Most Infrastructure Delays Are Locked In Before Construction Starts

Executive Summary

Australia continues to treat infrastructure failure as a delivery problem.

The evidence increasingly suggests otherwise.

Globally, fewer than 10 per cent of major projects are delivered on time and on budget. Bent Flyvbjerg’s database of more than 16,000 projects found that only 8.5 per cent achieved both targets, with average cost overruns approaching 30 per cent.

Olympic Games provide one of the clearest examples. The Oxford Olympics Study 2024 found the last three Summer Games averaged cost overruns of 185 per cent in real terms, with overruns described as “the norm”.

These failures are rarely caused by workers suddenly becoming less competent once construction begins.

The underlying problems are usually embedded much earlier:

• fragmented governance

• poorly sequenced decisions

• disconnected systems

• weak escalation pathways

• incomplete evidence

• operational outcomes ignored during planning

By the time delivery starts, much of the project’s risk profile is already fixed.

This matters because Australia is entering one of the largest infrastructure and housing delivery periods in its history while simultaneously facing:

• workforce shortages

• infrastructure backlogs

• rising construction costs

• increasing climate obligations

• tighter fiscal conditions

• public distrust following repeated overruns

Yet many organisations still respond by adding:

• more reporting

• more consultants

• more software

• more governance layers

without fixing the underlying decision architecture.

The problem is not simply whether projects are delivered efficiently.

It is whether the right decisions were made before financial and operational exposure became locked in.

This distinction is critical.

A project can appear “green” in reporting dashboards while carrying embedded delivery risks that only emerge years later through:

• claims

• redesigns

• scope disputes

• operational failures

• escalating lifecycle costs

• delayed benefits realisation

The issue is particularly relevant for councils, infrastructure agencies, and growth regions now dealing with housing pressure, asset renewal, and Brisbane 2032-related investment expectations.

Digital tools alone will not solve this.

BIM, digital twins, workflow systems, and dashboards can improve visibility, but only when integrated into governance structures that align:

• accountability

• escalation

• contractual obligations

• operational outcomes

• lifecycle value

Without that integration, digital systems often increase complexity rather than reduce it.

The challenge is therefore not primarily technological.

It is governance and decision sequencing.

Infrastructure projects do not usually fail during delivery.

Most failures are already forming before construction begins.

The Problem Definition

Most infrastructure organisations still assume delays and cost overruns occur during construction.

That assumption is comforting because it localises failure:

• contractors

• supply chains

• labour shortages

• weather

• inflation

These factors matter, but they are often secondary accelerants rather than root causes.

The more difficult reality is that projects frequently enter delivery carrying unresolved structural problems:

• incomplete scope definition

• weak operational requirements

• fragmented accountability

• unclear escalation pathways

• disconnected digital environments

• unrealistic assumptions

• political timing pressures

When this occurs, delivery teams inherit risks they did not create but are expected to absorb.

The consequences then appear later as:

• claims

• redesigns

• disputes

• approvals bottlenecks

• procurement churn

• rework

• operational inefficiency

At that point organisations often describe the outcome as “unexpected”.

In reality, many of these failures were structurally predictable.

Flyvbjerg describes this pattern as the “Iron Law of Megaprojects”:

over budget, over time, under benefits, over and over again.

This is not confined to megaprojects.

The same logic appears across:

• housing

• hospitals

• transport

• utilities

• local government

• schools

• digital transformation programs

The scale changes.
The pattern does not.

The Evidence Base

Global project performance remains structurally poor

Flyvbjerg’s global database covering more than 16,000 projects found:

• only 8.5% achieved cost and schedule targets

• only 0.5% achieved cost, schedule, and benefit targets simultaneously

That is not random variance.
It is systemic failure.

Olympic Games demonstrate the pattern clearly

The Oxford Olympics Study 2024 found:

• the last three Summer Olympics cost a combined USD $51 billion

• average cost overruns reached 185% in real terms

The study concluded:

“Cost overruns are the norm for the Games, past, present and future.”

This matters for Brisbane 2032 because Olympic programs compress existing governance weaknesses into fixed deadlines and public accountability environments.

The Games do not create governance problems.
They expose them.

Construction productivity remains stagnant

Australia’s infrastructure challenge is compounded by weak productivity growth.

