audit-copywriting — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited audit-copywriting (Agent Skill) and scored it 100/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 0 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 0 flagged
Every scanned point with the score it earned and what moved between them.
First recorded scan — no prior version to compare against.
The primary manifest — the file an agent reads to learn what this artifact does.
This audit does not write new copy. It detects problems in existing copy using a structured review sequence. The goal is to ensure all site text — headlines, subheadlines, body copy, CTAs, interactive element microcopy, error messages, tooltips, footer — is ready for a real visitor before launch.
Scope: all site pages (homepage, solutions/features, industry/use-case pages, process, pricing, contact) and all interactive demo or widget text.
Core principle: if you must choose between clear and creative, choose clear. Between specific and vague, choose specific. Between customer language and company language, choose the customer's.
| Level | Description | Action |
|---|---|---|
| Critical | Copy blocks conversion or contains placeholder text | Fix before launch |
| High | Copy weakens trust or fails to persuade | Fix before launch |
| Medium | Copy is functional but suboptimal | Fix within first week |
| Low | Minor improvement opportunity | Fix when convenient |
Before reviewing text quality, verify that each page has the correct architecture. Well-written copy on a poorly structured page does not convert.
Above the fold:
Page structure — minimum 5 distinct section types:
Section variety check: ❌ Weak: Hero → Feature 1 → Feature 2 → Feature 3 → CTA ✅ Strong: Hero → Social proof → Problem → How it works → Key benefits → Testimonial → Use cases → Objection handling → Final CTA
Flag any homepage that is a vertical list of features with icons.
Review in this order. Each sweep has a specific focus.
Can a non-technical decision-maker understand every sentence on first read?
Does every claim connect to a concrete customer benefit?
After every claim, ask "so what?" If the copy doesn't answer with a deeper benefit, the connection is missing. ❌ "We support PDF, DOCX, XLSX, PPTX and more" ✅ "Upload your documents as they are — no conversion required" ❌ "Semantic search engine" ✅ "Answers in seconds"
Is every claim supported by evidence or correctly softened?
| Problematic claim | Correction |
|---|---|
| "Market leader" | Remove or specify the metric |
| "Thousands trust us" | Don't use if not yet true |
| "The best solution" | "One of the few specialized for [market]" |
| "Guaranteed results" | "Demonstrable ROI in [timeframe]" |
| "Easy to use" | "First search in under 5 minutes" |
Is every claim concrete enough to be credible?
Full vague-to-specific replacement tables and word lists: see references/copywriting-patterns.mdQuick reference — eliminate these words:
What to check:
Does the copy make the reader feel something, or only inform?
Relevant B2B emotions: frustration of current state, relief of the solution, pride in the decision, confidence in implementation.
Techniques:
What to check:
Have we removed every reason not to click?
Weak CTAs to detect and replace:
| Weak | Strong |
|---|---|
| Submit | Request my proposal |
| Sign up | Create my account |
| Learn more | See how it works |
| Get started | Talk to an expert |
| Contact us | Book a 30-minute demo |
What to check:
Full objection anticipation map with 8 core B2B objections: see references/narrative-architecture.mdWell-written copy on a poorly structured argument does not persuade. This layer verifies sales logic across the site.
Full narrative architecture reference with all checks: see references/narrative-architecture.mdEach page must move the visitor through: Recognition → Understanding → Belief → Trust → Confidence.
| Stage | What the visitor needs | Where it must appear |
|---|---|---|
| Recognition | "This understands my problem" | Homepage problem section, industry pages |
| Understanding | "I get how this works" | How it works, process page |
| Belief | "This delivers what it promises" | ROI cases, testimonials, metrics |
| Trust | "I can trust this company" | Social proof, security signals, company info |
| Confidence | "I know what's next and it's low risk" | CTAs, pricing, contact page |
After leaving the site, three things must be remembered:
| Message | Must appear in |
|---|---|
| Core value proposition (one sentence) | Headlines, subheadlines, section summaries, CTA copy |
| Primary differentiator | Above fold, comparison section, FAQ, pricing justification |
| Risk reduction argument | Near CTAs, pricing page, process page, contact page |
Build a site glossary before launch. Verify consistent usage across all pages.
Glossary structure:
| Concept | Approved term | Terms to avoid |
|---|---|---|
| The product | [Official name] | "the platform", "the system" (use sparingly) |
| What it does | [Customer language description] | Generic tech terms without explanation |
| Customer documents | "your documentation" | "the data", "the content" |
| Monthly payment | [Approved local term] | Foreign finance terms if market is non-English |
| The customer | "your team" | "the end user" |
What to check:
Every headline must be classifiable into at least one recognized formula:
| Formula | Example |
|---|---|
| {Outcome} without {pain} | Answer compliance questions in seconds without reviewing hundreds of pages |
| Never {negative} again | Never lose time searching through documents again |
| {Benefit} for {audience} | Instant document search for legal firms |
| Turn {input} into {outcome} | Turn your documentation into searchable knowledge |
| {Question naming pain} | How many hours does your team spend searching each day? |
| {Outcome} in {timeframe} | Your documentation searchable in [X] weeks |
| Stop {pain}. Start {benefit}. | Stop searching. Start finding. |
If the project has no interactive elements, skip this section.
Search across all HTML files for unreplaced strings:
[COMPANY NAME], [Client name], [Your name], [Date], [Number]TODO:, FIXME:, PLACEHOLDER, Lorem ipsumXXX, 000, ???example.com, yourdomain.comCopywriting Audit — [Project Name]
Date: [Date]
Pages audited: [list]
Summary
Critical issues: X
High priority: X
Medium priority: X
Low priority: X
Placeholder strings found: X
Headline formula compliance: [X of Y headlines classifiable]
Narrative architecture: [Complete / Gaps found]
Overall copy readiness: [Ready / Needs work / Critical problems]
Critical Issues
[Issue title]
Layer: [Structure / Text quality / Narrative]
Sweep: [If Layer 2 — which sweep number]
Page: [page name]
Issue: [What is wrong and why it reduces conversion]
Fix: [Specific correction]
High Priority
[Same format]
Medium Priority
[Same format]
Low Priority
[Same format]
Recommended Fix Order
Placeholder text — cannot launch with unfilled strings
Headline formulas — first impression determines bounce rate
CTA copy — direct conversion impact
Specificity sweep — eliminate vague claims across all pages
Objection coverage — address unanswered buyer questions
Narrative architecture — ensure five-stage progression
Voice and tone consistency — polish passFull pre-launch checklist: see references/checklist.md~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.