Why Projects Go Over Budget: The 7 Root Causes of Cost Overruns

The project starts with a budget that was never realistic. Either the estimator lacked historical data, used methods unsuited to the project's complexity, or was under pressure to present an optimistic number. Early estimates are wide by design, and the industry classification says so explicitly: under AACE International's Recommended Practice 18R-97, a Class 5 concept-screening estimate (0 to 2% of the project defined) carries an expected accuracy range of -20% to -50% on the low side and +30% to +100% on the high side, narrowing to -10% to -20% and +10% to +30% only at Class 3, where 10 to 40% of the project is defined. Approving a Class 5 number as though it were a Class 3 budget books the overrun before any work starts. The fix: reference class forecasting, three-point estimation (optimistic / most likely / pessimistic), and mandatory independent estimate review before budget approval.

Requirements expand incrementally during execution. A 'small change' requested by a stakeholder adds a day here, a week there, and is never reflected in the budget. Without a formal change control process, scope grows while the budget stays fixed. Scope creep is particularly severe in IT projects, where requirements are documented late or ambiguously, and in government projects, where political interference adds features throughout delivery.

Budgets set without contingency reserves fail at the first problem. Contingency should be scaled to how much uncertainty the estimate actually carries, not set at a habitual round percentage. The UK's published benchmark for that scaling is the Green Book optimism-bias uplift, which is 24% of capital cost for standard buildings at outline-business-case stage, 44% for standard civil engineering, and up to 200% for equipment and software development. When a risk materialises (a subcontractor fails, material prices spike, or a technical problem requires rework) and no buffer was set aside, the overrun is immediate.

Errors in design discovered during construction require expensive rework: demolition, replacement, re-inspection, and programme delays. This is particularly common in fast-track projects where design and construction phases overlap. If foundation drawings are wrong after concrete is poured, the cost to fix can exceed the original design contract value. Professional indemnity insurance responds in theory, but claims are slow and disputed.

Material price spikes, long lead times, and subcontractor insolvency have driven significant project overruns since 2020. Timber, steel and electrical components all ran through sharp price cycles in 2021 and 2022, and a budget fixed before the cycle absorbs the whole of it. Even with fixed-price contracts, contractors have sought relief under force majeure or material escalation clauses. This cause is particularly hard to manage and often not covered by standard construction insurance.

Key personnel leaving mid-project, difficulty hiring specialists, or outsourced teams underperforming all add cost. In IT projects, developer turnover means knowledge transfer, onboarding time, and rework. In construction, skilled trades shortages have extended programmes significantly. Staff augmentation through consultants and contractors is expensive and reduces institutional knowledge.

PMI's 2013 Pulse of the Profession in-depth report, The Essential Role of Communications, put the figure precisely: of the $135 million at risk for every $1 billion spent on a project, $75 million (56%) is put at risk by ineffective communications. When project sponsors, contractors, and stakeholders don't share accurate status information, problems are hidden until they are unmanageable. Programme delays are not reported. Cost variances are buried in monthly reports. The project looks on track until it suddenly isn't. This is a governance failure as much as a communication one. Regular, honest EVM reporting with CPI and EAC prominently displayed is the counterweight.