Three PMOs, One Board: Choosing the Right P3O Model
Northgate Health Trust operates three division-level PMOs: Acute Services (14 staff), Community Services (9) and Estates & Facilities (6). Each uses its own templates, risk-register format and reporting cycle. The Trust Board receives three incompatible monthly reports and cannot compare delivery confidence across a £210m capital portfolio of 96 projects and 11 programmes. Internal audit found duplicated assurance reviews on four projects and no enterprise view of scarce clinical engineering resources. The COO proposes dissolving all three PMOs and creating a single 35-person central portfolio office reporting to the CFO, with all staff co-located at headquarters. Division directors refuse to release staff and warn that local knowledge will be lost. PPM maturity is low-to-moderate; two divisions have strong programme leadership, one does not. A business case is due in eight weeks and the CFO has said she will not approve a net cost increase without a clear value case.
1. Evaluate the COO's proposed central portfolio office against alternative P3O model types, and recommend a target model with justification.
Read the worked answer
Separate the problem from the proposed solution. The symptoms - incompatible reports, duplicated assurance, no enterprise resource view - point to the absence of a common operating model, not to the existence of three offices. A central office is one way to fix that; it is not the only way, and the proposal carries a net staff increase from 29 to 35, so it must win on value rather than on headcount logic.
Options:
- Central portfolio office: strongest control and consistency, single standards and reporting. But it concentrates transition risk, requires co-location, loses local knowledge and has already triggered active resistance from two divisions.
- Centre of excellence only: good for templates, training and tools, but too weak to deliver portfolio prioritisation, comparability or an enterprise resource view.
- Hub-and-spoke (federated): a corporate portfolio office owning prioritisation, enterprise reporting, assurance, the resource view and standards, plus a centre-of-excellence capability for methods and tooling, with divisional P3Os retained as spokes providing local delivery support and reporting functionally to the hub.
Recommendation: hub-and-spoke. The portfolio of 96 projects and 11 programmes needs enterprise prioritisation and resource visibility, which the hub provides. Two divisions have strong programme leadership and can anchor spokes; the weaker division can be supported by a hub-led transition team rather than being absorbed. Co-location is not necessary for the model to work and would destroy the local knowledge the directors are protecting. A hub of roughly 10–12 staff plus retained spokes should be cheaper and lower-risk than 35 central staff.
Conditions: a single data model and reporting standard, one risk taxonomy, assurance owned by the hub, a clear RACI between hub and spokes, and phased migration over 9–12 months with a decision gate. If the Board cannot mandate common standards, the hub will fail and the stronger central model becomes the fallback - so test the mandate before committing.
2. Beyond staff cost savings, what must the P3O business case contain, and what funding model should be recommended?
Read the worked answer
The business case must be benefits-led, because the proposal is not a cost-saving one. A credible case contains:
- Strategic case: which Board problems the P3O solves - comparability of delivery confidence, portfolio prioritisation, enterprise resource visibility, assurance efficiency.
- Options appraisal: central, hub-and-spoke, centre of excellence and do-nothing, assessed against cost, risk, deliverability and benefit.
- Service catalogue and customers: what is provided, to whom, at what service level. Without this the benefits cannot be traced.
- Benefits and dis-benefits: benefits include removal of duplicate assurance (four projects identified), faster and comparable reporting, scarce clinical engineering capacity released or redeployed, audit findings closed, better investment decisions. Dis-benefits include loss of local responsiveness, a transition productivity dip, dual-running cost and morale impact.
- Costs: transition, redundancy, tooling, training, dual running and steady-state run costs, with the 29-to-35 headcount movement shown honestly.
- Risks, dependencies, timeline and benefit owners, with baselines and review points at 6 and 12 months.
Funding: fund the hub centrally as an enterprise service, because its benefits accrue to the Trust rather than to any one division. Use recharge or transfer pricing only for genuinely discretionary spoke services, and do not introduce full chargeback at low maturity - it encourages divisions to bypass the P3O and recreate shadow capability. Provide transitional funding for 12–18 months, then review against measured benefits. Express cost in CFO language: cost per project, cost per £m of portfolio, assurance days avoided, reporting effort released.
3. The division directors threaten to withhold staff. How should this be handled so the business case remains deliverable?
Read the worked answer
Treat this as a mandate and stakeholder problem, not a negotiation to be won by argument. First, secure an explicit Board-level mandate for common portfolio standards and reporting; without it the hub has no authority and the business case is fiction. Second, remove the unnecessary threat by dropping compulsory co-location - the model does not require it - and by guaranteeing that spokes retain local delivery support roles. Third, give each director a stake: a named hub liaison, agreement that their strong programme leadership will co-design the common data model, and visibility of what they gain (resource conflicts resolved, fewer duplicate assurance reviews). Fourth, sequence transition so no division is asked to release critical staff before its spoke is stable. Finally, record the residual risk in the business case: if a division refuses, the hub model degrades and the Board must choose between accepting inconsistent reporting or mandating the central model. Do not hide that choice.