shape the design from the beginning.

The decision to defer benefits work was made based on advice from the DCDD chief executive officer who, it seems likely, felt that the “noise” produced through benefit management would not benefit DCDD.

For Acacia, every significant piece of work should have been able to answer: what improvement to healthcare is this supposed to produce, and how will we know whether it has?

Fourth mistake: clinical involvement weakened when it mattered most

Acacia actually began with very substantial clinical engagement. The Public Accounts Committee records thousands of requirements and hundreds of workshops involving clinicians and other health staff during the earlier stages.

The problem is that this level of engagement was not sustained.

The committee found examples of schedules being developed without NT Health involvement, work continuing with reduced clinical engagement, and later evidence that the program had become increasingly IT-centred rather than clinically led.

Eventually, a senior DCDD executive acknowledged that, at times, Acacia felt like something being done to NT Health rather than with NT Health.

That is a fundamental warning sign in healthcare.

Clinicians do not merely “use” a clinical system. The system changes how they admit patients, find information, order tests, manage medication, discharge patients and make clinical decisions.

Their involvement is therefore not consultation around the edges. It is part of designing the service.

Fifth mistake: the program’s internal referee was not sufficiently independent

A major program normally has a program management office, or PMO. For a lay reader, think of it as the program’s control room and referee.

Its job is not merely to organise meetings and update spreadsheets.

It should tell senior management objectively whether they are following the agreed plan, whether costs are increasing, whether milestones are genuine, whether risks are being hidden, whether promised benefits have disappeared, whether project plans are complete, whether problems are being escalated, and whether they are still building what government originally approved.

The evidence given to the Public Accounts Committee was that Acacia’s PMO did not perform this role adequately. It was said to lack independence, provide insufficient challenge, failed to enforce planning disciplines, and failed to ensure compliance with the very program – and project-management methods – that were supposed to control delivery.

If there was a significant opportunity for DCDD to shine, it may have been in the effective management of the program through a well-constituted PMO, empowered and staffed with skilled personnel. Instead, the head of the PMO, who has since been promoted several times, attempted to learn the complex art of program management and effective governance and reporting on the job, led by senior executives in DCDD who, the evidence suggests, had little if any experience in doing this properly.

That matters because once the independent control mechanism becomes too close to the delivery team, senior decision-makers can lose their reliable source of truth.

And then a much bigger problem arises.

Sixth mistake: money kept flowing without a sufficiently clear test of value

Perhaps the most important question of all is why money kept being spent without an understanding of whether value was being created.

Governance is not simply approving budgets – it is continually asking whether continuing to spend the money remains justified.

If a project originally expected to cost $100 is now going to cost $200, that does not automatically mean it should be stopped.

But somebody must ask what we are getting for the extra $100, what has already been delivered, what remains, what will finishing it now cost, whether the original benefits are still achievable, whether another approach would now represent better value, and whether the project should be stopped, redesigned or reduced.

Those questions become more important as expenditure rises.

Yet Acacia reached the point where almost the entire $335 million budget had been consumed while most of the planned stages, known as ‘functional groups’, remained undelivered.

The Public Accounts Committee even found difficulty reconciling some of the figures provided for the total program budget and recommended a financial audit.

This is not simply a question of why Acacia cost so much.

The more important question is why the governance system did not force a fundamental reassessment much earlier, as the gap between money spent and outcomes delivered became increasingly obvious.

Good governance exists precisely to make that intervention before the money is gone.

Seventh mistake: having lots of governance is not the same as being governed

The Acacia project had committees. Lots of them.

There were steering committees, implementation committees, clinical groups, specialty groups and working groups. The Public Accounts Committee was provided with an extensive governance structure.

But governance is not measured by the number of boxes on an organisation chart.

It is measured by whether those bodies can identify a problem, obtain reliable information, make a decision and ensure that something changes.

The Public Accounts Committee concluded that governance failures contributed to Acacia’s inability to deliver on time and on budget and recommended that the Territory’s information and communications technology governance framework itself be reviewed.

That is an important distinction.

A meeting is not governance. A report is not governance. A risk register is not governance.

Governance occurs when reliable information leads somebody with authority to make and enforce a decision.