Your own XD Thinking™ working paper summarises Infrastructure Australia findings showing construction productivity has remained effectively flat since the mid-1990s.

At the same time:

• infrastructure demand is increasing

• asset complexity is increasing

• compliance obligations are increasing

• coordination requirements are increasing

Yet delivery systems remain heavily fragmented.

Why Current Reform Efforts Often Fail

Most reform programs focus on symptoms rather than structural causes.

Examples include:

• new software platforms

• additional reporting layers

• digital dashboards

• workflow tools

• governance committees

• isolated BIM mandates

These can improve local efficiency while failing to improve system performance.

Why?

Because they often sit outside enforceable decision structures.

Your XD Thinking™ paper identifies this directly:

“technology, contracts, and policy have been pursued in parallel rather than as integrated governance.”

This is a critical distinction.

Digital systems do not automatically improve delivery outcomes if:

• incentives remain misaligned

• escalation occurs too late

• operational teams are excluded early

• lifecycle obligations are weak

• accountability is fragmented

In some organisations, additional systems actually increase coordination overhead.

Teams spend more time:

• reconciling information

• managing approvals

• translating between systems

• preparing reports

• escalating issues manually

while delivery confidence continues to decline.

The Governance Problem

Infrastructure projects are fundamentally decision systems.

Every major outcome is shaped by:

• who decides

• when decisions occur

• what evidence is required

• how risk is escalated

• whether operational impacts are considered early

• how accountability is enforced

Poor sequencing creates embedded exposure.

For example:

• approving projects before operational requirements are resolved

• locking budgets before scope maturity exists

• committing to timelines before utility coordination is complete

• separating digital delivery from operational governance

• treating lifecycle outcomes as post-project considerations

These are governance failures before they become delivery failures.

This is why many projects appear manageable early and deteriorate later.

The underlying risk already existed.
It simply had not surfaced yet.

The Implications for Councils and Infrastructure Agencies

This issue is becoming more significant across Australia because organisations are now operating under simultaneous pressure from:

• housing targets

• ageing assets

• climate obligations

• rising community expectations

• fiscal constraints

• workforce shortages

• infrastructure sequencing challenges

Councils and delivery agencies are increasingly expected to deliver:

• faster approvals

• more housing

• better infrastructure coordination

• stronger transparency

• improved lifecycle outcomes

using governance models often designed decades earlier.

The result is organisational friction.

Symptoms typically include:

• repeated audit findings

• fragmented evidence

• duplicated coordination effort

• delayed escalation

• unclear ownership

• growing operational exposure

• poor transition from CAPEX to OPEX

These are rarely isolated failures.

They are indicators of systemic governance strain.

Pathway to Improvement

Shift focus from delivery reporting to decision quality

Most reporting occurs after exposure already exists.

Organisations need earlier visibility into:

• governance gaps

• sequencing failures

• operational risk

• evidence maturity

• escalation delays

Integrate digital systems into governance structures

Digital tools should support enforceable operational outcomes, not exist as isolated technical environments.

Without governance integration, fragmentation simply becomes digital fragmentation.

Involve operational and lifecycle stakeholders earlier

Projects optimised only for delivery frequently create long-term operational inefficiency.

Facilities, asset management, maintenance, and operational teams need involvement before commitments are locked in.

Strengthen escalation pathways

Many delivery failures worsen because issues surface too late or move through fragmented decision chains.

Clear escalation structures reduce embedded exposure.

Treat infrastructure as a system, not isolated projects

Transport, housing, utilities, approvals, workforce capacity, and operations are interconnected systems.

Managing them independently creates compounding coordination failure.

Conclusion

Australia does not primarily have a construction problem.

It has a governance and decision sequencing problem.

The evidence is now overwhelming:

• cost overruns are persistent

• delays are systemic

• productivity remains weak

• fragmented reforms continue to underperform

Most importantly, many of these outcomes are forming long before delivery begins.

Projects rarely fail suddenly.

Risk accumulates gradually through:

• poor sequencing

• fragmented accountability

• disconnected evidence

• weak governance integration

By the time delivery teams encounter the consequences, many critical decisions have already been locked in.

The question for councils, infrastructure agencies, and delivery leaders is no longer whether these patterns exist.

The question is whether current governance structures are capable of identifying and correcting them early enough to matter.

Because once delivery starts, the window for low-cost correction rapidly closes.