Skill Locker
All skills
v1.3.03 loop iterations

Decision Matrix

Pros-and-cons lists pretend everything matters equally. The Decision Matrix forces you to name what *actually* matters most, weights the criteria, scores each option with evidence, and stress-tests the answer — so you can defend the choice when someone asks 'why this one?'

What this skill does

The problem with most decisions isn't that people don't think — it's that they think in the wrong order. They list options, list pros and cons, then pick the one that "feels right." Which means the decision is actually made by whichever factor they happened to think about most recently. Externalising the comparison flips that. You name what matters before you score, weight before you compare, and the answer falls out of evidence rather than vibes.

Five-to-eight criteria, no more. Beyond eight, several criteria are measuring the same underlying dimension — "ease of use" and "learning curve" are usability twice, "cost" and "ongoing maintenance" are total cost twice — and the matrix consolidates them. Each criterion gets a weight by pairwise comparison ("is cost more important than speed for this decision?"). Weights matter more than scores: two reasonable people score similarly but disagree about what matters most, and that's where decisions actually diverge.

Every score earns a justification note. "Scored 4 because £18k TCO over 12 months" is defensible. "Scored 4 because it feels right" is not. Then sensitivity analysis: if the top criterion's weight shifts by 10%, does the winner change? If the winning option's weakest score drops a point, does it still win? Two options within 0.3 points triggers the close-call protocol — the matrix says explicitly it isn't decisive, and you move to reversibility and regret-minimisation tiebreakers instead of pretending a near-tie is a clear answer.

It handles group decisions honestly: each stakeholder weights criteria independently, divergences above 15% surface a values disagreement before scoring. It handles personal decisions too — career, relocation, relationship — with emotional and identity criteria scored alongside rational ones, and a gut-check step after the math. And every output includes a "what you give up" section, because the best recommendation names the tradeoffs explicitly rather than pretending the winning option is superior in every way. A Quick Decision 5-minute mode exists for two-way doors where speed matters more than precision.

When this triggers

  • ·You're stuck between 2-5 options and the choice is non-trivial to reverse
  • ·Stakeholders disagree on a decision and you need a structured way to surface why
  • ·You've been going in circles weighing tradeoffs and need to externalise the comparison
  • ·You're evaluating vendors, tools, hires, or directions and need a defensible recommendation
  • ·You suspect your gut is telling you one thing but you can't articulate which factor is driving it

Example

Trigger

User: 'Choosing between three CRMs — HubSpot, Pipedrive, Salesforce. Need to decide this week.'

Output

Decision: Which CRM by Friday | Reversibility: medium (migration is real but not catastrophic at our size) Constraints (pass/fail) — all three options passed. WEIGHTED SCORING | Criterion (Weight) | HubSpot | Pipedrive | Salesforce | |---|---|---|---| | Cost — TCO 12mo (30%) | 4 — £18k | 5 — £9k | 2 — £42k | | Sales-team fit (25%) | 4 — modern UX | 5 — built for SMB | 3 — overkill | | Reporting depth (20%) | 4 — strong | 3 — adequate | 5 — best in class | | Integration with stack (15%) | 5 — native | 3 — Zapier | 4 — heavy lift | | Implementation speed (10%) | 4 — 2 wks | 5 — 1 wk | 2 — 3 mo | Weighted totals: HubSpot 4.20 · Pipedrive 4.15 · Salesforce 3.05 SENSITIVITY: HubSpot vs Pipedrive is within 0.05 — not decisive. If Cost weight rises to 35% (defensible), Pipedrive wins. If Reporting weight rises to 25%, HubSpot widens the lead. Close-call protocol triggered. Tiebreakers: · Reversibility: Pipedrive is easier to leave. · Optionality: HubSpot grows with you; Pipedrive may force a migration in 18 months at current growth. RECOMMENDATION: HubSpot — but only if 18-month commitment is realistic. If runway uncertainty is high, take Pipedrive and save £9k for now. Salesforce is out — sensitivity testing can't recover it.

Get this skill + 5 more

Get the full Research Intelligence pillar (6 skills) or the complete library.

Get the full stack — $299

What you get

  • 226-line SKILL.md, ready to drop into ~/.claude/skills/
  • Tested through 3 Karpathy-loop iterations (versions v1.0.0 → v1.3.0)
  • Triggers automatically when relevant — no command to remember
  • Lifetime updates as the skill is refined further

More from Research Intelligence

Browse the full library

297 skills across 31 categories. One purchase, lifetime updates.

See all bundles