rubicon Positioning
Positioning decides what every other module writes. A profile rewritten without it produces competent copy pointed nowhere, which is why "rewrite my headline" so often has to start here.
Positioning is a choice about who to lose. An executive positioned for everything is positioned for nothing, and at senior level the cost of vagueness is not fewer opportunities — it is worse ones.
Inputs
From salience-identity: career record, proof points, target roles, target audiences, and the
explicit "will not claim" boundaries. Ask only for what is genuinely missing.
The five questions
All five, answered concretely. Do not proceed on abstractions — a vague answer here produces vague output everywhere downstream, and the user will blame the copy.
1. Who, specifically?
Not "B2B companies". Company stage, size, model, and the situation they are in. Good: "B2B SaaS, $20M–$200M ARR, PE-backed or post-Series-C, where growth has flattened and the board is asking why spend stopped converting."
2. What problem?
The problem as the buyer experiences it, in their words — not the practice area name. Good: "We're spending more and getting less, and nobody can tell us whether it's the market, the product, or the marketing."
3. How do you solve it?
The method, and specifically what makes it repeatable rather than personal talent. Good: "Fix the measurement layer first, because most stalled funnels are a measurement problem presenting as a creative problem. Then rebuild demand against what the data actually says."
4. Why is that better than the alternatives?
Name the real alternatives: the agency, the internal hire, the other fractional, doing nothing. Doing nothing is usually the most common competitor and the one most analyses skip.
5. So what?
The consequence for the buyer if this works. Business outcome, not activity.
Differentiation test
Run every draft position through this. Failing any one sends it back.
| Test | Fails when |
|---|---|
| Substitution | Swapping in a peer's name leaves it equally true. Then it describes a category, not a person |
| Negation | The opposite is absurd ("I deliver poor results"). Then it is not a position, it is table stakes |
| Evidence | No verified fact supports it. It may still be right — but it is proposed, and the gap is now the priority |
| Exclusion | It repels nobody. A position that offends no one attracts no one |
| Defensibility | It cannot survive "say more about that" from a skeptical board member |
| Memorability | Someone who heard it once cannot repeat it a week later without notes. A position nobody can relay is a position that never travels through a network |
The substitution test is the one that kills the most drafts, and it should.
Check your own work
Before delivering, answer honestly:
- Is the differentiator something a peer could also claim? If yes, it is a category description.
- Does the audience describe a real person, or a demographic segment?
- If the user already had positioning, does this genuinely differ — or did I polish their old framing and present it as new? This is the most common failure and the hardest to notice from inside.
A useful question when differentiation is thin: "What have you done that a peer or an agency told you they don't do?" The answer is usually the real position.
Outputs
Written to ${SALIENCE_HOME:-~/.claude/salience}/positioning.yaml once approved.
- Positioning statement — internal, one paragraph, not copy
- One-liner — ten words or fewer, sayable out loud
- Short version — roughly 30 seconds, for the "what do you do" question
- Differentiators — three at most, each with the proof point behind it and its tier
- Audience variants — the same position expressed for each audience that matters
- Proof map — which fact supports which claim, and where the gaps are
Audience variants
One position, several expressions. The claim does not change; the emphasis does.
| Audience | Leads with |
|---|---|
| Board / search committee | Scope, judgment, track record at comparable scale |
| Founder / CEO | The problem and the speed of the fix |
| PE operating partner | Repeatability, measurement, time to value |
| Peer executive | Point of view and method |
| Prospective team | What it is like to work for this person |
Changing the claim per audience is not variation, it is inconsistency, and it is visible to anyone who reads two of them.
The employed-and-exploring case
An executive positioning for their next role while holding a current one has real constraints: discretion, loyalty, and the fact that current colleagues read the profile. Handle explicitly.
- Position around capability and point of view, never availability
- Never imply dissatisfaction with the current employer
- Signal direction through what is discussed, not through declared intent
- Keep availability in the recruiter-only setting, not in public copy
Career narrative
For a move, a pivot, or a gap, produce a narrative that is true, brief, and forward-facing. Three sentences: what the arc was, what the move is toward, why now. Never defensive, never over-explained. A long explanation signals a problem the short version did not have.
Where there is a genuine gap — a sabbatical, a layoff, a caretaking period — state it plainly in one clause and move to what came of it. The evasive version is always more damaging than the fact.
Anti-patterns
- Positioning on a claim with no proof and no plan to build any
- Positioning that describes the whole function ("full-stack marketing leader") — that is a job description, not a position
- Positioning for a role the user has not decided they want, to keep options open
- Adopting a "category of one" claim because it sounds strong. It has the highest ceiling and fails hardest under scrutiny; write it only when the proof exists
References
references/frameworks.md— the five questions expanded, competitive mapping, worked examplesreferences/executive-narrative.md— moves, pivots, gaps, and the employed-and-exploring case