Think right to left: why most projects start in the wrong place
Each year, the world invests approximately $3.4 trillion in megaprojects. These are the projects that shape economies, underpin energy systems and define industrial futures. Yet their track record is remarkably poor: the overwhelming majority run over budget, over schedule, and fail to deliver the benefits they promised. Only a vanishingly small fraction, around one in a thousand, can be considered truly successful.
This is not a marginal problem. It is systemic.
As Bent Flyvbjerg, the world’s leading authority on megaprojects, has observed, these projects obey an iron law: over budget, over time, over and over again.
And yet, despite decades of evidence, the industry continues to behave as though the next project will be different.
It rarely is.
The real failure happens before execution begins
There is a persistent narrative that megaprojects fail because of poor execution, contractors underperform, risks materialise, or unforeseen complexities arise. This is comforting, because it suggests the problem sits downstream, in delivery.
But the evidence points elsewhere. Megaprojects do not fail in execution. They fail in conception.
Budgets are underestimated, often significantly. Benefits are overstated, frequently to secure approval. Risks are inadequately understood, particularly those low-probability, high-impact black swan events. Projects are treated as unique, preventing meaningful learning from past experience. And once momentum builds, early decisions become politically and organisationally irreversible, even when the underlying case has weakened.
Flyvbjerg goes further: misinformation around costs, schedules, and benefits is not incidental; it is systemic.
This combination creates a powerful distortion: not the best projects get built, but the ones that look best on paper.
Or, as Flyvbjerg describes it, an inverted Darwinism—the “survival of the unfittest.”
The missing shift: thinking right to left
What sits beneath many of these failures is a more fundamental flaw in how projects are conceived.
Most megaprojects are planned left to right. We start with an idea, develop a concept, build detail, and only later ask whether the outcome truly delivers value.
By then, it is too late.
A more effective approach, central to Flyvbjerg’s more recent work, is to think right to left: to start with the desired outcome and work backwards.
- What does success actually look like, technically, economically, and operationally?
- What must be true for that outcome to be realised?
- And what is the simplest, most robust pathway to get there?
This is not just philosophical. It is practical.
It forces realism early. It anchors decisions in outcomes, not activity. And it creates a natural discipline against optimism bias, because the end state must stand up to scrutiny before the journey begins.
The front-end is where value is created, or destroyed
If failure is rooted in the front-end, then so too is success.
Front-End Loading (FEL), the sequence of identifying, selecting, and defining a project, is not merely preparatory work. It is where the majority of a project’s value is either created or lost.
- FEL 1 (Identify) defines the outcome, what success actually means.
- FEL 2 (Select) works backwards to identify the optimal system to achieve it.
- FEL 3 (Define) translates that pathway into an executable reality, protecting value through to Financial Investment Decision (FID).
Done properly, FEL is inherently right-to-left thinking.
The uncomfortable truth is that by the time a project enters execution, most of its fate is already set.
Yet this is also the phase that is most frequently rushed, under-resourced, or treated as a boxticking exercise.
A different approach: from deliverables to decisions
Improving megaproject outcomes requires a shift in mindset.
Traditional approaches focus on producing deliverables such as reports, models, engineering packages, often with limited regard for whether they meaningfully improve decisions. The result is volume without clarity.
A more effective approach starts with a different question: what decisions matter most, and what information is required to make them well?
This is the essence of a decision-quality mindset, but it is also inherently right-to-left.
Decisions are framed not around what can we do next? but around what must be true at the end?
It prioritises:
- clarity of outcomes,
- understanding of uncertainty,
- alignment on objectives,
- and timing of decisions relative to project maturity.
In this framing, analysis is not an end in itself, it is a tool to enable better choices.
Systems thinking in a world of complexity
Megaprojects are inherently complex systems. Energy projects, in particular, involve tightly coupled interactions between technology, economics, infrastructure and markets.
Optimising individual components in isolation is not enough. In many cases, it is actively harmful.
Right-to-left thinking reinforces this.
If you start with the end state, system performance, economics, operability, you are forced to consider the system as a whole. Local optimisation becomes irrelevant; system optimisation becomes essential.
This is where multi-domain modelling and systems engineering become powerful, not as academic exercises, but as practical tools for unlocking value.
Acting like owners, not contractors
Another persistent issue in megaproject development is misaligned incentives.
When teams are rewarded for scope, effort, or deliverable volume, the natural outcome is expansion, more work, more detail and more complexity. But more is not always better.
An owner’s mindset, particularly when combined with right-to-left thinking, asks a sharper question:
What is the minimum intervention required to achieve the desired outcome?
This means:
- challenging unnecessary scope,
- focusing on what truly moves the needle,
- and eliminating work that does not improve the end state.
It is a mindset that prioritises outcomes over activity.
Protecting value is as important as creating it
Even when value is successfully created in the early stages, it is fragile.
Scope creep, generic FEED approaches, overengineering, and siloed disciplines can all erode value as a project progresses. Without a clear thread linking early decisions to later definition, the original intent is often diluted.
Right-to-left thinking provides that thread.
By continuously anchoring decisions to the desired end state, projects maintain coherence as they move forward.
This includes:
- tailoring definition to the specific outcome, not applying standard templates,
- maintaining a decision focus throughout,
- and grounding assumptions in real-world data through benchmarking and reference class forecasting.
Greater certainty, in this context, is not about eliminating uncertainty. It is about designing with it, deliberately and transparently.
Evidence that it works
When done well, this approach delivers tangible results.
System-level optimisation has demonstrated significant reductions in capital expenditure. Integrated modelling has improved schedules by months, not weeks.
And more thoughtful design of complex systems, such as energy storage or compression, has unlocked efficiencies that conventional approaches would miss.
These are not marginal gains. They are step changes.
Megaprojects don’t have to fail
The narrative that megaproject failure is inevitable is both pervasive and wrong.
Failure is not a consequence of scale or complexity alone. It is the result of how projects are conceived, evaluated, and defined.
By combining decision quality, systems thinking, and right-to-left planning, it is possible to fundamentally change outcomes.
The challenge is not technical capability. It is discipline.
Start with the end. Work backwards with intent. Protect value at every step.
And deliver with a level of certainty the industry too often assumes is unattainable.
It isn’t.
Why this matters to io
For io, this is not a theoretical critique, it is the core of what we do. We exist precisely at the point where projects are most vulnerable and most valuable: the front-end. Our role is to define the end state clearly, then work backwards to shape the optimal path to get there.
We cut through optimism bias, challenge flawed assumptions, and bring clarity where others bring volume. We are not there to validate a preferred answer, we are there to find the right one.
By combining systems thinking, decision quality, and deep supply-chain insight, we shape projects before they harden, and protect value before it leaks away.
In a world where most megaprojects fail quietly and expensively, io’s proposition is simple but uncompromising: start with the end, get the front-end right, and everything that follows has a fighting chance.
