In this article
In this article
Capital projects are critical to the success of executives, leaders, workforces, and their organizations too. Capital projects involve large investments or upgrades to physical assets such as infrastructure, facilities, equipment or technology that can have a significant impact on the organization’s operations and performance. By understanding how to approach capital project types and their intake, evaluation, prioritization, selection, and allocation; leaders will achieve their strategic goals.
These strategic goals could include increasing efficiency, expanding its product offerings, entering new markets, or improving its competitive position. Unfortunately, capital projects frequently fail to meet their schedules and overspend on their budgets, let’s see why capital projects fail from credible sources.
The scale of that failure is documented rather than anecdotal. Bent Flyvbjerg’s research on megaprojects found that nine out of ten have cost overruns, that overruns up to 50% in real terms are common and overruns above 50% are not uncommon, and that the pattern has held for the seventy-year period for which comparable data exist. He calls it the iron law of megaprojects: over budget, over time, over and over again.
McKinsey research, from 2016, puts the same problem in operational terms, finding that large capital projects typically take 20% longer to finish than scheduled and cost up to 80% more than expected. Failure at that rate is not caused by construction error. How capital projects fail is a question about decisions, not construction: failure is settled earlier, in how projects are defined, selected, executed and validated, which are the four stages this article works through.
Here we discuss the four main reasons capital projects fail so that you can identify and avoid pitfalls with your capital projects:
Failure is decided at these four points rather than during construction: how the project is defined, whether it should have been selected, how it is executed, and whether anyone checked the result.
1. Poor Project Definition: Scope, Schedule and Budget
How much definition is enough is a measurable question rather than a matter of judgement. AACE International’s Cost Estimate Classification System, set out in Recommended Practice 17R-97, classifies capital cost estimates from Class 5 to Class 1 according primarily to the maturity of project definition, while recognizing that expected accuracy varies with project-specific uncertainty and risk. This means an early Class 5 estimate should not be treated as though it were a definitive control baseline: as scope, quantities and risks become better defined, normal estimate development can otherwise be reported as an apparent cost overrun. The remedy is to stage the definition: the depth of definition demanded at each gate is what fixes the estimate class, which is the subject of FEL 0 to FEL 3.
Under-Analyzed
Undertaking poorly defined projects is an obvious risk. The motivation, objective and required investment of time and money in an initiative require definition to even assess success or failure. Nevertheless, in practice this happens all too often. Investments are made in very roughly defined pet-projects to the perplexed incredulity of other stakeholders. We did what? With no yardstick to measure the inputs or outcomes, it can be assumed that a significant proportion of these investments would have incurred an excessive commitment of resources for the benefits produced.
Over-Analyzed
By contrast, over analysis can also be a cause of failure. Too much time is invested designing and estimating bad ideas. Because of ‘analysis-paralysis’, over-analyzed projects start too late. Deferring benefits. Or worse: projects are delivered too late to mitigate the operating risk they were intended to address. A key indication of over-analysis is when investment proposals over-emphasize the ‘what’ and the ‘how’ rather than the more important ‘why’.
Under-Estimated
Rosy ‘happy-day’ estimates help get projects over the line. Often with catastrophic consequences. Suppliers get forced to deliver to unrealistic constraints and quality suffers. Or work is abandoned. Insufficient internal resources are allocated for effective design, validation or change management. Consequently, projects are delivered but not adopted due to the lack of preparation and training. Or operating outcomes materially fail to deliver the expected benefits. Avoidable risks are not fully mitigated with dramatic impact.
Project ‘contingency’ funding is too often fully utilized, and not reserved for truly un-predictable events. An estimate is only as reliable as the definition beneath it. So release funds in proportion to how well the scope is understood. At the earliest classes, approve only enough to get the definition to the next gate. The remedy for a habitually under-estimated portfolio is a later funding gate, not a tighter number.
Over-Estimated
This is when estimates are too generous. Too much funding contingency is provided. Too many scarce resources are assigned. The project is comfortably delivered. But at what cost? Over-investing in critical project A, may mean that just as important projects B and C are not delivered. As stewards of an organization’s resources, executives are compelled to seek maximum value without fear or favor.
