Landingpage cro
Skill Gekkos-tech/agency-os/plugins/ad-creative/skills/landingpage-cro
The complete Claude Code toolkit for web & advertising design agencies — 6 plugin bundles for the full agency workflow
npx -y skills add Gekkos-tech/agency-os --skill landingpage-croAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
2 things to look at
- 27 days oldThe repository was created 27 days ago. New is not bad, but a brand new repository carrying a familiar-sounding name is the shape a typosquat arrives in, and there has been no time for anyone else to find a problem with it.
- 4 stars4 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
What its author says it does
Copied from the file, not written here
Structures and audits landing pages for conversion — the bridge between design, SEO, and copy. Use when the user asks about "landing page structure", "CRO", "conversion optimization", "why isn't the page converting", "landingpage aufbauen", or needs a section-by-section page architecture, friction audit, message-match check, or A/B testing plan. Covers the conversion narrative sequence (hero → credibility → problem/solution → offer → proof → objections → risk reversal → CTA), DACH trust elements, mobile-first conversion rules, and a measurement plan. NOT for visual design systems, typography, or component styling (use design-foundation skills), NOT for SEO audits, rankings, or keyword research (use seo-marketing skills), NOT for writing the copy itself (use ad-copy).
SKILL.md
8.5 KB, ~1.8k tokens by cl100k_base, as published. Nobody here has run it
Landing Page CRO — Conversion-Focused Page Structure
Design the argument of a landing page: which sections, in which order, doing which persuasive job. Output is a structural blueprint plus audit findings — not visual design, not finished copy.
© GEKKOS Tech. Licensed under MIT.
Two modes
- Build mode (user is creating a page): run the message-match check, then produce the section sequence with per-section jobs and must-contain elements, then the measurement plan.
- Audit mode (user asks "why isn't it converting"): fetch/read the page, check message match first, then walk the section sequence for missing/misordered jobs, then run the friction checklist. Report findings ordered by expected conversion impact, not by page order.
Message-match rule (check this first, always)
The single highest-leverage CRO factor. Two equations must hold:
- Ad promise == hero promise. Whatever the ad (or email, or post) promised is the first thing the hero must confirm — same offer, same benefit, same wording family, same visual world. A visitor who clicked "30% off onboarding" and lands on a generic homepage-style hero bounces before reading anything else.
- Keyword == H1 intent. For search traffic, the H1 must answer the query's intent (not necessarily echo the keyword verbatim, but resolve it). Someone searching "payroll software comparison" must land on a comparison, not a signup pitch.
If traffic comes from multiple distinct promises, recommend separate landing page variants per promise — one page per message, not one page for all messages.
The conversion narrative — section sequence
Default sequence for a cold-to-warm conversion page. Each section has ONE job; a section doing two jobs is doing neither. Layout patterns per section are in references/section-patterns.md.
| # | Section | Job | Must contain |
|---|---|---|---|
| 1 | Hero promise | Confirm the click was right, state the core value in 5 seconds | H1 = main benefit (message-matched), supporting subline (how/for whom), primary CTA, hero visual showing product/outcome, above the fold on mobile |
| 2 | Credibility strip | Answer "can I trust these people?" before any selling | Client/press logos OR review score OR key number ("12,000 customers"); compact, no headline needed |
| 3 | Problem / solution | Create recognition ("that's my problem"), then map product to it | Problem stated in the visitor's words, 2–4 pain points, solution framed as the mechanism (HOW it fixes it, not just "it fixes it") |
| 4 | Offer detail | Make the offer concrete and evaluable | What exactly they get (features as benefits), pricing or pricing-path, scope/timeline, visual of the actual product/service |
| 5 | Social proof | Let others make the claims the brand can't | 2–3 testimonials with full name + photo/company (specific results > praise), case numbers, ratings with source |
| 6 | Objection handling / FAQ | Remove the last "yes, but…" | 4–7 real objections (price, effort, switching, "will it work for MY case", contract terms) answered honestly; sourced from sales/support if available |
| 7 | Risk reversal | Make saying yes safer than saying no | Guarantee, free trial, free cancellation, "no credit card required", data portability — whichever is true |
| 8 | Final CTA | Convert the reader who scrolled everything | Repeat primary CTA + one-line benefit recap; optionally a lower-commitment secondary path (newsletter, demo) for not-ready visitors |
Adaptation rules:
- Hot/retargeting traffic: compress 2–3, expand 6–7 — they know the problem, they have objections.
