Persuading

Purpose

Apply the principles that determine whether value communication lands — not tips or templates, but the structural causes of why genuine value fails to be recognised, and how to fix them.

This skill is a specialisation of communicating-in-writing, applied when the goal is to make genuine value visible. It adds three things that communicating-in-writing does not cover: how to excavate value the communicator cannot see clearly (the translation problem), how to position that value against alternatives (the reference point problem), and how to manage the psychological barriers that make self-promotion feel wrong even when the value is real.

The default methodology is extract → frame → express. Work through the phases in order. Skipping extraction and going straight to expression is the most common cause of value communication that fails.

This skill operates at medium freedom — follow the extract-frame-express sequence in order, but adapt the depth of each phase to the mode and context.

Scope

Covers three operating modes depending on how this skill is invoked:

1. Preparation mode — dialogue with a user preparing for a pitch, self-promotion task, or positioning decision. Work through the methodology interactively.
2. Review mode — diagnosing existing content (a LinkedIn post, proposal, cover letter) to identify where it fails and how to fix it.
3. Agent mode — producing persuasive content autonomously from a brief, without dialogue.

Out of scope: negotiation and interest-based persuasion, manipulation techniques, copywriting templates, cross-cultural adaptation, format-specific execution (slide rendering, PDF generation).

The Translation Problem (Why This Skill Exists)

Most value communication fails before a word is written. Three systematic miscalibrations cause this:

The Curse of Knowledge — the better someone knows their subject, the worse they communicate it to those who don't. Experts skip context, assume shared vocabulary, and abstract where they should concretise. The fluency effect compounds this: expertise makes a subject easy for the expert to process, and they mistake their own ease for the subject's simplicity.

The visibility gap — people systematically underinvest in making their value visible, then overestimate how apparent it already is. Excellent work does not advertise itself. Making good work noticeable is part of doing good work.

The empathy gap in self-promotion — self-promoters overestimate positive reactions and underestimate negative ones. They cannot easily imagine that their audience doesn't share their enthusiasm. This produces either under-communication (saying nothing) or miscalibrated communication (saying too much about the wrong things).

These are calibration problems, not character flaws. The methodology corrects them structurally.

Inputs

Before applying this skill, establish:

• Subject — what or who is being positioned (a person, product, proposal, idea)
• Audience — who needs to recognise the value, and what they already know and care about
• Goal — what should happen after the communication lands (a decision, a response, a belief change)
• Context — where this communication will appear and what constraints apply
• Existing material — any prior attempts, background documents, or source content (required for review mode)

In preparation mode, gather these through dialogue. In agent mode, they must be provided upfront. In review mode, the existing content substitutes for most inputs but audience and goal must still be established.

Outputs

This skill does not produce files directly. What it produces depends on mode:

• Preparation mode — a positioning strategy: a clear articulation of the value, the right frame, and talking points the user can act on
• Review mode — a structured diagnosis: which phases the existing content fails, why, and specific revisions to fix each failure
• Agent mode — the written artefact itself (post, cover letter, proposal, pitch narrative, or other), complete and ready to use

Phase 1 — Extract

Excavate the genuine value before attempting to communicate it.

In review mode, audit rather than excavate. Check which jobs the existing content addresses and which it ignores. Apply the "So What?" test to claims already made. Identify what is missing or miscalibrated before moving to Phase 2.

Identify the real job being done.

Apply the Jobs to Be Done lens. People do not hire products or people for their features and credentials — they hire them to get a job done. Ask:

• Functional job: what practical task does this solve?
• Emotional job: how does the audience want to feel after this? (confident, relieved, respected, safe)
• Social job: how do they want to be perceived by others because of this decision?

Most value communication addresses only functional jobs. Emotional and social jobs are often more powerful motivators. Surface all three and identify which is primary for this audience.

Map the competitive landscape honestly.

Apply Dunford's five questions:

1. What would the audience do if this didn't exist? (real alternatives, not imagined ones)
2. What does this offer that those alternatives genuinely lack?
3. What does that translate to — not capabilities, but outcomes?
4. Which audience segment genuinely cares about that value?
5. Which framing makes the strengths self-evident before any argument is made?

The competitive alternative question is the most important. It establishes the reference point. People never evaluate anything in isolation; they evaluate relative to something. Control what they compare against, and you control the evaluation.

Apply the "So What?" test to every claimed benefit.

Features and credentials are not value. Translate them:

• Claim: "Five years of experience in distributed systems" → So what?
• Outcome: "Reduced deployment failures by 60% in three production environments" → So what?
• Deeper value: "The team ships faster and the on-call rotation sleeps"

Keep asking until you reach the genuine human outcome. That is what belongs in the communication.