Unspecified Results
Starting any endeavor without a plan may achieve something. But normally this course of action leads to sub-optimal results. Even if benefits are derived, could more have been achieved? Could they have been achieved for less? Assigning corporate resources to a project manager without any specification of outcomes is a dereliction of executive responsibility and very likely to be a primary cause of failure.
Over-Specified
Expecting to have a very detailed specification at the start of a project is also, however, unrealistic. Or worse: it leads to sub-optimal outcomes. Design is a phase of a project: it is not done effectively or efficiently prior to project initiation. Over specifying a solution is indicative of a poor project delivery methodology. The project team should be given clear targets and reasonable allocation of initial funding. Project delivery methodologies should actively mitigate technical and commercial risks early in the project lifecycle: this helps refine forecasts to complete and expected benefits on realization.
Provided a project continues to indicate high value return from future investment it should be progressed. If not, it should be halted as soon as possible. Therefore, over-investing in all aspects of design too early is inefficient if the project is never going to achieve completion. Modern project delivery methodologies employ agile principals including iterative risk-mitigating cycles to ensure prioritized design and delivery of critical features first, acknowledging that some features may never be designed or delivered as more important requirements emerge and investment thresholds are reached.
Started Too Soon
Projects that seek to replace assets or mitigate compliance or other commercial risks too soon are a more subtle form of failure. Whilst these projects may be successful in their intended purpose, again the question is ‘at what cost’? Given resource constraints it is important that projects that can be safely deferred, are deferred. This allows more critical projects to commence sooner. This risk of committing resources too early is hard to manage, as most sponsors are conservative by nature and err on the side of caution.
Delivered Too Late
Of course, delivering projects too late is possibly even more ruinous. In practice, organizations get the timing ‘about-right’ not through effective prioritization but application of excessive resources to accommodate parallelization. Because projects are inappropriately prioritized, they tend to start too soon, and then take too long as resources are stretched too thinly. This incurs excessive cost and delivery risk.
2. Misaligned Portfolio Selection
What systems do you have in place to choose the right capital projects and reject the wrong ones? Here are some of the reasons capital projects make the cut when perhaps they shouldn’t have.
Selection failure has a signature you will recognize. Score projects on several dimensions with no agreed weighting, and the ranking cannot be reproduced. Hand the same portfolio to a different committee and you may well get a different answer. That is why consistent classification and a fixed scoring model do more for portfolio quality than the most persuasive business case in the room. Getting from why the wrong projects get chosen to how a defensible ranking is produced is a portfolio management problem, and we set out the method in our guide to how to prioritize capital projects.
Undue Influence
In many organizations, to get a project approved, it’s not ‘what you know’, it’s ‘who you know’. The power of influence stems right from the top with executives pushing pet-projects. The more capital funding approval gets devolved, the more pet-projects spring-up. It’s an unavoidable human characteristic, that we’re unduly influenced by persuasive argument. And with a lack of high-quality and comparable data, the only basis for a decision will be persuasive argument. The slicker the presentation, the more astute the pitch, the greater the likelihood of human-bias and sub-optimal projects being selected.
Poor Quality Information
Executives will rely on their intuition, experience, and persuasion when the information is presented is unreliable. Information is poor quality if it is incomplete, misleading, invalid, or inaccurate. This may be because of delays in collation, errors in preparation and transcription, or distortion. A common example of misleading information is the reliability of estimates. Often costs or benefits are presented as a single best-case number. In reality, a broad range of possible outcomes is possible, and failure to represent this information effectively can result in executives making inappropriate choices.
Incomparable Options
A key requirement for executives to fairly assess a list of possible investment needs and opportunities is to have reliable bases for comparison. Trying to compare replacement and growth initiatives is not practical, for example, as replacements projects are primarily motivated by risk to operations, whilst growth opportunities are primarily assessed on net present value. When executives are not provided relevant classifications (such as the investment reason), or the comparison metric is not reliably determined (e.g. inconsistent discount rates applied), it is very likely that the project selection will be inadvertently biased and therefore sub-optimal.