- High-ticket B2B: insert a "process/how we work" section after 4; the offer is a relationship, buyers need to see the path.
- Simple/cheap offer: sections 3+4 can merge; never cut 2 or 5.
- Never place the first CTA below section 3 — some visitors arrive ready.
Friction audit checklist
Run in audit mode; recommend preemptively in build mode.
- Form length: every field costs conversions. Justify each field against "do we need this BEFORE the sale?" Name+email beats full address forms for lead gen; move qualification to step 2 (multi-step forms convert better than long single forms).
- Cognitive load: one page = one goal = one primary CTA verb. Count distinct decisions the page asks for (plans, options, links); more than one primary decision → simplify or sequence.
- CTA competition: navigation menus, footer link farms, outbound links, and multiple differently-worded CTAs all leak clicks. Landing pages for paid traffic should strip or minimize global navigation.
- Load speed: every second of load costs conversions, most brutally on mobile paid traffic. Flag: hero images >200KB, render-blocking scripts, autoplaying video above the fold, unoptimized fonts. (Deep performance auditing → hand to the SEO/technical skills.)
- Clarity frictions: unexplained jargon in H1, benefit hidden behind cleverness, price nowhere findable, CTA label vague ("Submit", "Mehr erfahren" as primary).
Trust elements for DACH markets
German-speaking visitors are measurably more trust-sensitive. On any DE/AT/CH page, check:
- Impressum + Datenschutz visibly linked (legally required in DE; absence destroys trust AND is abmahnfähig). Footer at minimum.
- Real company data: full legal name, address, contact — anonymity reads as scam.
- Reviews with sources: Google, Trusted Shops, ProvenExpert, OMR Reviews — named platforms beat unsourced star graphics.
- Certifications/badges where genuine: TÜV, Trusted Shops, ISO, "Made in Germany", DSGVO/GDPR compliance statement near forms.
- Payment/data reassurance at the form: SSL mention, "Ihre Daten werden nicht weitergegeben", recognizable payment logos for shops.
- Du/Sie consistency with the ad that brought the visitor (mismatch = message-match violation).
Mobile-first conversion rules
Most paid traffic is mobile; audit mobile first, desktop second.
- Above-fold offer clarity: on a 390px viewport, H1 + subline + CTA must be visible without scrolling; hero image must not push the CTA below the fold.
- Sticky CTA: persistent bottom CTA bar (or sticky header CTA) once the hero scrolls away; label stays constant with the primary CTA.
- Tap targets: min ~44×44px, full-width buttons on mobile, adequate spacing — no adjacent-link misfires.
- Forms on mobile: correct input types (email/tel keyboards), autofill enabled, no dropdowns where 3 radio buttons work.
- Thumb reach: primary actions in the lower half of natural viewport positions where possible.
- Test the page at 390px width before shipping any structure recommendation.
Measurement plan
Every blueprint or audit ends with a measurement section:
- Primary conversion: the ONE action that defines success (purchase, qualified lead, booked demo). One per page.
- Micro-conversions: scroll depth past section 3, CTA clicks, form starts vs. form completions, FAQ interactions, time on pricing. These diagnose WHERE the page loses people.
- What to A/B test first, in order of typical leverage: (1) hero H1 / offer framing, (2) hero CTA label + placement, (3) social proof position (above vs. below offer), (4) form length, (5) risk-reversal prominence. Test one variable at a time; state a hypothesis per test; don't test button colors while the headline is unproven.
Boundaries
Structure and argumentation only. Visual design (colors, typography, components) → design-foundation skills. SEO audits, rankings, technical crawl issues → seo-marketing skills. Writing the actual headlines, body copy, and CTAs → ad-copy skill; this skill specifies WHAT each section must say, ad-copy writes HOW.