In my view, there was a tendency towards leadership with a “hail fellow well met” approach to professional relationships. People focused on being “great guys” rather than “good people”. Hard questions were not pursued to closure and hard decisions were not taken at appropriate times.

So what should government do now?

The answer to the question of what the government should do now is not simply to “get Acacia back on schedule”.

A schedule was never the fundamental problem.

Nor should the government immediately decide that it must finish every part of the original Acacia plan simply because so much money has already been spent. Past expenditure cannot be recovered by spending more money badly.

The program needs a genuine reset, and I think six issues must be addressed.

First, the government should decide what the Territory’s future health service actually needs. What should a doctor, nurse, administrator or remote clinician be able to do when the transformation is finished? What should a patient’s experience look like? Which existing systems must disappear? Which systems should remain? What information must be available, where and when?

That future state needs to be understandable in plain English before another large technology decision is made.

Second, NT Health must own that future state. Its leadership should be accountable for the clinical transformation, changes in working practice, adoption and benefits.

DCDD should remain a critical partner, but its accountability should concentrate on delivering secure, reliable and cost-effective technology and IT services.

Third, the remaining work should be broken into real projects. Each should have a clear purpose, cost, timetable, owner, deliverables and definition of completion.

In the city analogy, instead of saying “we are working on the eastern district”, the government should know that it is building a particular road, substation or hospital, who is responsible for it, what it costs, when it will open and what must be true before it is accepted as finished.

Fourth, the government needs an independent program control function that reports the truth, whether the news is good or bad.

Its job should be to maintain one authoritative picture of scope, cost, schedule, risk, dependencies and benefits.

That gives ministers and senior executives something they have desperately needed throughout Acacia: a single source of truth.

Fifth, the benefits must be managed from the beginning.

So before approving each significant piece of work, the government should identify: what improvement is expected, how it will be measured, what the current baseline is, who owns the improvement, and what will happen if it does not materialise.

The Public Accounts Committee itself concluded that benefits do not automatically flow from installing a system, and that they must be actively managed throughout the project lifecycle.

Sixth, the government should require an independent business case before committing substantial additional money.

The Public Accounts Committee has already recommended this. It said further funding should be subject to an independently prepared business case developed through rigorous clinical engagement and clearly identifying costs, objectives, benefits, governance and implementation timeframes.

That review should not begin with the question of how much money is needed to finish Acacia. It should ask what NT Health now needs, what already exists that is useful, what remains worth doing, and what the safest and best-value way of getting there is.

PRINCE2 and MSP: why both matter

This is where the seemingly dry subject of project-management methodology becomes very practical.

PRINCE2 helps the government build each piece properly. It provides discipline around responsibility, cost, stages, risks and the continuing justification for an individual project.

MSP helps the government make sure all those individual pieces combine to deliver the intended transformation. It maintains the link between projects, organisational change and benefits.

Using PRINCE2 without effective program management can result in the government successfully delivering individual bits of technology while the overall transformation fails.

Using program management without disciplined projects can produce an impressive vision that never gets reliably built.

The government needs both.

One asks: are we doing this piece of work properly?

The other asks: are these pieces of work collectively producing the future we promised?

That is the distinction Acacia appears to have progressively lost.

The lesson from Acacia

The lesson should not be that the government is incapable of delivering large digital programs. Nor is the lesson that technology is inherently too complicated.

The lesson is much simpler.

You cannot manage a transformation by managing the technology alone.

A successful health transformation requires clinical leadership, technology expertise, disciplined projects, independent program oversight, measurable benefits and clear accountability.

Acacia had many of the individual ingredients.

Some of the required governance methods were documented. Benefits arrangements were designed. Clinical groups existed. Reviews repeatedly identified problems.

The failure was that these things did not consistently operate together as a management system capable of saying: this is what we promised; this is where we are; this is what we have spent; this is what has actually improved; this is what remains; and does continuing still represent value for Territorians?

That is what the government now needs to rebuild.

The objective should not simply be to finish Acacia.

It should be to ensure that the Northern Territory finishes with a health system that works better, a government that knows whether its investment delivered value, and the permanent capability to manage the next generation of digital transformation properly.