Too Many Competing Criteria
3. Capital Project Execution
Even well-defined, appropriately selected capital projects can fail in the execution phase.
Capital project execution is where overruns become visible, not where they are created. Every failure mode below is a planning, monitoring or coordination gap, which means the project was off-plan for some time before it was reported as off-plan. The practical consequence is that periodic reporting against an approved budget cannot protect a capital project. Only continuous forecasting against approved budget can, because it surfaces the deviation while there is still scope to act on it.
Overloaded People
To err is human. Through lack of expertise, time, or motivation, people don’t always do the right thing. One of the most critical people involved with project execution is the project manager. In many organizations, this is a scarce resource, and individuals are often overloaded and lack the capacity to ensure project success. Business resources are essential to define requirements and validate outcomes. They too are often under extreme pressure with business-as-usual activities and are often not backfilled to ensure effective engagement and change management. Supplementing internal people with external contractors may fail as they may be more aligned to their own daily rate and contract duration than the business outcomes.
Technical Risk
Contemporary projects typically involve the introduction of new technology. This introduces the inherent risk of technical failure in the organization’s unique operating environment. Conversely, failure to adopt new technology may incur additional cost or competitive disadvantage. Projects fail because technical risks are not actively addressed throughout the project lifecycle.
An Immature Delivery Process
Projects fail in the execution phase because of a lack of process to mitigate inherent human and technical risks. The art of effective project delivery is to establish an effective methodology for managing these risks effectively. Where project execution methodologies are poorly defined or immature, inherent risks are more likely to eventuate.
No Ongoing Planning
The most critical responsibility of project managers is to actively plan the execution of projects, and apply mitigating actions when deviations occur. Projects that proceed without ongoing planning of activities and resource consumption are doomed to failure.
Unmonitored Variance
A plan is only effective if it is monitored. Adverse variations to plan indicate the need for mitigating action. Projects fail when management do not have reliable cost, schedule, scope, risk and other Key Performance Indicators to monitor, or do not action exceptions when they occur.
Uncoordinated Dependencies
Large scale capital projects are often related and conducted in pursuit of defined strategic goals and objectives. There is inter-dependency of deliverables and resources. Lack of co-ordination of these projects therefore can be a cause of failure.
The Wrong Suppliers Engaged
The key external facing activity of project execution is procurement. There is invariably more than one option for solving a problem or addressing an opportunity, and alternative solutions may be proposed by various suppliers. Projects can fail when the wrong suppliers are engaged. This can be equally a result of undue reliance on incumbents, or ineffective qualification of new partners.
4. Weak Project Validation and Accountability
Project sponsors are subject to a moral hazard when they are not held accountable for the promises contained in their business cases. Most business cases are a finally balanced trade-off of risk and return. Slightly understating the one, or overstating the other, may be all that’s needed to get approval to proceed. Therefore, a common reason for projects failing to deliver to expectation is that outcomes are not formally assessed, responsible parties are not held accountable, and practices are not improved.
Validation is the failure mode that compounds. A capital project that is never reviewed after completion cannot inform the next estimate, so the definition error described in section one repeats at the next funding round with the same confidence attached to it. It is the reason why capital projects fail repeatedly in the same organization for the same reasons. Post-investment review is therefore part of the capital process itself, not an assurance activity bolted on outside it.
No Post-Investment Reviews
To determine the success or failure of a capital project it is necessary to conduct a post-implementation review. Costs are easy to measure. But on their own are not a determinant of failure. Capital projects that come within budget can be failures. Capital projects that cost double their initial estimates may be deemed successful. Because success or failure depends on the benefit derived. Delivering a capital project on budget that fails to achieve its objectives is failure. Capital projects that deliver a significantly higher net present value, even at twice the investment, are a success. Unfortunately, assessment of benefits is much harder than measuring costs.
Capital investments, by their very nature, are long-term in nature. Or mitigate risk which in practical terms may be difficult to measure. This may require independent reviews over many years. Nevertheless, failure to conduct a post-implementation review is likely to encourage, or hide, project failures.
No Accountability
Capital projects are very likely to fail if they are nobody’s responsibility. Many individuals are typically involved in a project definition, selection, execution, and review. These individuals, their roles and their contributions, comments, endorsements, and approvals are intended to help deliver success. Failure to retain these records will increase the risk of dereliction of responsibility, which ultimately may lead to an increase in adverse project outcomes.
No Improvement
No successful organization or entrepreneur has succeeded without failures along the way. A complete lack of failures may indicate that the project portfolio has become too timid and conservative. Failures present an invaluable opportunity to learn. Ongoing, and disproportionate project failures, are, however, more likely when these learning opportunities are not acted upon. This may require adjustments to the way your projects are defined, selected, executed, or validated. As Henry Ford is supposed to have said: “The only real mistake is the one from which we learn nothing”.
How Capital Projects Fail in Practice
How capital projects fail is easiest to see where the cost and schedule outcomes were independently reported. The four failure modes above are all visible in public-record projects, and each of the following failed at a different stage, which is the point. There is no single cause to design against.
Berlin Brandenburg Airport: A Definition Failure
Scheduled to open on 30 October 2011, Berlin Brandenburg Airport opened on 31 October 2020, nine years and a day late. The project broke ground in 2006 on a cost assessment of about €2.83 billion, and reported spending passed €7.3 billion before it opened. The decisive failure was the fire safety system. The design was bespoke, it was never certified, and more than 20,000 defects were logged against the fire detection and emergency lighting alone. Construction had started before that design was resolved: an over-specified solution imposed on an under-defined scope, which is the combination described in section one.
Sydney Light Rail: A Procurement and Selection Failure
The CBD and South East Light Rail business case summary of November 2013 estimated A$1.6 billion. By the time the main works contract was signed in December 2014 the budget had risen to A$2.1 billion, and the NSW Auditor-General found that A$517 million of that A$549 million increase was mispricing and omissions in the business case, not scope change and not negotiation. The revised figure announced in November 2019 was A$2.9 billion, and a follow-up audit put the total above A$3.1 billion. There were execution problems too. A contractor dispute cost the state a further A$576 million to settle in 2019. But the damage was done at selection: the project was approved on numbers that could not hold, and almost all of that first gap was sitting in the business case before anyone broke ground.
Snowy 2.0: A Failure to State the Range
Announced in March 2017 with an indicative capital cost of about A$2 billion and a four-year build, Snowy 2.0 was reassessed by a feasibility study released before the year was out at A$3.8 to A$4.5 billion. The figure nearly doubled before construction began, on definition work alone. A 2023 reset put the cost at A$12 billion with completion moved to December 2028. By March 2026 actual spend had reached A$11.1 billion, and in June 2026 the Australian National Audit Office found significant deficiencies in the project’s management, including no agreed baseline schedule. A further cost reassessment is still to report. The original estimate being wrong was never the problem. A pre-feasibility number is expected to be wrong, within a stated range. The problem was that it went out as a commitment with no range attached.
Read across the three and the pattern holds. Not one of them failed on construction alone. Each was already locked into an outcome it could not deliver before delivery began. That is why classification and staged definition of capital project types shape the final number long before execution discipline gets a say.
How to Reduce Capital Project Failure
As we conclude this article, covering the four key reasons why capital projects fail, it would be remiss not to offer a solution to the problem. Capital project failure is related to the way projects are collected, defined, ranked and selected. Get this right and your project portfolio will align to organizational objectives with consideration to human resource and budget constraints. Understanding how capital projects fail is only useful if the process changes as a result. Stratex Online is capital planning and CapEx Management software built around that sequence: demand, business case, portfolio selection, expenditure request and forecast, so that the reasoning behind each approval is retained and available at post-investment review rather than reconstructed from email.
