The rating on the page in front of you is Medium-Low. Planned completion sits inside the next six months. Every conversation the team has had with you at governance sits behind that mark, and everything you now decide about the last stretch of the program will turn on how you read it.
In Brief
- A Medium or lower delivery confidence rating reflects what the project team believes about its own trajectory.
- One-third of projects reporting a Medium or lower rating have planned completion inside the next six months.
- The rating cannot distinguish confidence in the plan from confidence in the current plan holding to completion.
- An independent view initiated four months out shapes decisions; initiated eight weeks out, it documents them.
- The sponsor's accountability at completion is untested unless the project has been read from outside the delivery structure.
The rating is the delivery team’s assessment of the delivery team’s own trajectory. That is worth stating plainly, because a Medium-Low rating is often read as though it were a neutral observation of the project rather than a self-report from inside it. Delivery confidence ratings are built inside the delivery structure that produced them. They carry the pressures the team is working under and the reading the team has taken of what leadership expects to hear. That is not dishonesty. It is what a self-reported assurance metric is designed to do.
A Medium-Low rating is a self-assessment
The Digital Transformation Agency’s Major Digital Projects Report 2026 (MDPR-2026) covers 103 active projects across 43 agencies, with $9.7 billion in total budget commitment and $5.9 billion of that in digital technologies (Digital Transformation Agency, 2026a). Sixty percent of Tier 1 and Tier 2 projects report a Medium-High or High delivery confidence rating. That figure is essentially flat against the 60.7% reported the year before. What has moved is where the dollars sit: 69.5% of the total budget is now behind projects reporting Medium-High or High confidence, up from 52.9% (Digital Transformation Agency, 2026a). Confidence has consolidated around the money, not around the count of projects.
The figure that matters for a sponsor holding a Medium or lower rating six months out from completion sits further down the same report. In the DTA’s own words, one-third of projects reporting a Medium or lower delivery confidence have a planned completion date within the next 6 months. The agency’s framing is that these projects are more susceptible to schedule slippage and quality risks, particularly where dependencies and residual issues converge late in the delivery cycle (Digital Transformation Agency, 2026a). One in three sponsors in a comparable position across the Commonwealth is looking at the same window from the same seat.
Read carefully, the rating in that window is doing more than one thing at once. It carries what the team knows about its own progress. It carries what the team believes about how the remaining work will land. And it carries a judgement the team has made about the sensible level of confidence to send up the assurance chain. Under stable delivery conditions those readings tend to line up. Inside the final six months of a complex program under schedule pressure, they can separate, often without the team producing the rating being able to see that they have.
The team's self-assessment is usually accurate
The organisational conditions that make delivery confidence assessments useful are the same ones that make them a partial view. Delivery confidence is a self-report. The people producing the rating are the same people whose credibility and next contract cycle depend on the rating not being materially wrong. That is what the assurance process asks of them: give the accountable delivery leadership the responsibility to declare, in their own judgement, how the work is going. Judgement inside a running program is bounded by what the program is set up to see.
What the program can see six months from completion is largely execution. Burn against schedule, defect trends, integration progress, readiness for cutover. What the program cannot easily see is whether the underlying diagnostic that shaped the work eighteen months ago has held: the assumption about user need, the read of the operational environment, how dependencies with other agencies were sequenced. Execution can look on track against a plan that is no longer answering the right question. A delivery confidence rating cannot separate those two conditions. At the level the rating is captured, they look the same.
The DTA report treats independent assurance as a decision-quality mechanism rather than a compliance one. In 2026, 88.3% of projects had an independent delivery confidence rating, up from 80.3% the year before (Digital Transformation Agency, 2026b). The Deputy CEO, Simon Quarrell, described high-performing projects as those where the delivery organisation uses assurance as a strategic tool to improve the quality of key decisions and stay on track (Digital Transformation Agency, 2026c). An independent view reads the same project from outside the pressures shaping the internal view. Its usefulness lies in whether it can tell you the confidence expressed is confidence in the plan, or confidence in the current shape of the plan holding through to completion.
The sponsor must answer what the team can't
For the sponsor holding a Medium or lower rating with six months to run, what follows is specific.
You are now the accountable holder of a question the delivery team is structurally not positioned to answer: is the program on track to deliver the outcome the executive committee approved, or on track to deliver what the plan has come to describe? The rating in front of you speaks to the second. The first is a different question and needs a different reader.
The useful window for an independent view is narrower than the calendar suggests. An assessment initiated four months out has time to shape decisions about scope, sequencing, and integration readiness. The same assessment initiated eight weeks out becomes a documentation exercise. It names what could have been done differently rather than shaping what can still be done. Value drops sharply across the last three months of a program because the space in which decisions can still be taken is closing. The rating on the page is not the trigger for the assessment. The calendar is.
Doing nothing has a specific cost. The sponsor accepts the delivery team’s assessment as the assessment, retains full accountability for what happens at completion, and answers to the executive committee and to the Minister with a view of the project that has never been tested against an independent read. If the rating holds and the program lands, that risk was invisible. If the rating does not hold, the gap between what the sponsor believed and what the organisation delivered is what the post-implementation review will name, and by that point the record already exists.
A rating is not independent assurance
The rating in front of you is doing its job. It is telling you what the delivery team believes, expressed through the discipline the DTA has established for reporting delivery confidence. That is genuinely useful information, and it is the information the assurance process was designed to produce. What it cannot do is answer the question a sponsor holds at completion: was the project delivering the outcome the executive committee approved eighteen months ago, or delivering what the plan came to describe over the intervening period?
The rating is a reading of the project by the people delivering it. An independent view is a reading of the same project from outside the delivery structure. Both belong on the record when the delivery window is inside six months, when the rating sits at Medium or lower, and when completion sets the terms of what the organisation lives with next. The sponsor who initiates the second reading while the calendar still allows a decision to be taken is accountable at completion for a project that has been read from two positions, not one.
What this means for senior leaders
- The rating in front of you is a self-report from inside the delivery structure, not an independent reading of the project. Treat it as one input to your decision, not the decision itself.
- Six months from completion, the rating can see execution but not whether the diagnostic that shaped the work still holds. Work can track cleanly against a plan that has quietly stopped answering the original question.
- The question you answer for at completion — is the program delivering the outcome the executive committee approved — is one the delivery team is not positioned to answer. It needs a reader from outside the delivery structure.
- The window for an independent view is set by the calendar, not the rating. Initiated four months out it shapes scope and sequencing; initiated eight weeks out it only documents them.
- Accepting the rating as the assessment carries a cost you hold personally: accountability at completion for a project that was read from one position when it could have been read from two.
References
- Digital Transformation Agency. (2026a). Major Digital Projects Report 2026: Project performance. Digital Transformation Agency, Australian Government.
- Digital Transformation Agency. (2026b). Major Digital Projects Report 2026: Key figures. Digital Transformation Agency, Australian Government.
- Digital Transformation Agency. (2026c, February 9). Release of the Major Digital Projects Report 2026 [Media release]. Digital Transformation Agency, Australian Government.
