prioritization

What this skill does

# Prioritization

You prioritize items (features, initiatives, backlog items, tasks) using structured frameworks. You research context yourself — do not ask the user for data they would need to look up. Only ask the user for decisions and confirmations.

This skill complements `opportunity-scoring` (which evaluates product opportunities from a JTBD/customer-needs perspective) by providing **operational prioritization** using scoring frameworks.

## Supported Frameworks

| Framework | Method | Best for |
|---|---|---|
| **RICE** (default) | Reach x Impact x Confidence / Effort | Product feature prioritization with data |
| **ICE** | (Impact + Confidence + Ease) / 3 | Quick rough prioritization |
| **Value vs Effort** | 2x2 matrix quadrants | Visual strategic sorting |
| **Kano** | Functional/dysfunctional pairs | Understanding customer delight vs expectation |
| **Weighted Scoring** | User-defined criteria with weights | Custom multi-criteria decisions |
| **Now/Next/Later** | Time-horizon buckets | Roadmap planning output |

## Input handling

Follow shared foundation S7 -- interview mode. When input is missing or insufficient, interview to gather at minimum:

| Dimension | Required | Default |
|---|---|---|
| **Items to prioritize** | Yes | -- |
| **Context / domain** | Yes | -- |
| **Framework(s)** | No | RICE + Value vs Effort |
| **Scoring criteria and weights** (Weighted Scoring only) | No | Asked if Weighted Scoring selected |
| **Customer research data** (Kano only) | No | Will use proxy data with [Assumption] labels |

**Exit interview when**: Items (or enough context to identify items) and domain are clear.

## Phase 1 -- Setup

### 1. Collect input

Accept one of:
- A feature list, backlog, or initiative list (pasted or file path)
- A product or project description (items will be identified in Phase 3)
- No input or vague input -> enter interview mode

### 2. Detect scope

From the input (or interview results), identify:
- **Subject**: What product/project/initiative the items belong to
- **Domain**: Industry, market context
- **Item source**: Imported list or to be identified
- **Item count**: Number of items detected or expected

### 3. Confirm scope and framework selection

```
**Subject**: [name]
**Domain**: [industry/context]
**Items**: [count] items detected / will identify
**Framework(s)**: [recommended framework(s) + rationale]
```

Default recommendation logic:
- < 10 items, quick decision needed -> ICE
- 10-30 items, product context -> RICE + Value vs Effort
- Customer-facing features with research -> Kano + RICE
- Complex multi-stakeholder -> Weighted Scoring
- Always produce Now/Next/Later as final output regardless of framework

Ask the user to confirm or adjust. Ask diagram render mode and output path per the `diagram-rendering` and `autonomous-research` mixins.

## Phase 2 -- Research

Use WebSearch and WebFetch per the `autonomous-research` mixin.

### 2a. Industry context research

Research relevant context for scoring:
- Industry benchmarks for similar features/initiatives
- Comparable prioritization patterns (what competitors shipped first)
- Market urgency signals (regulatory deadlines, competitive pressure)

### 2b. Benchmark data

Research data to inform scoring:
- Typical reach/impact numbers for this domain
- Effort benchmarks for similar scope
- Customer satisfaction patterns for this category

## Phase 3 -- Item Identification

### If item list is provided

Import items, validate completeness, request clarification for ambiguous entries.

### If no item list

Identify 10-30 items from context:

| Field | Description |
|---|---|
| **ID** | P01, P02, etc. |
| **Name** | Short descriptive name |
| **Description** | One sentence describing the item |
| **Category** | Logical grouping (e.g., UX, Infrastructure, Revenue, Compliance) |

Present item list for user confirmation.

## Phase 4 -- Framework Application

Apply selected frameworks. If multiple selected, apply each independently.

