Minimum Viable Governance for Complex Projects
Complex projects do not fail because they have too little paperwork. They fail when decisions arrive late, risks stay hidden, authority is unclear, and controls give leaders confidence without giving teams direction.
That is the problem minimum viable governance is designed to solve.
In oil and gas, EPC, construction, energy, infrastructure, mining, utilities, and marine operations, governance is essential. No serious capital project can run on trust alone. The scale is too large, the exposure is too high, and the number of interfaces is too great. Owners need assurance. Contractors need clear authority. Lenders, regulators, partners, and boards need evidence that decisions are controlled and risks are being managed.
The difficulty starts when governance grows faster than the project’s actual risk profile.
A project can reach a point where every change needs several approvals, every report repeats information already submitted elsewhere, and every stage gate asks for evidence that adds little to the decision. Teams spend more time proving control than exercising control. Senior leaders become document reviewers instead of decision-makers. Project controls become an administrative burden rather than a management system.
Minimum viable governance is not weak governance. It is disciplined governance. It keeps the controls that protect value, safety, compliance, and accountability. It removes the controls that create delay, duplication, and false comfort.
Governance Becomes A Problem When Control Work Replaces Control Thinking
Project governance should answer a few practical questions.
Who is authorized to decide?
What evidence is needed?
Which risks require escalation?
What must be assured before the next commitment?
Who is accountable after the decision?
When those questions are clear, governance supports delivery. When they are buried under templates, committees, review cycles, and repeated status packs, governance starts to work against the project.
The warning signs usually show up long before leaders acknowledge them.
Decision papers circulate for weeks because every function wants a comment, but no single person owns the decision. Change requests wait for a board meeting even when the cost and schedule impact sit well within delegated authority. Stage gates focus on document completeness rather than whether the project is ready to commit capital, mobilize resources, or accept operational risk.
Reporting becomes another common failure point. A major project may produce weekly dashboards, monthly owner reports, steering committee packs, lender updates, internal controls submissions, contractor performance reports, and assurance summaries. Each report may have a valid audience. Yet many draw from the same source data, ask similar questions, and consume the same project controls capacity.
The result is not better project governance. It is slower decision-making and weaker attention on the work that needs judgment.
For complex project delivery, this matters because delay is rarely neutral. Slow approvals can hold up procurement. A delayed variation decision can affect subcontractor productivity. A gate review that asks the wrong questions can push risk forward while giving sponsors the impression that the project has passed a meaningful test.
Governance should create confidence because decisions are informed and accountable. It should not create confidence because the project has produced a large volume of paper.
Minimum Viable Governance Keeps The Controls That Matter Most
A lean governance model starts by separating essential controls from inherited habits. The aim is not to make governance lighter for its own sake. The aim is to make it proportionate to risk, value, timing, and consequence.
A minimum viable governance model usually includes four elements.
Clear Authority
Delegated authority must be specific enough to support fast decisions. A delegation matrix that looks complete but still requires informal executive approval for most meaningful decisions does not help delivery.
Good delegated authority defines:
Financial approval limits
Commercial approval rights
Schedule and milestone decision rights
Technical concession authority
Contract change approval levels
Risk escalation thresholds
Emergency decision routes
The test is simple. If a project manager, package manager, contract manager, or project director cannot quickly identify who can approve a decision, the governance model is not yet clear.
In major capital projects, authority should also recognize timing. Some decisions lose value if delayed. A procurement substitution may need input from engineering, quality, operations, and commercial teams, but the approval path must match the urgency and risk. Holding every decision for a senior committee can create more exposure than it avoids.
Meaningful Accountability
Accountability is stronger when fewer people can hide inside the process.
Many governance systems create shared review rights without clear ownership. Several groups review a risk, a change, or a gate submission, but no one owns the outcome. This leads to late objections, repeated comments, and defensive behavioundergo the same level of review as a change that affectsr.
Minimum viable governance assigns accountability at the right level. It distinguishes between:
The person who recommends
The person who reviews
The person who approves
The person who owns the result
That distinction matters. A project controls lead may provide schedule impact analysis. A commercial lead may assess contract exposure. A technical authority may confirm engineering acceptability. The accountable decision-maker must then decide, record the basis, and carry the consequence.
Risk-based Assurance
Not every work package deserves the same governance weight.
Risk-based governance directs assurance effort towards decisions and activities with the highest potential impact. These may include major contract awards, critical path changes, design maturity, safety-critical interfaces, commissioning readiness, regulatory commitments, long-lead equipment, marine logistics, or brownfield tie-ins.
A low-risk document revision should not face the same level of review as a change affecting operating integrity or project economics.
Risk-based assurance also changes the nature of stage gates. A gate should not exist simply because the calendar says it is time. It should confirm that the project is ready to move into a higher level of commitment. That may mean approving funding, entering fabrication, commencing offshore work, energizing systems, or handing over to operations.
Reliable Management Information
Good governance depends on good information, but more reporting does not always mean better information.
The best project controls reports support decisions. They show movement, variance, forecast, risk, and required action. They don't just list activities completed in the reporting period.
