influence-diagramming

What this skill does

# Influence Diagramming

You perform autonomous influence analysis. You research influence dynamics 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 `stakeholder-mapping` (which identifies and classifies stakeholders) by analyzing the **relationships and influence flows** between them.

## Phase 1 — Setup

### Input handling

Follow shared foundation §7 — interview mode. When input is missing or insufficient, interview to gather at minimum:

| Dimension | Required | Default |
|---|---|---|
| **Project/initiative context** | Yes | — |
| **Stakeholder mapping output** | No | Will identify stakeholders itself |
| **Known relationships** | No | Will be researched/inferred |
| **Focus** | No | All (change management, risk, communication, coalition) |

**Exit interview when**: Project context is clear enough to identify stakeholders and research influence dynamics.

### 1. Collect input

Accept one of:
- A project or initiative description
- A file path to a stakeholder mapping report (imports stakeholders)
- Pasted stakeholder list or business case content
- No input or vague input → enter interview mode

### 2. Detect scope

From the input (or interview results), identify:
- **Project/initiative**: What is being planned or executed
- **Stakeholder source**: Imported from mapping or to be identified
- **Known relationships**: Any explicitly mentioned influence dynamics
- **Focus areas**: Change management, risk, communication, coalition building (default: all)

### 3. Confirm scope

Present the detected scope to the user for confirmation:

```
**Project**: [name]
**Stakeholder source**: [imported from mapping / will identify]
**Known relationships**: [listed or "will be researched"]
**Focus**: [areas]
```

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. Influence dynamics research

Research typical influence patterns for this type of project/industry:
- Common power structures and organizational politics
- Typical influence dynamics in similar initiatives
- Best practices for influence management in this domain
- Known challenges with stakeholder influence in similar contexts

### 2b. Domain-specific context

Research factors that shape influence relationships:
- Industry governance structures and decision-making patterns
- Regulatory influence channels
- Typical coalition formations in similar organizations/projects
- Change management influence research for this domain

## Phase 3 — Stakeholder Import or Identification

### If stakeholder mapping is provided

Import stakeholders with their attributes:
- Read the stakeholder mapping report
- Extract: ID, Name/Role, Organization, Category, Power, Interest, Attitude
- Present imported list for user confirmation
- Note any stakeholders that should be added or removed for influence analysis

### If no stakeholder mapping

Identify 10-20 key stakeholders with basic attributes:

| Field | Description |
|---|---|
| **ID** | S01, S02, etc. |
| **Name/Role** | Generic role or specific title |
| **Organization** | Department or external entity |
| **Attitude** | Supportive / Neutral / Resistant |

Present stakeholder list for user confirmation before proceeding.

Target: 10-20 stakeholders. For small-scope projects, minimum 6 with explicit note.

## Phase 4 — Influence Relationship Assessment

For each pair of stakeholders with a meaningful relationship, assess:

| Field | Description | Values |
|---|---|---|
| **From** | Influencing stakeholder | Stakeholder ID |
| **To** | Influenced stakeholder | Stakeholder ID |
| **Direction** | Flow direction | Unidirectional (→) / Bidirectional (↔) |
| **Strength** | Influence intensity | 1-5 (1=minimal, 2=weak, 3=moderate, 4=strong, 5=dominant) |
| **Type** | Influence mechanism | One or more of 7 types (see below) |
| **Nature** | Relationship quality | Supportive / Neutral / Adversarial |
| **Basis** | Why this influence exists | Brief explanation |

### Seven influence types

| Type | Definition | Example |
|---|---|---|
| **Formal authority** | Decision-making power from position | CEO approves budgets |
| **Resource control** | Controls budget, staffing, equipment | CFO controls funding |
| **Information access** | Privileged access to critical data | Data analyst controls reporting |
| **Expertise** | Domain knowledge others depend on | Architect defines technical direction |
| **Social capital** | Trust, relationships, reputation | Senior developer mentors the team |
| **Political leverage** | Alliances, institutional memory | Long-tenured manager knows the history |
| **Referent power** | Charisma, role model status | Respected leader inspires followership |

### Rules

- Not every pair needs a relationship — only meaningful ones
- A single relationship can have multiple influence types
- Bidirectional relationships may have different strengths in each direction — record as two entries
- Indirect influence (A→C→B) is captured through the pathway analysis in Phase 8, not here

### Influence relationship table

| # | From | To | Direction | Strength | Type | Nature | Basis |
|---|---|---|---|---|---|---|---|
| 1 | S01 | S02 | → | 4 | Formal authority | Supportive | CEO sponsors the initiative |
| 2 | S03 | S04 | ↔ | 3 | Expertise, Social capital | Neutral | Peer architects collaborate |

## Phase 5 — Influence Matrix

Build an N×N matrix where:
- Rows = influencing stakeholder (outgoing influence)
- Columns = influenced stakeholder (incoming influence)
- Cell = influence strength (0 = no relationship, 1-5 = strength)

### Matrix format

| | S01 | S02 | S03 | ... | **Out Total** |
|---|---|---|---|---|---|
| **S01** | — | 4 | 0 | ... | [sum] |
| **S02** | 2 | — | 3 | ... | [sum] |
| **S03** | 0 | 3 | — | ... | [sum] |
| **In Total** | [sum] | [sum] | [sum] | ... | |

### Matrix analysis

- **Top influencers**: Highest outgoing total — these stakeholders exert the most influence
- **Most influenced**: Highest incoming total — these stakeholders receive the most influence
- **Influence ratio**: Outgoing / Incoming — >1 = net influencer, <1 = net influenced
- **Reciprocity**: Count of bidirectional relationships — indicates collaborative vs hierarchical dynamics

## Phase 6 — Network Analysis

### Centrality metrics

Calculate for each stakeholder:

**Degree centrality** = (number of connections) / (N-1)
- High = many connections = hub

**Betweenness centrality** = fraction of shortest paths between all pairs that pass through this stakeholder
- High = controls information flow = gatekeeper
- Simplified calculation: for each stakeholder, count how many other pairs would lose their connection if this stakeholder were removed

**Closeness centrality** = (N-1) / (sum of shortest distances to all other stakeholders)
- High = can reach everyone quickly = connector
- Use path length = 1 for direct connections; for indirect, count minimum hops

### Network roles

| Role | Identification criteria | Significance |
|---|---|---|
| **Hub** | Degree centrality > 0.6 | Central connector; failure disrupts many relationships |
| **Gatekeeper** | Betweenness centrality > 0.4 | Controls information flow; bottleneck risk |
| **Bridge** | Connects two otherwise disconnected subgroups | Critical for cross-group communication |
| **Broker** | High betweenness + connections to multiple coalitions | Mediator between factions |
| **Isolate** | Degree centrality < 0.1 | Excluded from influence network; risk of disengagement |
| **Connector** | High closeness centrality | Can reach all stakeholders efficiently |

Thresholds are guidelines — adjust if the network is very small or very large.

### Centrality analysis table

| ID | Stakeholder | Degree | Betweenness | Closeness | Role(s) | Significance |
|---|---|---|---|