Check for curse-of-knowledge symptoms.

• Is any term unfamiliar to the audience? Replace or define it.
• Is any context assumed that the audience doesn't have?
• If a stranger read this cold, what would they be confused about?

Phase 2 — Frame

Determine the evaluative context before choosing what to say.

In review mode, diagnose the existing frame before proposing a new one. What category does the content implicitly claim? What reference point does it establish? Is the subject positioned as protagonist or guide? Identify the framing failure before rewriting.

Choose the right category.

Category choice triggers a set of assumptions — about price, alternatives, and expectations — before any other communication occurs. The wrong category loses the conversation before it begins.

A database positioned as competing with Oracle activates one set of assumptions. The same database positioned as a BI tool activates a different set. Neither changed what the product does. The frame changed what the audience evaluated it against.

Apply this to individuals: "senior developer" activates different assumptions than "platform engineer who has led three zero-downtime migrations — the person you call when the deploy can't fail." The first is a job title; the second is a positioning statement that narrows the category, names the outcome, and implies the audience (high-stakes infrastructure teams). Be the best option in a narrower frame rather than an average option in a broad one.

Control the reference point.

Loss-framed messages are often more motivating than gain-framed ones: what is lost by not acting is felt more acutely than what is gained by acting. But the effective choice depends on the audience's current state.

The principle: decide explicitly what the audience should compare this against. Present the status quo as the benchmark, then show the improvement. Or surface the cost of inaction, then offer the alternative. The reference point is a choice, not a given.

Position yourself as guide, not protagonist.

The most common communication mistake: talking about yourself when you should be talking about the audience's transformation. The StoryBrand principle: the audience is the hero. You are the guide — demonstrating understanding of their problem and credible ability to resolve it, without competing for the protagonist role.

Translate: instead of "I have 10 years of experience in X", ask what the audience's problem is and what transformation they want. Show you understand that, then provide the evidence that makes you the credible guide to it.

Sequence emotion before evidence.

Emotional processing precedes rational evaluation. The audience's felt response to your communication shapes their reception of your argument before they consciously consider it. This is not a counsel to manipulate — it is a sequencing rule.

Practical order: (1) establish that you understand the audience's situation (emotional recognition), (2) surface the problem or tension clearly (complication), (3) provide the evidence and argument (resolution). Evidence delivered before emotional recognition lands poorly because the audience hasn't yet agreed there is a problem worth solving.

Phase 3 — Express

Write and organise the communication so the value you've extracted lands clearly in the frame you've established.

In review mode, diagnose before revising. Score the existing content against the SUCCESs checklist and identify the Proof + People + Place gaps. Fix the root causes, not surface symptoms. Rewrite only what the diagnosis points to.

Apply the SUCCESs diagnostic.

Use this as a checklist before and after drafting:

Mediocre content actively harms credibility in saturated contexts. Rare excellent content compounds over time. The quality bar matters.

Lead with the answer, build with evidence.

State the core claim first. Audiences need the conclusion before the reasoning — not after. This applies to proposals, cover letters, LinkedIn posts, and pitch narratives equally.

Exception: when the audience needs emotional preparation before they can accept the conclusion (strongly contrary to current belief, delivering difficult news), build briefly to the claim. This is deliberate, not a default.

Build credibility through the Proof + People + Place structure.

Three credibility signals multiply when combined:

• Proof — verifiable data, specific outcomes, quantified results
• People — third-party validation: testimonials, endorsements, citations, who else says this
• Place — contextual grounding: where did this happen, in what conditions, for whom

Research on LinkedIn content found that posts with all three signals generate 333% higher engagement than posts lacking them, and the signals multiply rather than add. In any high-stakes communication, check for all three.

Use third-party validation strategically.

Third-party confirmation of claims is more credible than self-assertion and psychologically cheaper to deploy. Build structures that generate it rather than relying on direct boasting:

• Request LinkedIn recommendations for specific outcomes, not general praise
• In proposals and applications, include case study quotes rather than abstract claims
• Co-create content with respected collaborators — their association transfers credibility
• Reference who else has endorsed or used your work before making claims about it

Apply the Grunt Test before finishing.

Could someone with no prior knowledge understand what you offer and why they should care within five seconds? If not, the problem is clarity, not the offer. Revise the opening until it passes.

Addressing Self-Promotion Discomfort

For people who find self-promotion uncomfortable, three reframes matter:

Your discomfort is invisible; your value is also invisible.

