Capital Allocation Skill

Capital Allocation Skill

Allocating capital is the core executive job: a fixed pot, more good ideas than money, and the need to say no on the record. This skill scores initiatives by expected return and strategic fit per unit of cost, allocates against the cap (honouring must-funds), and makes the cut line explicit — so funding is a defensible portfolio choice, not the loudest voice in the room.

Required Inputs

Ask for these only if they aren't already provided:

• The cap — the total budget or headcount to allocate, and the period.
• The initiatives — each with its cost, expected return (revenue, savings, or a strategic value), and strategic fit.
• Constraints — anything that must be funded (compliance, keep-the-lights-on) or can't be partially funded.
• The objective — what you're optimising: near-term return, strategic positioning, or a balance.

Output Format

Capital Allocation: [pot], [period]

1. Objective & cap — what you're optimising and the total available.

2. Scored initiatives — a table; score = expected value × strategic fit, normalised per unit cost:

3. The allocation — funded vs. unfunded against the cap, with budget utilisation. Must-funds first, then highest score/$ until the cap binds.

4. The cut line — the marginal initiative that just missed, and what it would take to fund it (the most useful number for the debate).

5. Rationale & trade-offs — why the portfolio is balanced this way, what's deliberately not funded, and the reversibility of each bet.

6. Re-evaluation triggers — what would change the allocation mid-period (a bet pays off early, a must-fund grows).

Programmatic Helper

scripts/capital_allocate.py (stdlib only) does the allocation deterministically — must-funds first, then by score-per-cost until the cap binds — and reports the cut line: