Key Takeaways
- IPA's 2025 Industry Benchmarking Consortium data show that average schedule slip on large capital projects has roughly doubled over two decades and now sits at around 18%, with execution speed declining by approximately 20% over the same period.
- 73% of capital project teams are missing critical functions, using inexperienced staff in leadership roles, or are generally understaffed , a condition that IPA's database links to approximately 25% worse cost outcomes compared with well-staffed owner teams.
- When programmes span multiple time zones, EVM data quality degrades predictably. The US Department of Energy's EVMS Interpretation Handbook (2016) requires prime contractors to maintain reconcilable subcontractor data each reporting period - a discipline that is operationally demanding in distributed delivery models and routinely absent.
- Peer-reviewed research published in Organization Science (2024), drawing on communication data from 12,038 employees, confirms that time-zone separation produces sizable, measurable reductions in rich synchronous communication between teams - regardless of the collaboration technology available.
- Only approximately 50% of project sponsors are formally appointed with documented roles and responsibilities, according to IPA research, a governance gap that consistently correlates with worse delivery outcomes.
The structure of Energy, Minerals and Resources (EMR) capital project delivery has changed materially over the past two decades. A project sanctioned by an operator headquartered in London or Singapore is routinely designed in an engineering hub in Mumbai or Chennai, procured across multiple jurisdictions, and constructed on a remote site in West Africa, the Pilbara, or northern Canada. The owner team, the EPC contractor, and the subcontractors may collectively span eight or more time zones.
This is not an edge case. The IEA reports that upstream oil and gas investment reached US$570 billion in 2024, with NOCs in the Middle East and Asia accounting for almost the entire year-on-year increase. The IEA's Global Critical Minerals Outlook 2024 estimates that approximately US$590 billion in new mining capital investment will be required by 2040, with production increasingly concentrated in jurisdictions where owner organisations have limited operating history. Wood Mackenzie recorded nine LNG final investment decisions in 2025, totalling approximately 72 mtpa, distributed across North America, the Middle East, and East Africa.
The scale of distributed delivery is not the problem in itself. The problem is that governance structures calibrated for co-located teams have not adapted to reflect the operating model that has become standard across the sector.
Why Does Distributed Delivery Create Specific Governance Problems?
The term "distributed team" is often treated as a logistics question, a matter of collaboration tools and time-shifted meetings. The evidence positions it as a structural governance problem.
McKinsey's 2022 review of 532 capital projects - 62% of which were megaprojects above US$1bn - found average cost overruns of at least 79% against initial budgets and average schedule delays of 52%. A separate McKinsey analysis of more than 300 billion-dollar-plus megaprojects found average overruns of approximately 80%. EY's study of 192 mining and metals projects found that 64% ran over budget, over schedule, or both, with an average cost overrun of 39%. These outcomes are not explained by technical difficulty alone. McKinsey's analysis of 48 troubled megaprojects attributed cost and time overruns primarily to poor execution in 73% of cases - and in distributed programmes, poor execution has a specific character that is now empirically documented.
Where Do Distributed Capital Programmes Actually Fail?
Three failure modes appear consistently across the research literature on distributed capital governance.
Owner-team understaffing. IPA's project-team research finds that 73% of capital project teams are missing critical functions, using inexperienced personnel in leadership positions, or are generally understaffed. The performance consequence is quantified: teams that cede cost estimating, scheduling, and project controls to an EPC contractor - without maintaining owner-side capability to independently challenge that data - deliver projects at approximately 25% worse cost competitiveness on average. Strong owner teams deliver projects at an average 25% lower cost than contractor-led or understaffed counterparts, per the same IPA research. In a distributed programme, understaffing is compounded: the owner's capacity to challenge contractor-reported data is reduced at exactly the point where that challenge is most needed.
EVM data integrity. Earned Value Management is the standard instrument of capital governance in the EMR sector. It is also fragile in multi-contractor, geographically distributed environments. The US Department of Energy's EVMS Interpretation Handbook (2016) is explicit: "failure to integrate data reported in subsystems invalidates the usefulness of reported earned value information." Prime contractors are required to maintain reconcilable subcontractor EVM data, with accounting differences explained each reporting period. When engineering is in one country, construction management in another, and the project site in a third, that reconciliation discipline - straightforward in a single-site team - becomes operationally demanding. IPA's 2016 project-controls survey found that more than 50% of project-controls specialists report their organisations are inadequately resourced to perform this function, and that the industry's Project Controls Index has remained in the "Fair" band since 2004 with no material improvement. The board pack's EVM trend lines may be more optimistic than the underlying data warrant.