Research on the illusion of transparency shows people dramatically overestimate how visible their internal states are to others. Your anxiety when self-promoting is far less visible than it feels. Your credentials and expertise are also less visible than you assume — they must be explicitly communicated. Both effects run in the same direction: you need to communicate more, not less.

Self-promotion discomfort is disproportionately distributed.

Research documents a persistent gap where equivalent performance is rated significantly lower by some groups than others, due to socialised backlash risk rather than strategic modesty. "Just be more confident" is structurally inadequate advice. The practical correction is to build structures that generate third-party validation rather than requiring direct assertion — this bypasses the psychological cost of self-promotion without sacrificing visibility.

Demonstrating is not the same as declaring.

The most effective and psychologically accessible form of visibility is content that demonstrates thinking and value without requiring explicit self-assertion. Writing, case studies, and teaching create visible evidence that audiences evaluate directly. This is self-promotion that does not feel like self-promotion — and research on thought leadership confirms it is more trusted than direct marketing by a significant margin.

Anti-Patterns That Backfire

These approaches reliably damage credibility and should be avoided:

Humblebragging. Bragging masked by complaint or false humility ("I'm so overwhelmed by all these speaking invitations") performs worse than straightforward bragging and worse than honest complaints — on likability, competence ratings, and resource allocation. The mechanism is perceived insincerity. The combination of self-promotion and false modesty creates cognitive dissonance that observers read as manipulative. If you are going to acknowledge an achievement, do it directly or let someone else do it.

Over-claiming without substance. Positioning that outpaces the actual evidence triggers backlash when the gap is detected. Surface-level authenticity signals (casual tone, behind-the-scenes content) show negative correlation with engagement when not accompanied by substantive credibility. The sustainable direction: under-claim and over-deliver consistently. Reputation compounds; a single credibility collapse takes years to recover from.

Category-level over-reach. Positioning in a competitive category where you cannot plausibly be best activates unfavourable comparisons before you make any argument. "Best in class" in a category dominated by well-known alternatives is noise; "best for [specific segment]" in a narrower frame can be meaningful.

Credential-leading without outcome-connecting. Listing qualifications before establishing the audience's problem fails twice: it talks about you when the audience is thinking about themselves, and it leaves the translation from credential to outcome unfinished. Lead with the problem, surface the outcome, then connect the credential as evidence.

Performative authenticity. Claiming to be authentic, relatable, or genuine as a strategy is self-defeating. Authenticity is a perception produced by consistent behaviour and verifiable substance — it cannot be asserted. "Real" and "authentic" as explicit claims function as red flags in an audience already sceptical of marketing language.

Self-Assessment Checklist

Use this as an evaluate-fix-re-evaluate loop. In preparation mode, apply Rounds 1–2 to the positioning strategy. Rounds 3–4 apply when the strategy is expressed as written content (agent/review modes).

Round 1 — Extraction check (see Phase 1 for full guidance)

• [ ] Functional, emotional, and social jobs identified — which is primary for this audience?
• [ ] Competitive alternatives question applied — what does the audience compare this against?
• [ ] "So What?" test applied to every benefit claim — have you reached a genuine human outcome?
• [ ] Curse-of-knowledge cleared — no unexplained jargon, no assumed context?

Round 2 — Frame check (see Phase 2 for full guidance)

• [ ] Category frame correct — does it trigger the right assumptions before the argument begins?
• [ ] Audience positioned as hero, subject as guide?
• [ ] Emotional recognition before evidence and argument?
• [ ] Reference point explicit — does the audience know what they're comparing against?

Round 3 — Expression check (see Phase 3 for full guidance) [agent/review]

• [ ] Opening passes the Grunt Test — clear claim within five seconds?
• [ ] Proof + People + Place signals present, specific, and verifiable?
• [ ] SUCCESs checklist passes — which dimension is weakest?
• [ ] Anti-patterns absent — no humblebragging, no over-claiming, no performative authenticity?

Round 4 — Calibration check [agent/review]

• [ ] If someone from the audience read this cold, what would they be confused by? What would they not believe?
• [ ] What is the strongest objection to this communication, and is it addressed?
• [ ] Does the communication ask for one thing, clearly, at the end?

Edge Cases

When the subject genuinely lacks competitive differentiation. Do not position around strengths that don't exist. The honest diagnosis is: "The value is not yet communicable because it hasn't been built." Redirect to building a track record rather than framing the absence. The skill cannot make weak value appear strong — it makes real value visible.

When the audience is hostile or sceptical. Hostile audiences process counter-attitudinal arguments through a rejection filter. Lead with the areas of agreement first. Establish shared stakes before introducing the contested claim. Third-party validation carries more weight here than direct assertion. Do not open with your strongest argument — open with the least objectionable one.