A useful governance pack should help leaders answer:
What has changed since the last review?
Which decisions are needed now?
Which risks have increased?
Which assumptions are no longer valid?
Where does the forecast differ from the approved baseline?
What support does the project need from sponsors or functional leaders?
If a report doesn't support a decision, trigger action, or provide required assurance, challenge it.
Where Project Controls Often Become An Obstacle
Project controls should help leaders manage cost, schedule, risk, scope, change, and progress. In mature organizations, the project controls function often becomes the main source of delivery intelligence.
The problems start when controls become detached from decisions.
Duplicate Reporting
Duplicated reporting is one of the clearest signs of inefficient governance. Teams may rework the same data into different formats for separate forums, each with slightly different commentary and status language.
The project team may then spend days reconciling differences caused by reporting format rather than actual project movement.
A better approach is to create a single source of controlled project data, with tailored views for different governance bodies. The board may need exception reporting and major decision points. The delivery team may need package-level trends and blockers. The assurance team may need evidence of compliance and risk treatment. These are different views, not different truths.
Excessive Stage Gates
Stage gates can protect major commitments, especially in capital-intensive sectors. They help confirm definition, readiness, funding, technical maturity, and operating alignment.
Yet stage gates lose value when they become ceremonial. If the outcome is already assumed, or if the gate focuses mainly on whether forms are complete, the process adds cost without improving control.
A useful gate has a real decision attached to it. The decision may be to proceed, stop, recycle, reduce scope, change execution strategy, release funds, or delay commitment until defined actions close.
For high-risk projects, the gate may need deep assurance. For lower-risk projects, a focused readiness review may provide enough control. The governance model should allow that distinction.
Approval Processes That Do Not Match Risk
Many organizations apply the same approval process to all changes above a low financial threshold. This can overload committees and slow down minor decisions while still failing to focus on changes with larger strategic impact.
Financial value matters, but it is not the only measure of risk. A relatively low-cost change can affect safety, start-up, permits, operability, or the critical path. A higher-cost change may be commercially straightforward if it falls within contingency and doesn't affect key milestones.
Approval processes should consider cost, schedule, technical risk, contractual exposure, safety, compliance, and stakeholder impact. This prevents low-risk decisions from clogging the system and high-risk decisions from being treated as routine.
Governance Forums Without Decision Discipline
A governance forum should have a clear purpose. Some forums decide. Some review. Some advise. Some assure. Confusion between these roles creates delay.
If a steering committee only receives updates, it should not be labeled a decision forum. If a risk committee can escalate but not approve treatment actions, its limits should be clear. If a change board can approve budget but not schedule relief, the process must state where schedule decisions go.
Decision discipline also means recording what was decided, who approved it, what assumptions were used, and what actions follow. Without that record, the same issue often returns in a different form.
A Practical Model For Minimum Viable Governance
Building a minimum viable governance model is not an academic exercise. It starts with the current governance burden and tests each control against its purpose.
This model works well for complex projects where the challenge isn't the absence of governance, but too much low-value governance.
Governance area | Minimum viable test | What to remove or reduce |
Authority | Can the right person approve the decision at the right time? | Shadow approvals, unclear escalation, informal veto points |
Assurance | Does the review address material risk before commitment? | Checklist reviews with no decision consequence |
Reporting | Does the information support action or assurance? | Repeated status packs, cosmetic formatting, duplicated commentary |
Stage gates | Is there a real proceed, stop, recycle, or commit decision? | Ceremonial gates and document-only reviews |
Controls | Do cost, schedule, risk, and change data inform decisions? | Reports that describe history but do not show forecast or action |
Accountability | Is one role accountable for the outcome? | Collective approval with no clear owner |
This kind of model should not be imposed as a slogan. It needs to be designed around the project type.
An offshore brownfield modification has different governance needs from a utility-scale solar project. A mining expansion has different interface risks from a marine logistics program. An EPC lump-sum contract has different reporting and change-control requirements than a reimbursable construction management model.
The governance design should follow the risk.
How To Reduce Governance Without Losing Control
Reducing governance can make senior leaders nervous, especially in organizations that have recently had delivery problems. That concern is understandable. Removing controls without understanding risk can create serious exposure.
The answer is to simplify with discipline.
Map The Current Decision Path
Start with a small number of common decisions. Examples include contract variations, schedule baseline changes, technical concessions, contingency drawdown, procurement substitutions, and gate approvals.
For each decision, map:
Who prepares the submission
Who reviews it
Who comments on it
Who approves it
Who can stop it
How long the process usually takes
What evidence is required
Which systems or reports are updated afterward
This often reveals hidden delays. A process may show three formal approval steps but include ten informal review points. Those informal steps are often where time is lost.
Classify Governance By Risk And Consequence
A minimum viable model needs clear thresholds. The project should define what makes a decision low-, medium-, or high-governance intensity.
Useful criteria include:
Impact on approved budget
Impact on critical path
Safety or environmental consequence
Regulatory or permit effect
Contractual exposure
Technical complexity
Interface impact
Reputational or stakeholder sensitivity
Effect on operations or maintainability
This lets the team apply more assurance where it matters and less where it doesn't.
