Guide
Project prioritization: how to arbitrate a portfolio
Scoring projects is easy. Deciding which ones stop is the hard part — and it only happens when the ranking meets capacity.
Prioritizing projects means answering a question most organizations avoid: given what we can actually staff and fund, which projects start, which wait, and which stop. A ranked list that never confronts capacity is not a prioritization — it is an opinion poll, and everything on it stays open.
Score on few criteria, and always the same ones
Four or five criteria are enough: strategic contribution, expected value, cost, risk, and sometimes regulatory obligation. Beyond that, the score becomes unreadable and every project finds a column where it shines. What matters far more than the number of criteria is that they are identical for all projects and stable from one cycle to the next — otherwise you are comparing rankings, not projects.
A score is an input, not the decision
Scoring gives an order. It does not give a plan. Two projects with the same score may need the same three people in the same quarter, and only one can happen. The decision appears when you go down the ranked list and stop at the point where capacity runs out — that line is the real portfolio, and everything under it must be told so explicitly.
Mandatory projects come out first
Regulatory, security and end-of-support projects are not arbitrated: they consume capacity before anyone scores anything. Putting them in the same ranking as discretionary projects distorts the whole exercise, because they will be done anyway. Take them out, subtract what they consume, and arbitrate on what is left — which is usually a much smaller envelope than expected.
Re-arbitrate on a rhythm, not on demand
A portfolio prioritized once a year is fiction by month four. A portfolio re-arbitrated at every request is chaos. The workable middle is a fixed cycle — quarterly for most organizations — where new candidates are compared against the ones already running, including the possibility of stopping something. A portfolio where nothing ever stops has no prioritization, whatever its scoring model says.
Arbitrating with FoxPlan
FoxPlan holds the scoring criteria, the cost and the resource demand on the same portfolio objects, so a ranking can be read against available capacity rather than beside it. Scenarios let you test an arbitration — this project in, that one postponed a quarter — and see the effect on workload and budget before committing, and the strategic objectives tree shows what the retained portfolio actually serves.
