5.1. Define and Establish Project Governance
💡 First Principle: Governance is the structure that determines who can approve what, and at what threshold — without it explicitly defined, escalation and decision-rights become ad hoc, and the exam tests whether you proactively establish that structure rather than improvise it when a decision is contested.
Enablers: describe and establish structure/rules/procedures/reporting/ethics/policies using organizational process assets (OPAs), define success metrics, and outline governance escalation paths and thresholds.
Escalation thresholds matter concretely: a governance structure that defines "changes affecting budget by more than 10% require steering-committee approval" gives the PM (and the team) a clear, depersonalized rule to point to — versus a project where every borderline decision becomes a political negotiation because no threshold was ever set.
Who governs, and what they are for
The project charter, issued by the sponsor or initiator, formally authorizes the project and gives the project manager authority to apply organizational resources. Governance then spells out what the PM may decide alone and what must go higher.
The steering committee (or an equivalent governance board) provides guidance, makes strategic decisions such as go/no-go at stage gates, and keeps the project aligned with organizational objectives. Because it carries cross-functional representation, it is also the right forum when two functional departments both claim a shared critical resource — the PM brings the conflict to the body that has authority over both, rather than picking a winner.
Governance defines who decides what, at what threshold, scaled to risk and impact. A sponsor who insists on approving every task assignment on a 40-person project is not practicing strong governance; approving decisions far below any reasonable threshold is micromanagement, and it slows the project without reducing risk.
Governance also sets the boundaries you work within. When the steering committee says there is no more money and the date cannot move, adding people, crashing, or dipping into reserves are off the table — the strongest answer usually optimizes the process itself, for example removing non-value-adding work with lean methods and using retrospectives to surface improvements.
Build it from what the organization already has
Your first move is to check the organizational process assets — existing policies, approval workflows, governance templates, lessons learned — before drafting anything new. If a well-tested framework already exists, adapt it and tailor only where this project genuinely differs. A new governance policy rolls out in a fixed order: draft it from existing OPAs, confirm it with the steering committee, communicate it to the team and affected stakeholders, then apply it going forward.
| Organizational process assets | Enterprise environmental factors | |
|---|---|---|
| What they are | Plans, processes, policies, procedures, templates and knowledge bases the organization owns | Conditions outside the team's immediate control that influence or constrain the project |
| Examples | Approval workflow, governance template, lessons-learned repository | Organizational culture and governance framework, IT infrastructure, regulations, market conditions |
| How you use them | Draw on them and update them | Account for and work within them |
Tailoring applies to governance too. Forcing a small, low-risk internal project through the weekly reviews and approval layers built for the largest initiatives creates overhead out of proportion to its risk. In the other direction, an innovative, uncertain project inside a centralized, stage-gated organization should not ignore the gates — the PM proposes iterative decision-making and learning cycles within the existing gates and justifies the flexibility. When a PMO tailors a framework for the whole organization, the finishing move is a tailoring process based on project size and complexity plus a mechanism for ongoing improvement.
Offices that support governance
| Type | Level of control | What it does |
|---|---|---|
| Supportive PMO | Low | Provides templates, training and lessons learned |
| Controlling PMO | Moderate | Requires compliance with frameworks and reviews |
| Directive PMO | High | Directly manages the projects |
| Value delivery office (VDO) | Enabling, not controlling | Coaches teams and leaders, builds adaptive capability, mentors sponsors and product owners, supports leaner governance |
An organization that wants to assess and improve its capability across projects, programs and portfolios is working in organizational project management (OPM), for which PMI publishes The Standard for Organizational Project Management.
Escalation paths and thresholds
A threshold turns "should this go up?" into a lookup. Without one, every 2% overrun can trigger a full committee review. When a category of decision has no threshold, the answer is to work with the steering committee to define one — not to guess, and not to escalate everything.
When a risk grows past the level governance says must be reported — a low-rated risk that now threatens the core deliverable — you notify the committee promptly.
⚠️ Exam Trap: Establishing thresholds before they are needed and defining success metrics at kickoff are proactive governance; convening an emergency review after a threshold was already breached is reactive. When both appear as options, PMI favors the proactive one.
Success metrics and reporting
Success metrics belong to governance because agreeing them up front prevents the end-of-project dispute over whether the project delivered. A sponsor measuring revenue impact while the team optimizes for defect-free delivery is a governance gap: define the metrics now, with both parties. A reporting structure — who receives which information at what cadence — is part of the same design; without it, updates drift to an inconsistent mix of people. Governance that lives only in "everyone just knows how it's done" does not survive turnover, so write it down.
Ethics as policy, not personality
An explicit ethics policy gives everyone a shared, referenceable standard for ambiguous situations, instead of individual judgment that bends under pressure. Under the PMI Code of Ethics, honest reporting is not negotiable: if a sponsor asks you to drop negative performance data from a steering committee deck, discuss the concern and present the data alongside a remediation plan. When a committee asks where cost is heading, give it the estimate at completion — the most likely final cost — not the original budget. A vendor offering you event tickets during bid evaluation creates a conflict of interest: decline and disclose it, as the Fairness value requires.
⚠️ Exam Trap: Quietly omitting an unfavorable test result "to avoid alarming the sponsor" before a funding decision is never the answer, however it is framed. Report it transparently and pair it with the plan to address it.
⚠️ Exam Trap: Escalating every ambiguous decision to the sponsor by default. The better-scoring answer is usually to have established (or, if missing, to now establish) clear governance thresholds so routine decisions don't require case-by-case escalation.
