positioning-engine

Orchestration Log: When this skill is activated, append a log entry to outputs/orchestration_log.md:

```

### Skill Activation: Positioning Engine

Timestamp: [current date/time]

Actor: AI Agent (positioning-engine)

Input: Paper draft + [N] comparison papers identified

Output: Differentiation matrix with [N] dimensions, positioning_analysis.md saved

```

Positioning Engine

Core Principle

"How is your paper different from X?" is the question every reviewer asks. This engine produces a systematic answer. It identifies the 5-10 most similar existing papers, builds a structured differentiation matrix, and generates a positioning statement that makes the unique contribution explicit and defensible.

The output directly strengthens the Introduction (gap + contribution paragraphs) and the Discussion (theoretical implications).

When to Activate

• User says "position my paper", "how is this different from X?", "differentiation"
• User says "compare to related work", "positioning analysis", "unique contribution"
• During Phase 2 (Framing) to sharpen the gap and contribution
• When a reviewer challenges the novelty or contribution
• User runs /analyze-positioning

Prerequisites

• draft.md or paper.tex exists (to understand the paper's claims)
• references.bib and/or literature_base.csv exist (comparison candidates)
• Research questions and contribution are at least tentatively defined

Step 1: IDENTIFY Closest Competitors

Finding the Most Similar Papers

Sources for comparison papers:

1. From the paper itself: Papers cited in the Introduction and Related Work

that address the same or very similar research questions

1. From literature_base.csv: Papers with the highest topical overlap
2. From snowballing: Forward citations of the paper's key references that

appeared after those references were published

1. Direct search: Search for papers with very similar titles or identical keywords

Selection Criteria

Select 5-10 papers that are the closest competitors — papers that a reviewer might say "but Author X already did this." These are papers that:

• Address the same or very similar research questions
• Study the same phenomenon in a similar context
• Use a similar methodology
• Apply the same theoretical lens
• Were published recently (high overlap risk)

Output

For each selected paper, extract:

{
  "bib_key": "[key]",
  "title": "[title]",
  "authors": "[authors]",
  "year": [year],
  "venue": "[venue]",
  "rq": "[their research question or objective]",
  "method": "[their method]",
  "theory": "[their theoretical lens]",
  "context": "[their empirical context]",
  "key_finding": "[their main finding]",
  "similarity_to_ours": "high/medium",
  "citation_count": [N]
}

Step 2: BUILD Differentiation Matrix

Matrix Structure

Build a table comparing the competitor papers against YOUR paper across key dimensions:

## Differentiation Matrix

| Dimension | Our Paper | [Author1 (Year)] | [Author2 (Year)] | [Author3 (Year)] | ... |
|-----------|-----------|-------------------|-------------------|-------------------|-----|
| **Research Question** | [our RQ] | [their RQ] | [their RQ] | [their RQ] | |
| **Theoretical Lens** | [our theory] | [their theory] | [their theory] | [their theory] | |
| **Method** | [our method] | [their method] | [their method] | [their method] | |
| **Context/Sample** | [our context] | [their context] | [their context] | [their context] | |
| **Time Period** | [our period] | [their period] | [their period] | [their period] | |
| **Technology Focus** | [our tech] | [their tech] | [their tech] | [their tech] | |
| **Unit of Analysis** | [our UoA] | [their UoA] | [their UoA] | [their UoA] | |
| **Key Contribution** | [our contribution] | [their contribution] | [their contribution] | [their contribution] | |
| **Scope** | [our scope] | [their scope] | [their scope] | [their scope] | |

Dimension Selection

Choose 6-10 dimensions that are most relevant for differentiation. Common dimensions:

| Dimension | When to Use | |-----------|------------| | Research Question / Objective | Always | | Theoretical Lens | When theory is a differentiator | | Method | When method is a differentiator | | Context / Industry / Sample | When context is new | | Time Period / Technology Generation | When recency matters | | Level of Analysis (individual, team, org, industry) | When level differs | | Unit of Analysis | When what's being studied differs | | Geographic / Cultural Context | For cross-cultural studies | | Dependent Variables / Outcomes | When outcomes differ | | Stage of Adoption / Maturity | For implementation studies | | Practitioner vs. Academic Focus | For applied research |

Step 3: ANALYZE Positioning

Gap Identification

From the differentiation matrix, identify:

1. Unique dimensions: Where does our paper differ from ALL competitors?
2. These are the strongest positioning arguments

1. Underrepresented combinations: What combination of dimensions is new?
2. Even if individual dimensions are not new, the combination may be

1. Temporal advantage: What has changed since the competitors published?
2. New technology, new phenomena, new policy context

1. Methodological gap: What method has NOT been applied to this topic?
2. E.g., many surveys, no case studies; many qualitative, no quantitative

1. Contextual gap: What context has NOT been studied?
2. E.g., many US studies, no European context; many large enterprises, no SMEs

Positioning Patterns

Classify the paper's positioning into one or more patterns:

| Pattern | Description | Strength | |---------|-------------|----------| | New phenomenon | Studies something that didn't exist before | Very strong | | New context | Applies known theories to unstudied context | Strong | | New method | Uses a method not yet applied to this topic | Strong | | New theory | Applies a theory not yet used for this topic | Medium-Strong | | Integration | Combines previously separate research streams | Medium | | Replication | Tests prior findings in new setting | Medium | | Update | Revisits topic after significant change | Medium | | Deeper | Goes beyond surface-level prior work | Depends on execution |

Step 4: GENERATE Positioning Statement

Positioning Paragraph Template

Generate a paragraph that can be directly inserted into the Introduction (Gap section) or Related Work:

While prior work has made important contributions to understanding [topic],
significant opportunities for advancement remain. Table [N] compares our study
to the most closely related work. [Author1] (Year) [their focus], but [key
difference from our paper]. Similarly, [Author2] (Year) [their contribution],
yet [what they don't address that we do]. [Author3] (Year) comes closest to
our work in [dimension], but differs in [key differentiator: method/context/theory].

Our study extends this body of work in [N] key ways. First, [unique dimension 1].
While prior studies have [what they did], we [what we do differently]. Second,
[unique dimension 2]. Third, [unique dimension 3, if applicable].

Contribution Refinement

Based on the positioning analysis, refine the contribution statement:

## Refined Contribution Statement

Based on the positioning analysis, the paper's unique contribution is:

1. **[Contribution 1]** — This is unique because no prior study has [specific gap].
   Closest competitor: [Author (Year)] who [what they did, but not this].

2. **[Contribution 2]** — This extends [theory/method] by [specific extension].
   Unlike [Author (Year)] who [their approach], we [our approach].

3. **[Contribution 3]** — Practical contribution: [specific practical value].
   This goes beyond [Author (Year)]'s [their practical contribution] by [how].

Step 5: SAVE Output

##