When direct self-assertion is structurally impossible. Anonymous submissions, regulated industries, and contexts with explicit prohibition on self-promotion require indirect strategies: work samples speak, referees assert, portfolio items demonstrate. The methodology still applies; the expression channel changes.

When the subject is genuinely novel — no existing category applies. Category-less subjects need to create the category before competing in it. Name what the problem is, demonstrate it exists, then introduce the solution. The audience cannot evaluate what they cannot name.

Evaluation Scenarios

Scenario 1 — Preparing a pitch for a role or contract

Test input: A user says: "I'm preparing for a senior platform engineering role at a Series B fintech. My background: 8 years backend, last 3 years leading infrastructure at a payments startup — we went from 99.5% to 99.97% uptime. The role is about building reliability for their core transaction system. I'm not sure how to pitch without sounding like I'm just listing my CV."

Expected behaviour: Working in preparation mode, the skill asks about the audience (what specific reliability problem does this role exist to solve?), the competitive alternatives (who else is likely applying, and what do they offer?), and the emotional and social jobs (what does the hiring manager need to feel confident about — the system, or their own decision?). The skill does not move to expression until extraction and framing are complete. The output is a positioning strategy: the candidate framed as the guide who has solved this exact problem before, with one concrete outcome as the lead rather than a credential list.

What changed: The frame. The content is drawn from the same experience; the order and lens are different.

Scenario 2 — Diagnosing a LinkedIn post that isn't getting traction

Test input: A product manager shares a LinkedIn post announcing a new feature launch. The post reads: "Excited to share that we've launched [Feature X] — a new capability that lets users customise their dashboard views. This was 6 months in the making, work by a brilliant team. Check it out in the app!" The post has 12 likes and no comments after 48 hours.

Expected behaviour: In review mode, the skill diagnoses the post against the SUCCESs framework before suggesting any revision. Expected diagnosis: Simple (partial — "customise dashboard views" is the claim, but it's about the product, not the audience); Unexpected (fails — "excited to share" is the most predictable LinkedIn opening); Concrete (fails — no specific numbers, no before/after, no user outcome); Credible (fails — no Proof + People + Place signals); Emotional (fails — no social or emotional job addressed — who does this make the user look like to their colleagues?); Stories (fails — no narrative arc). The revision addresses Unexpected and Concrete first, then adds one social proof signal. The diagnostic is systematic, not impressionistic.

What changed: Root causes identified and fixed, not surface symptoms polished.

Scenario 3 — Writing a proposal for a new initiative

Test input: A team lead wants a proposal for introducing automated incident response tooling. Background: current mean time to resolve (MTTR) is 47 minutes; industry benchmark is 12 minutes; on-call engineers report high stress and context-switching during incidents. Budget required: £40k. Audience: CTO and VP Engineering who have not previously prioritised this.

Expected behaviour: In agent mode, the skill applies the extract → frame → express sequence without prompting. Phase 1 identifies the status quo (47-minute MTTR, stressed on-call rotation) as the reference point and the gap (35 minutes above benchmark) as the complication. Phase 2 frames the proposal as solving a problem the CTO already has — not as an ask for new budget but as a fix for a known cost. Phase 3 structures output as: situation (current state and its cost) → complication (gap and consequence) → resolution (the tooling and what it changes) → evidence (benchmark comparison, case reference if available) → call to action (decision required). Proof + People + Place signals are embedded. The reader is met where they are before the solution is offered.

What changed: Structure and sequence — the reader's problem is named before the solution is proposed.

Scenario 4 — Helping someone uncomfortable with self-promotion

Test input: A user says: "I need to update my LinkedIn profile but I hate talking about myself. I'm a UX researcher — I've run 200+ user interviews, two of which directly changed product direction. My manager says my profile is too modest. I don't want to sound like I'm bragging."

Expected behaviour: In preparation mode, the skill does not offer encouragement and does not simply suggest listing accomplishments. It reframes the task: the profile exists to make value discoverable for people who already need what this person offers — it is a service, not boasting. It then guides extraction: the two product-direction changes are the lead, not the 200+ interviews (outcomes, not activity metrics). The social job is surfaced: what does a hiring manager or collaborator need to believe about this person? The skill suggests building the third-party validation section (recommendations, featured work) as the primary credibility vehicle, minimising direct self-assertion. The user can produce a credible profile without feeling they are bragging because the task has been reframed from "promote yourself" to "make your evidence accessible."

What changed: The psychological frame and the extraction lens — outcomes lead, third-party signals carry the credibility weight.