Prioritize a portfolio after eligibility gates

A portfolio performs two different functions: exclude opportunities that cannot appropriately operate under current conditions, then rank eligible alternatives. Before scoring, examine whether data use is permitted in context, quality can be tested, an accountable decision owner exists and the team can respond to failure. These are context-specific eligibility gates, not points that implementation speed can offset. If a material data-use right is unresolved, hold the opportunity until that right is established. High expected benefit does not make it eligible. Preserve the exclusion reason and the evidence needed to reconsider it.

After the gates, choose criteria that do not count the same benefit twice. Criteria might cover expected value, feasibility and strategic learning, with weights summing to one and clearly defined common scales. The weighted score is the sum of each weight multiplied by its criterion score. Record who selected the weights and why. Test sensitivity: does the winner change after a reasonable weight adjustment or a reduction in an uncertain estimate? Ranking is not an automatic decision. Examine dependence on the same team or dataset; a promising portfolio may fail if every project needs the same data engineer in the same week.