Project governance specialists for programmes that need clear decisions

Start with the decisions that stall

Most governance problems show up as delay. An approval sits with someone who is not sure they hold authority, or a risk waits for a forum that meets fortnightly, and the programme absorbs the cost in idle weeks.

Before writing any policy, list the decisions your programme makes repeatedly and mark where each one currently stops. That list is the real brief for governance design, and it is usually shorter than the policy document someone has already drafted.

Name the decision, the person who currently blocks it, and the cost of the wait. When you can point to five such decisions, you have enough to design a model that fixes real delays rather than describing an ideal process nobody follows. The brief should also name who decides, what must be true before a shortlist, and what would make you stop. The brief should also name who decides, what must be true before a shortlist, and what would make you stop.

Define decision rights and forums

Decision rights describe who can say yes, who must be consulted and who only needs to know. A RACI chart is a common way to record this, but the value comes from specificity: name the role, not the department, and state the threshold at which authority moves up.

Ambiguity here is what produces escalation by email chain. If a programme manager is unsure whether a change of a certain value needs sponsor approval, the safe move is to ask, and asking costs days. Writing the threshold down removes that hesitation.

Forums need the same treatment. State which body meets, how often, what it decides and who must attend. A steering group that meets monthly but is asked to approve weekly releases is a structural mismatch, not a discipline problem.

Keep the record short and current. A one-page decision rights table that people actually read beats a forty-page governance manual that nobody opens after the launch meeting. The table should name each forum, its chair and the decisions it owns.

Design stage gates with evidence rules

A stage gate is a decision to proceed, pause or stop. It only works if everyone knows what evidence must exist before the review, and who confirms that the evidence is genuine rather than asserted.

Define entry criteria for the gate and the specific documents, sign-offs or test results that confirm readiness. Without that, the gate becomes a meeting where opinions compete and the loudest voice decides, which is the opposite of governance.

Gates also need a named chair with authority to stop work. If the chair can only recommend, the gate is advisory and the programme will learn to route around it within a quarter. The brief should also name who decides, what must be true before a shortlist, and what would make you stop. The brief should also name who decides, what must be true before a shortlist, and what would make you stop The brief should also name who decides, what must be true before a shortlist, and what would make you stop.

Build escalation and risk routes

Escalation is a route with a clock, not a reaction to frustration. For each risk category, state who is informed first, who decides, and how long each step may take before it moves up. The clock is what separates an escalation path from an org chart.

A risk that waits three weeks for a forum is a risk that has already changed shape by the time it is discussed. By then the cheap options have gone and the conversation is about damage rather than choice.

Score risks on likelihood and impact, but also on decision latency: how long the programme can afford to wait before the risk forces a decision. That third dimension is what tells you which risks cannot wait for the monthly meeting.

Review the routes after the first few escalations. If most issues bypass the defined path, the path is wrong, not the people using it, and it should be redrawn before the next phase begins. The register should show owners and review dates, not just a severity score.

Make reporting serve the forum

Status reporting should mirror the decisions the forum makes. If the steering group decides on funding, scope, risk acceptance and vendor commitments, the pack needs a section for each, with the decision required stated plainly at the top rather than buried on slide fourteen.

Activity summaries belong in an appendix, if anywhere. A pack that lists tasks completed tells the forum what the team did, not what the forum must now decide, and it trains sponsors to skim.

Use data your teams already produce. A reporting design that requires a new timesheet or a second tracking tool will decay within two months, because the people filling it get nothing back for the effort.

Test the pack against the last three meetings. If the forum spent its time on items the pack did not flag, the reporting is describing the programme rather than serving the decision. The brief should also name who decides, what must be true before a shortlist, and what would make you stop.

Plan the handover from the start

Governance designed by a specialist should be run by your team. From the first week, identify who will own the model, the registers and the reporting after the engagement ends, and involve them in the design rather than handing them a document at the close.

If no one is named, the model will lapse quietly once attention moves elsewhere. Registers go stale, gates get skipped under delivery pressure, and the programme returns to the pattern that prompted the engagement in the first place.

Handover includes templates, a written operating guide and a walkthrough with the people who will run each part. It also includes a review date, so the model gets tested against a real gate rather than assumed to work.

A specialist who leaves behind a programme office that can run the model without them has done the job. One who remains indispensable has only moved the dependency, not resolved it. The brief should also name who decides, what must be true before a shortlist, and what would make you stop.

Conclusion

Good governance is quiet. Decisions get made on time, risks reach the right people while options remain, and records answer questions months later. Design it around your real decisions, keep the artefacts short and current, and plan the handover before work begins. TalentSkape can match you with a specialist to set that model and leave your team running it.