Purpose
Evaluate whether a specific piece of content genuinely serves its intended reader at their stage in the buying journey — producing a weighted relevance score, a dimension-level diagnosis of where fit breaks down, and a ranked set of edits that will move the number. Replaces gut-feel "does this feel on-brand?" reviews with a repeatable content-to-customer-need scoring rubric.
When to Use
Invoke this skill when the user asks to:
- Score or grade a piece of content for relevance to a target persona
- Check whether an asset matches the customer journey stage it's meant for
- Audit a content plan or editorial calendar for persona-message fit
- Decide whether to publish, revise, kill, or repurpose a draft
- Diagnose why a published asset isn't resonating or converting
- Compare two content directions for the same audience
Inputs Required
Before starting, collect:
- Content to score: The full draft, outline, or content plan item (headline, body, CTA)
- Content type: Blog post / email / ad / landing page / social post / whitepaper / video script
- Target persona: Role, goals, pain points, objections, and level of sophistication
- Journey stage: Unaware / Problem-aware / Solution-aware / Product-aware / Purchase-ready
- Intended action: What the reader should think, feel, or do next
- Brand voice notes: Tone, terminology, and positioning constraints (if any)
If persona detail is thin, infer a working persona from the content and state the assumption. If journey stage is missing, diagnose the stage the content currently speaks to and flag any mismatch with intent.
Instructions
Step 1: Establish the Fit Baseline
Restate, in 2–3 sentences, who this content is for and what job it must do at this journey stage. Name the single most important thing the reader needs from content at this stage:
- Unaware: Name a problem they haven't articulated yet — no product talk
- Problem-aware: Validate the pain and frame the stakes — light on solutions
- Solution-aware: Compare approaches and teach evaluation criteria — not yet selling the product
- Product-aware: Differentiate this product against alternatives — proof and specifics
- Purchase-ready: Remove friction and risk — pricing clarity, guarantees, next step
Flag immediately if the content's actual stage disagrees with the intended stage — this is the most common and most costly relevance failure.
Step 2: Score Across 6 Relevance Dimensions
Score each dimension 0–10 with one-sentence evidence quoting or paraphrasing the content. Apply the weights shown to compute the final score.
Dimension 1 — Persona Alignment (weight ×2.0) Does the content speak to this persona's actual role, sophistication, and vocabulary?
- Deduct for jargon above or below the reader's level, wrong seniority framing, or generic "everyone" targeting
- Full marks require language the persona would use themselves
Dimension 2 — Journey Stage Match (weight ×2.5) Does the content deliver what this stage requires (per Step 1)?
- Deduct heavily for premature selling (product talk to unaware readers) or under-selling (no CTA for purchase-ready readers)
- This is the highest-weighted dimension because stage mismatch invalidates otherwise strong content
Dimension 3 — Pain/Need Resonance (weight ×2.0) Does the content name a pain or desire the persona genuinely feels, in their emotional terms?
- Deduct for feature-first framing, assumed pains not tied to the persona, or benefits stated as claims without stakes
- Full marks require at least one line the reader would recognize as "that's exactly my situation"
Dimension 4 — Message Clarity & Specificity (weight ×1.5) Can the reader extract the core message in one pass, backed by specifics?
- Deduct for abstraction, hedging, buzzwords, and unquantified claims
- Reward concrete numbers, named scenarios, and a single dominant idea
Dimension 5 — Proof & Credibility (weight ×1.0) Does the content earn belief appropriate to the stage?
- Later stages demand more proof (data, cases, specifics); earlier stages need credibility of framing, not testimonials
- Deduct for claims that outrun the evidence provided
Dimension 6 — Action Fit (weight ×1.0) Is the CTA the right size for the stage and does it lower the reader's next-step friction?
- Deduct for asking too much too early or offering no clear next step when the reader is ready
- Full marks: the ask matches reader readiness and states exactly what happens next
Scoring math: Multiply each raw score by its weight, sum, and normalize to a 0–100 scale (total possible weighted points = 100). Present the final number plus the per-dimension breakdown.
Step 3: Diagnose the Weakest Links
Identify the 2–3 lowest-scoring dimensions and explain, with reference to the actual content, why each is dragging the score. Distinguish between:
- Fatal flaws: Stage mismatch or wrong persona — these cap the ceiling and must be fixed first
- Leaks: Weak resonance, vague messaging, or thin proof — fixable with edits
- Polish: Tone and CTA refinements — worth doing but not score-defining
Step 4: Rank Improvements by Score Impact
Produce a numbered list of 5–8 specific, applyable edits ordered by how much each will lift the score. For each: name the dimension it fixes, the concrete change (rewrite the line, cut the section, reframe the CTA), and the estimated point gain. Include at least one before/after rewrite of the single highest-leverage line.
Step 5: Verdict
Give a clear call — Publish as-is / Revise (specify must-fix edits) / Repurpose for a different stage / Rebuild — with a one-line rationale tied to the score.
Output Format
- Structure: Fit baseline → scored dimension table → weakest-link diagnosis → ranked improvements → verdict
- Length: 700–1,000 words
- Format: Scored table for the 6 dimensions; numbered ranked list for improvements; prose for diagnosis and verdict
- Sections: Fit Baseline → Relevance Score (with per-dimension table and final /100) → Diagnosis → Ranked Improvements → Verdict
- Tone: Editorial director reviewing a draft — direct, evidence-cited, no praise padding
Quality Checklist
Before delivering output, verify:
- [ ] All 6 dimensions are scored 0–10 with one line of content-specific evidence each
- [ ] The weighted final score is computed correctly and normalized to /100
- [ ] Any stage or persona mismatch is flagged explicitly as a fatal flaw
- [ ] Ranked improvements are ordered by score impact, not listed arbitrarily
- [ ] At least one concrete before/after rewrite of the highest-leverage line is included
- [ ] The verdict is one of the four defined calls — no hedging
Notes
- Journey stage match is weighted highest on purpose: excellent copy aimed at the wrong stage still fails. Diagnose stage before critiquing craft.
- When scoring a full content plan rather than one asset, score each item and surface gaps — journey stages with no content and stages that are over-covered.
- Low scores driven by "polish" dimensions are worth publishing through; low scores driven by "fatal flaws" are not — the ranked list should make that trade-off obvious to the buyer.
How to use this skill
This file is a Claude Code skill in Anthropic's standard format — markdown with YAML frontmatter at the top.
Installing in Claude Code:
~/.claude/skills/<folder-name>/SKILL.md— auto-loads in every session anywhere on your machine (recommended for skills you'll use across multiple projects)<your-project>/.claude/skills/<folder-name>/SKILL.md— auto-loads only when working in that project (recommended for project-specific skills)
Restart Claude Code or open a new session to register.
Using in Claude.ai, ChatGPT, Gemini, or other AI tools:
- Drag this file into a new chat, or copy and paste its contents
- Send your task or request as your next message — the AI will apply the skill where it fits
For Claude Code skill details: https://docs.anthropic.com/en/docs/claude-code/skills