The Human Factors Behind Project Failures

Understanding the human architecture of a project team is not a soft intervention, it is a critical governance issue. The factors which determine whether a project succeeds or fails are human, but they are measurable, and they are addressable…if you know where to look.

The timeline was reasonable and the engineering was sound. The budget had contingency built in and yet, somewhere between kick-off and the first major milestone, something went wrong…

This is not an unusual story. It is, in fact, the dominant story of large-scale projects worldwide.

Bent Flyvbjerg, professor at Oxford and arguably the world's leading authority on megaproject performance, has spent decades cataloguing why ambitious initiatives underdeliver. His research, synthesised in How Big Things Get Done, analysed thousands of projects across industries and continents. The conclusions are simple; the overwhelming majority of major projects run over time, over budget, or both. Not because of technical problems, but because human problems were ignored.

The Forecasting Trap

One of Flyvbjerg's most important observations is the prevalence of the “planning fallacy”, a term originally coined by Daniel Kahneman and Amos Tversky, the near-universal tendency for project teams to underestimate costs, overestimate benefits, and dismiss the relevance of comparable past experience. Teams anchor their forecasts to the project in front of them, treating it as singular and special, based on optimism rather than evidence.

The Heroic Narrative Problem

There is a tangential kind of project failure that begins not with a miscalculation but with a performance. A senior stakeholder wants the initiative to look bold, the timeline gets compressed to signal ambition, the budget gets trimmed to demonstrate fiscal discipline, the scope expands because saying no to an influential voice carries its own risks. The plan, in other words, is shaped by what is commercially desirable rather than what is achievable. A more charitable reading is that many projects would never get started if their true costs were known upfront, but the bottom line is that when the people responsible for delivering a project are also responsible for selling it, the incentives are misaligned from the start.

James Reason, the cognitive psychologist best known for his work on human error, described organisations which suppress inconvenient signals as “pathologically silent”; systems in which the information needed to avert failure exists somewhere but never reaches the people with the authority to act on it. Stakeholder pressure is one of the most reliable mechanisms for producing this.

The Commitment Escalation Problem

Once a project has momentum, something psychologically powerful kicks in. Behavioural scientists call it “escalation of commitment”, the tendency to pour further resources into a failing course of action because of what has already been invested. In project environments, this manifests as a kind of institutional stubbornness. Early design decisions calcify into assumptions, and assumptions calcify into constraints. Critical questions stop being asked, not because the answers are known, but because asking them has become uncomfortable. The project becomes something to defend rather than interrogate.

We call these contextual or situational biases. The advent of sophisticated AI systems for project forecasting, scheduling and execution can help. They expose unrealistic forecasts, challenge unreasonable assumptions and surface the future implications of mistakes or delays which have been kicked down the road. These tools strip decisions of their immediate context and place them in the frame of the entire project, along with all their interactions and knock-on effects.

Different Minds, Different Rhythms

However, beyond these biases lies a deeper problem: The interpersonal biases and behavioural preferences which shape day to day communication and decision making. These are not a function of context, but of the sheer diversity of cognitive and behavioural styles that any serious project team will contain.

Some people reason from data and reach conclusions cautiously, while others synthesise rapidly and act on instinct. Some need extensive consultation before committing to a decision; others experience that consultation as delay. Some thrive on structured process; others find it suffocating. Some manage conflict directly; others find directness aggressive.

None of these preferences is inherently superior. But when they are unacknowledged, they create friction which compounds over time; misread intentions, eroded trust, decisions that feel imposed rather than shared.

Edgar Schein, the MIT organisational psychologist who spent decades studying culture and leadership, argued that the deepest and most consequential organisational dynamics are the ones that remain tacit, the unspoken assumptions about how decisions should be made, what counts as good communication, who is expected to lead and who to follow. Left unexamined, these assumptions inevitably collide.

Human Architecture as a Critical Governance Issue

Making the intangible human factors visible is not a soft intervention. It is a critical governance one.

The organisations which consistently deliver complex projects well tend to share a characteristic that goes beyond rigorous planning or experienced leadership. They have invested in understanding how their teams actually work, not how they are supposed to work on paper, but the genuine patterns of communication, decision-making, and conflict resolution which operate beneath the surface.

This means creating a shared language for behavioural difference, one that allows a team to discuss how it makes decisions, resolves disagreements, and navigates uncertainty without these conversations becoming personal or charged. When this language exists, misalignment surfaces early, while it can still be addressed. When it doesn't, it surfaces late, in the form of escalation, attrition, and failure.

Flyvbjerg's research shows that the projects most likely to succeed are those that take the longest to plan and the shortest to execute, those that front-load the hard thinking rather than discovering problems mid-delivery. Understanding the human architecture of a project team is part of that front-loading. It is not a cultural nicety, it is risk management.

The Gantt chart will always have its place, but the factors which really determine whether a project succeeds or fails are the ones it cannot represent: the conversation that didn't happen, the assumption that went unchallenged, the team that never quite understood itself well enough to work as one.

These factors are measurable. And they are addressable…if you know where to look.

MindAlpha's People Behind the Racks Project was created to help organisations understand their behavioural dynamics across leadership, communication, planning, decision-making, execution and adaptation, creating the shared language complex projects depend on.