A pharmaceutical development decision commits years of capacity and crores of spend, and it is often made on a slide deck and the memory of whoever was in the room. Months later, when someone asks why a particular molecule was chosen over another, the honest answer is difficult to reconstruct. Portfolio selection deserves better than advocacy.
Why "gut plus a deck" fails at scale
Committee-based selection is not wrong because committees are wrong — experienced heads bring judgement no model replaces. It fails because the reasoning is not captured. Market data, feasibility, and each function's view live in separate files, so the decision cannot be reconstructed, compared, or defended after the fact. When the shortlist is challenged, there is nothing to point to but a conclusion.
A rubric the whole committee scores against
The alternative is to make every function score every candidate molecule against the same fixed set of criteria. A workable rubric covers technical feasibility, manufacturing complexity, market attractiveness, competitive position, NPV or ROI outlook, investment required, clinical differentiation, and regulatory risk. Each criterion forces a specific question and is naturally owned by a specific function — feasibility by R&D, complexity by operations, attractiveness by marketing, and so on.
The rubric does not remove judgement. It structures it, so that the shortlist is the product of a consistent method rather than the loudest voice, and so that it can be explained a year later by someone who was not in the room.
Evidence before opinion
Scoring is only as good as the data behind it. Loading market datasets — volume, value, reference pricing, regional pull — and attaching them to the molecules they describe means the conversation starts from what the data says. A strategy or attractiveness view lets the committee argue over a short list rather than a long one, focusing effort where the decision is genuinely close.
The decision trail is the real asset
The output that matters most is not the shortlist itself but the record behind it: who scored what, on which evidence, and who approved the result. When that shortlist is signed off through staged electronic approval with the reasoning attached, the portfolio decision becomes auditable. Spend on de-prioritised molecules is avoided at the gate rather than discovered late, and the selection can withstand scrutiny from a board, a partner, or a regulator.
From quarters to weeks
A selection cycle that runs on circulating decks and reconvened committees can consume a full quarter. The same decision, made against a rubric with the data loaded and the approval built in, compresses dramatically — not because anyone is rushed, but because the friction of assembling and reconciling evidence disappears. The decision is faster and, more importantly, it is defensible.