Remove Duplicated Reviews
Duplicated reviews often survive because each function once had a valid reason to be involved. Over time, those reviews become permanent.
A practical clean-up should ask whether each reviewer is required for one of three reasons:
They provide evidence the decision-maker needs
They hold delegated authority for part of the decision
They provide mandatory assurance for risk, compliance, or technical integrity
If none of these applies, the review may be optional, advisory, or unnecessary.
Make Stage Gates Decision-led
A stage gate should have a clear decision question. For example:
Is the scope mature enough to approve the next funding tranche?
Is engineering definition sufficient to start procurement?
Is construction readiness adequate for mobilization?
Is commissioning risk acceptable for energization?
Is the asset ready for handover to operations?
This approach strengthens project assurance because the review focuses on readiness and risk, not on document volume.
Protect Escalation Routes
Minimum viable governance depends on fast escalation. If a decision exceeds authority, affects critical risk, or materially deviates from the sanctioned case, it must move quickly to the right level.
Don't treat escalation as failure. In complex projects, escalation is a normal control mechanism. Failure occurs when teams delay escalation because the route is unclear or politically difficult.
Keep The Audit Trail Simple But Complete
Governance reduction should not mean poor records. Major decisions still need evidence.
A simple decision record should capture:
The decision made
The accountable decision-maker
The options considered
The basis of approval
Key risks and assumptions
Conditions or follow-up actions
Date and authority level
This protects accountability and helps future reviews understand why the project acted as it did.
What Good Minimum Viable Governance Looks Like In Practice
A project with effective minimum viable governance feels different. It is not uncontrolled. It is more deliberate.
Decisions move at the lowest appropriate level. Senior forums spend less time reviewing routine changes and more time resolving material risks, commercial exposure, strategic trade-offs, and sponsor-level constraints.
Reports become shorter but more useful. They show trends, exceptions, forecast movement, threats, opportunities, and required decisions. They avoid repeating information that is already available elsewhere.
Stage gates become sharper. A gate review may still be demanding, especially before high-value commitments, but participants understand the decision being made. The evidence supports that decision. Actions are specific. Conditions are tracked.
Project controls teams also spend their time differently. Instead of maintaining multiple reporting products, they analyze variance, test forecasts, support change control, challenge schedule logic, and make risk visible. That is where project controls creates value.
For senior leaders, the benefit is more efficient governance without weakening assurance. For delivery teams, the benefit is less waiting and less duplicated work. For sponsors, the benefit is clearer accountability and a better view of material risk.
Success isn't fewer meetings or fewer templates. Those may be outcomes, but they are not the goal. The goal is better control with less friction.
Common Mistakes When Applying The Concept
Minimum viable governance can be misused if treated as a cost-reduction exercise. The phrase can sound attractive to teams frustrated by bureaucracy, but poor application can remove necessary controls.
The main mistakes are predictable.
Removing Assurance From High-risk Decisions
Some decisions need serious review. Don't rush major contract awards, safety-critical design changes, start-up readiness, regulatory commitments, or large commercial settlements through weak governance.
Minimum viable does not mean minimum effort. It means the right effort.
Confusing Speed With Quality
Faster decisions are only useful when they are also better decisions. A rushed approval with poor evidence can create greater delay later through claims, rework, failed inspections, or operational problems.
The right model reduces waiting time, not thinking time.
Leaving Authority Unclear
Governance cannot improve if decision rights remain vague. Removing forums without strengthening delegated authority may speed things up briefly, then become chaotic when difficult decisions arise.
Authority must be visible, accepted, and used.
Keeping Reports But Removing Conversations
Some organizations cut meetings but keep the same reporting load. Others cut reporting but leave no forum for cross-functional decisions.
Governance works when information, authority, and accountability connect. Removing one part without redesigning the whole flow can weaken control.
A Better Question For Complex Projects
The practical question is not, “How much governance can we remove?”
A better question is, “Which governance activities improve decisions, assurance, authority, accountability, or risk control?”
If an activity supports one of those outcomes, it may be worth keeping. If it does not, it should be redesigned, combined, automated, reduced, or removed.
For major projects, this requires judgment. A governance step that looks unnecessary on one project may be essential on another. A monthly steering group may be too slow during a critical recovery period but suitable during steady execution. A detailed gate review may be essential before offshore mobilization but excessive for a low-risk internal milestone.
Minimum viable governance works best when it is treated as a living system. As a project moves from concept to FEED, execution, commissioning, handover, and closeout, the governance model should change. Risk changes. Authority needs to change. Assurance needs change. Reporting needs change.
Static governance rarely fits a dynamic project.
The organizations that manage this well tend to share a few habits. They review governance burden as part of project health. They ask whether decision cycle times are affecting delivery. They challenge duplicated reporting. They keep delegated authority current. They make escalation acceptable. They use project assurance to test readiness, not punish teams after the fact.
Most of all, they treat governance as a means of control, not as proof of control.
Project controls should help a project see clearly, decide quickly, and act responsibly. When controls become an obstacle, the answer is not to abandon governance. The answer is to make it viable again.
