prioritize

Installation
SKILL.md

Score and rank features or initiatives so teams build the highest-value things first.

Step 1: Gather the List

Collect the items to prioritize. If the user hasn't provided them, ask. Each item needs a clear, concise name and a one-sentence description of what it delivers.

Don't prioritize fewer than 3 items (just pick the best one) or more than 25 (split into themes first, then prioritize within each).

Step 2: Choose a Framework

Pick one based on what the team actually has available:

If you have... Use Why
Usage data, reach estimates, engineering sizing RICE Rigorous, accounts for confidence in your estimates
JTBD need statements with importance/satisfaction data Opportunity scoring Finds underserved needs directly
A large list and need a quick first pass Impact/Effort matrix Fast triage, no math required
Limited data but reasonable intuition ICE Lighter-weight RICE, moves fast

If unsure, default to Impact/Effort matrix for triage, then apply RICE to the top-right quadrant.

Step 3: Score Each Item

RICE

Score each item on four dimensions, then calculate:

RICE Score = (Reach x Impact x Confidence) / Effort
Dimension What it measures Scale
Reach How many users/accounts affected per quarter Actual number (e.g., 500 users/quarter)
Impact How much this moves the needle per user 3 = massive, 2 = high, 1 = medium, 0.5 = low, 0.25 = minimal
Confidence How sure you are about the above estimates 100% = high, 80% = medium, 50% = low. Below 50% means you need research, not prioritization
Effort Person-months of work Actual estimate (e.g., 2 person-months)

Opportunity Scoring

For each need identified through JTBD work:

Opportunity Score = Importance + max(Importance - Satisfaction, 0)
Dimension Scale
Importance 1-10: How important is this need to the job performer?
Satisfaction 1-10: How well do current solutions satisfy this need?

Scores above 12 are strong opportunities. Scores above 15 are urgent.

Impact/Effort Matrix

Plot each item on a 2x2:

          High Impact
               |
   Quick Wins  |  Big Bets
   DO FIRST    |  PLAN CAREFULLY
  -------------|-------------
   Fill-ins    |  Money Pits
   DO LAST     |  DON'T DO
               |
          Low Impact
  Low Effort ------- High Effort

Score impact 1-5 and effort 1-5 if you want numbers. Otherwise, just place items on the grid.

ICE

ICE Score = Impact x Confidence x Ease
Dimension Scale
Impact 1-10: How much will this move the target metric?
Confidence 1-10: How sure are you about impact and ease?
Ease 1-10: How easy is this to implement? (10 = trivial, 1 = months of work)

Step 4: Rank and Present

Sort items by score descending. Present as a table:

| Rank | Item | Score | R | I | C | E | Rationale |
|------|------|-------|---|---|---|---|-----------|
| 1    | ...  | ...   | . | . | . | . | ...       |
| 2    | ...  | ...   | . | . | . | . | ...       |

Adapt columns to match the chosen framework. Always include the final score, component scores, and a brief rationale for each item.

Group results into tiers when scores cluster:

  • Tier 1 — Build now. Clear leaders.
  • Tier 2 — Build next. Strong but not urgent.
  • Tier 3 — Build later or revisit. Low scores or low confidence.

Step 5: Sanity Check

After ranking, review the output critically:

  • Does the top item actually feel like the top priority? If not, identify what the framework missed (strategic alignment, dependencies, political reality).
  • Are there items with low confidence dragging down or inflating scores? Flag them separately as "needs research before prioritizing."
  • Do dependencies change the order? Item #3 might need to ship before Item #1 can work.
  • Is there a quick win in the top 5 that could ship this week? Call it out. Momentum matters.

When framework output and intuition disagree, don't just override the numbers. Name the disagreement explicitly: "RICE ranks X first, but the team believes Y is more urgent because [reason]." Then decide deliberately.

Output Format

Deliver:

  1. The framework chosen and why
  2. A scored, ranked table
  3. Tier groupings with recommended action (build now / build next / revisit)
  4. Any flags: low-confidence items, dependency issues, intuition mismatches
Weekly Installs
9
GitHub Stars
34
First Seen
1 day ago