Reporting latency and time-zone effects. A 2024 study published in Organization Science, drawing on communication data from 12,038 employees across a large multinational, found that temporal distance produces sizable and measurable reductions in rich, synchronous communication between teams - even when employees adjust working hours to compensate. A separate 2024 paper in the International Journal of Project Organisation and Management, replicating an earlier study conducted a decade prior, concluded that temporal distance still materially affects perceived decision quality in distributed project teams, regardless of current collaboration technology. Weekly reporting cycles designed for a co-located team are structurally insufficient in a programme spanning eight or more time zones: by the time a delay signal reaches the owner board, the cost consequence of responding has typically grown.
What Does Rigorous Governance Require Across Time Zones?
The evidence does not support governance-by-dashboard as the primary response. McKinsey's case evidence for digital control towers - including a documented saving of more than US$75 million on a single petrochemical project - explicitly conditions those results on a sound operating model and disciplined process change. Technology applied on top of a broken governance model produces better-formatted inaccurate data.
What the evidence does support is a set of non-negotiable owner-side capabilities that distributed delivery makes more, not less, critical:
- Formal sponsor appointment. IPA finds that only approximately 50% of project sponsors are formally appointed, and fewer than half have received documented roles and responsibilities. IPA's database consistently links this gap to worse delivery outcomes. In a distributed programme, where the project director may be managing teams across five time zones, an undocumented sponsorship structure leaves accountability gaps that are difficult to close after authorisation.
- Owner retention of the project-controls function. Per IPA, cost estimating, scheduling and project controls must remain under owner authority - not delegated to the EPC contractor. This is the only reliable basis for independent challenge of contractor-reported EVM data. It is also the function most commonly absent when programmes encounter trouble.
- Independent project assurance. Where distributed delivery introduces reporting latency and reduces synchronous oversight, an independent assurance function with direct access to the board - not routed through the project director - provides the verification layer that physical co-location previously made implicit. One global energy company used this approach to revamp its entire US$50 billion capital portfolio, per McKinsey's 2024 analysis.
- Stage-gate discipline at board level. IPA's IBC 2025 data find that only 16% of projects in their sample entered authorisation with "Best Practical" Front End Loading. As covered in more detail in our post on what boards should be asking about their capital projects, boards that sanction before FEL has reached at least IPA's "Good" standard are accepting governance risk before a single contractor has been engaged, and in a distributed programme, that risk compounds rapidly once execution begins.
For programmes of significant scale, a structured digital governance system - integrating project-controls data across the full lifecycle - can provide the single source of truth that distributed delivery requires. The qualifying condition, consistently documented in the literature, is that the system must sit on top of a sound operating model: a properly staffed owner team, a formally appointed sponsor, and an independent challenge function with genuine authority.
The Governance Gap Is Structural, Not Incidental
IPA's most recent IBC data, published in June 2025, conclude that the industry has lost approximately 20% in execution speed over the past two decades, and that average schedule slip has roughly doubled. That deterioration has occurred over the same period that distributed delivery has become the default model in Energy, Minerals and Resources EMR capital programmes, and over the same period that digital collaboration tools have been most widely adopted.
The data are consistent across IPA, McKinsey, EY, and the academic literature: distributed delivery does not inherently produce worse outcomes, but governance structures calibrated for co-located teams consistently fail when applied without modification to a distributed operating model. The specific failure modes - owner understaffing, EVM data degradation, and reporting latency - are empirically documented, quantified, and addressable. The question for boards and project directors is whether the governance model currently in use was designed for the operating model actually being deployed. The research suggests that, in most cases, it was not.
For a deeper look at how these failure modes connect to risk management in mega-projects, the principles are closely related.
PDAS provides independent governance and assurance for Energy, Minerals and Resources (EMR) capital programmes, across geographies, time zones, and delivery phases. If your programme governance was designed for a co-located team, we can help you assess what needs to change.
