When Digital Projects Go Wrong, It Usually Starts Long Before Launch

The Bureau of Meteorology’s website overhaul has become a headline example of a public digital project going off the rails. A decade-long modernisation program, with the initial cost of $4.1 million and a final cost of $96.5 million, nine contract extensions, and a public backlash that arrived the moment the new interface went live.

The temptation is to blame the design, the technology, or even the agencies involved. But the real issue sits further upstream. This is a governance and scope control problem long before it is a user experience problem.

Where Large Projects Actually Fail

Major digital transformations don’t collapse at the finish line. They unravel in the early stages, where decisions about scope, ownership and procurement shape everything that follows.

When requirements are unclear, when oversight is fragmented, and when contracts are repeatedly extended without fresh evaluation, costs grow quietly in the background until they’re impossible to ignore.

By the time the public sees the interface, the critical choices that determined the final price have been locked in for years.

Procurement Discipline Is Not Optional

The BoM program shows how quickly budget and scope can drift when procurement structures are not designed for accountability.

A contract originally valued at $31 million expanded to $78 million through nine extensions. Another grew from $11 million to $35 million.

These are not failures of expertise. They are failures of governance.

Clear delivery milestones, independent technical validation, fresh tender processes and tighter controls around vendor performance are not bureaucratic hurdles. They are risk management.

Public digital infrastructure depends on them.

The Hidden Complexity Doesn’t Justify the Blowout

There is no question that BoM’s systems are complex. Ingesting and processing vast data streams, rebuilding vulnerable legacy infrastructure after a cyber intrusion, and ensuring national resilience is not inexpensive work.

But complexity does not excuse poor programme control.
Government agencies manage complex engineering, transport and defence projects every day with strict oversight. Digital should be treated no differently.

A Public Reminder for Future Programs

The BoM website became the visible symbol of the entire $96.5 million spend, even though only a fraction covered the user-facing interface. That disconnect is part of the problem.

For public-facing digital work, transparency matters.
So does clarity around what is being built, why it costs what it does, and how success will be measured.

Why This Matters to the Industry and to Us

At Spark, we see these challenges from the other side. Organisations come to us after experiencing drifting scopes, unclear deliverables or project fatigue.
The pattern is always the same:

• the brief wasn’t clearly defined
• multiple stakeholders pulled the project in different directions
• decisions weren’t documented or validated
• vendors worked to hours rather than outcomes
• user needs were never fully surfaced

The result is wasted time, strained budgets and a digital product that doesn’t deliver what people need.

How We Approach It Differently

Our work across government, construction, professional services and health has taught us that project success hinges on the foundations:

• clear scope defined at the outset
• a measurable delivery plan
• transparent communication with all stakeholders
• user-centred validation before and after design
• strict controls that protect budgets from drift
• in-house delivery paired with specialist partners only where needed

These principles aren’t glamorous, but they are what stops a $4.1 million project becoming a $96.5 million one.

Final Thought

The BoM story isn’t about pointing fingers. It’s about recognising the importance of disciplined governance in digital transformation, and the responsibilities that come with public trust and public money.

It’s also a reminder to every organisation, large or small, that the decisions made at the start of a project determine its fate at the end.