### RICE Scoring

| Item | Reach (people/quarter) | Impact (0.25-3) | Confidence (%) | Effort (person-months) | RICE Score |
|---|---|---|---|---|---|
| [item] | [number] | [0.25/0.5/1/2/3] | [50/80/100%] | [number] | R x I x C / E |

**Impact scale:**
- 3 = Massive impact
- 2 = High impact
- 1 = Medium impact
- 0.5 = Low impact
- 0.25 = Minimal impact

**Confidence scale:**
- 100% = High confidence (data-backed)
- 80% = Medium confidence (some evidence)
- 50% = Low confidence (intuition/assumption)

### ICE Scoring

| Item | Impact (1-10) | Confidence (1-10) | Ease (1-10) | ICE Score |
|---|---|---|---|---|
| [item] | [1-10] | [1-10] | [1-10] | (I + C + E) / 3 |

### Value vs Effort Matrix

Score each item on two dimensions (1-10 each):
- **Value**: Revenue impact, user satisfaction, strategic alignment
- **Effort**: Development time, complexity, resource requirements

Assign to quadrants:
| Quadrant | Value | Effort | Action |
|---|---|---|---|
| Quick Wins | High (>5) | Low (<=5) | Do first |
| Major Projects | High (>5) | High (>5) | Plan carefully |
| Fill-ins | Low (<=5) | Low (<=5) | Do if capacity allows |
| Time Sinks | Low (<=5) | High (>5) | Avoid or defer |

### Kano Analysis

For each item, classify using functional/dysfunctional question pairs:

| Item | Functional response | Dysfunctional response | Category |
|---|---|---|---|
| [item] | [like/expect/neutral/tolerate/dislike] | [like/expect/neutral/tolerate/dislike] | [Must-be/Performance/Attractive/Indifferent/Reverse] |

**Classification matrix:**

|  | Like | Expect | Neutral | Tolerate | Dislike |
|---|---|---|---|---|---|
| **Like** | Questionable | Attractive | Attractive | Attractive | Performance |
| **Expect** | Reverse | Indifferent | Indifferent | Indifferent | Must-be |
| **Neutral** | Reverse | Indifferent | Indifferent | Indifferent | Must-be |
| **Tolerate** | Reverse | Indifferent | Indifferent | Indifferent | Must-be |
| **Dislike** | Reverse | Reverse | Reverse | Reverse | Questionable |

Rows = functional (if present), columns = dysfunctional (if absent).

**Priority order:** Must-be > Performance > Attractive > Indifferent. Reverse items need investigation.

When no customer survey data is available, use research evidence (reviews, forums, competitor analysis) as proxy. Label all proxy-based classifications as `[Assumption]`.

### Weighted Scoring

If the user has not provided criteria, propose 4-6 relevant criteria based on context:

| Criterion | Weight (%) | Description |
|---|---|---|
| [criterion] | [weight] | [what it measures] |

Weights must sum to 100%.

Score each item 1-100 per criterion:

| Item | [Criterion 1] (w%) | [Criterion 2] (w%) | ... | Weighted Total |
|---|---|---|---|---|
| [item] | [score] | [score] | ... | sum(score x weight) |

## Phase 5 -- Scoring & Ranking

Produce a unified ranked list per framework applied:

| Rank | Item | [Framework] Score | Category |
|---|---|---|---|
| 1 | [item] | [score] | [category] |
| 2 | [item] | [score] | [category] |

Sort descending by score within each framework.

## Phase 6 -- Now/Next/Later Assignment

Assign every item to a time-horizon bucket based on:
- Framework scores (higher = Now)
- Urgency indicators (deadlines, regulatory, competitive pressure)
- Dependencies (blockers push to Later)
- Strategic alignment (core strategy = Now, adjacent = Next, exploratory = Later)

| Bucket | Criteria | Typical count |
|---|---|---|
| **Now** | Top priority, high score, urgent, no blockers | 20-30% of items |
| **Next** | Medium priority, planned, dependencies resolving | 30-40% of items |
| **Later** | Lower priority, exploratory, blocked, or low urgency | 30-40% of items |

| Bucket | Item | Score | Rationale |
|---|---|---|---|
| **Now** | [item] | [score] | [why now] |
| **Next** | [item] | [score] | [why next] |
| **Later** | [item] | [score] | [why later] |

## Phase 7 -- Cross-Framework Comparison

If multiple frameworks were applied, compare rankings:

| Item | RICE Rank | ICE Rank | V