Landing Page Complete Audit (12 Modules)
Purpose
Run a complete, standalone audit of an existing landing page across all 12 modules in one invocation. Deliver one coherent report with a scorecard, module findings, priorities, three proposed experiments and a 30-day roadmap. This skill does not require or call the separate module skills; it contains the complete workflow itself. Each separate module skill remains independently usable.
Workflow
1. Confirm the minimum audit context
Review the user's message and attachments before asking questions. Do not ask for facts already provided. Before the full audit, obtain essential context if missing:
- An accessible landing-page URL, or screenshots/recording/page text if the URL cannot be accessed.
- Business/brand and product or service.
- Intended audience/market and primary conversion goal.
- The offer and its key terms at a level sufficient to judge the page's promise.
Ask a concise, consolidated follow-up for only essential missing items. Do not require unavailable analytics, A/B test history or metrics. Request optional details when available: business model, secondary goal, ICP pains/desired outcomes/objections, pricing, differentiators, ad/campaign and traffic sources, brand constraints, competitors, testimonials/credentials, traffic/conversion data, analytics tools and prior tests. Mark anything missing as unavailable, and proceed once the minimum facts needed for grounded review are provided.
If the page cannot be accessed, request screenshots, a screen recording, exported text or HTML. Restrict conclusions to the material supplied. Do not claim to have tested interactions, mobile behavior, performance or accessibility when only static evidence is available.
2. Inspect the page and collect evidence
Review the entire page from hero to footer, including visible headings, copy, images, navigation, forms, CTAs, price/terms, trust elements and interactive steps that are actually accessible. Understand the intended visitor, traffic source (if supplied), offer and desired action. Record a brief audit scope and evidence basis.
For every significant finding, distinguish:
- Observed: directly visible or actually tested.
- Reported: supplied by the business/user.
- Inference/Hypothesis: plausible but unverified.
- Recommendation: proposed action, not a proven result.
Tie findings to a section, element, supplied asset, source URL or measurement when possible. Quote exact copy only when available; otherwise label it as a paraphrase. Never fabricate facts, user research, analytics, testimonials, results, guarantees, certifications, competitors, page behavior or test outcomes.
3. Evaluate all 12 modules
Complete each module below independently. Do not skip a module silently. If evidence is insufficient, mark it Not Evaluated or report a limited review, explain why, and exclude it from the score average.
- UX Audit: navigation and page structure; information architecture; visual hierarchy; content organization; readability/typography; spacing consistency; findability; interactions; cognitive load; clarity of visitor journey.
- CRO Audit: conversion-goal clarity; competing actions and distractions; friction; form/checkout/enquiry complexity; offer presentation; objection handling; decision barriers; five-second clarity; message relevance; potential friction reduction. Treat conversion effects as hypotheses unless tested.
- Content & Copywriting Audit: headline/subheadline and opening; problem/solution explanation; feature-to-benefit translation; emotional relevance; clarity; persuasion; structure; consistency; objection handling; closing CTA. Use AIDA, PAS, BAB or another framework only when it fits. For high-impact copy issues, show existing copy or marked paraphrase, weakness, why it matters, replacement copy and rationale. Do not rewrite the entire page unless asked.
- CTA Audit: primary/secondary, hero, mid-page, final and sticky CTAs where present; wording, action clarity, visibility, contrast, size/accessibility, placement, frequency, consistency, offer relevance, destination/functionality and mobile use if inspected. Mark untested destinations as unverified.
- Trust & Credibility Audit: testimonials, reviews, case studies, results, credentials, certifications, client/partner logos, guarantees/refunds, security/privacy, identity/contact details and authenticity/relevance of proof. Never create or embellish proof.
- Mobile Experience Audit: responsive reflow; text/headings; button and touch usability; navigation; image scaling; forms; spacing; overflow; fixed/sticky UI; CTA visibility and mobile conversion path. State whether evidence is actual mobile testing, responsive inspection, supplied screenshots or unverified.
- Performance Audit: available loading data/Core Web Vitals; images/fonts; JS/CSS; render-blocking resources; layout shifts; third-party scripts; lazy loading; network/resource efficiency. Identify lab vs field data and provenance. Reference thresholds: LCP good ≤2.5s, needs improvement >2.5–4s, poor >4s; INP good ≤200ms, needs improvement >200–500ms, poor >500ms; CLS good ≤0.1, needs improvement >0.1–0.25, poor >0.25. Do not invent measurements or claim a pass without data.
- Competitor Benchmarking Audit: compare only competitors supplied by the user or authorized for review. Compare positioning/headline, audience, value proposition, offer/pricing presentation, benefits, CTA, structure, visuals, trust, UX, differentiation and journey. Cite the pages and access date when possible. Never infer competitors' conversion rates or business results without data.
- Offer & Value Proposition Audit: offer clarity, outcome, audience fit, perceived value, features vs benefits, differentiation, pricing/terms, presentation, bonuses, risk reversal, guarantee and audience-offer fit. Use the Value Equation diagnostically: Dream Outcome, Perceived Likelihood of Achievement, Time Delay, Effort/Sacrifice. Do not invent offer elements.
- Message Match & Traffic Alignment Audit: compare supplied ad/campaign/email/search promise with page headline, audience, offer, pricing/conditions, visuals, CTA, search intent and conversion journey. If source materials are unavailable, state that complete message-match evaluation is not possible; do not infer an unseen campaign.
- Accessibility & Inclusive Design Audit: use WCAG 2.2 as a reference. Review contrast, readability, heading hierarchy, semantics, alt text, keyboard navigation, focus, form labels/errors, screen-reader considerations, controls, reduced motion and inclusive language. Distinguish visual inspection, automated checks, manual keyboard tests and assistive-technology tests. Never claim WCAG compliance without adequate testing.
- Strategic Recommendations & Optimization Roadmap: consolidate evidence-based opportunities, dependencies and business relevance; prioritize by potential impact, effort, evidence and confidence. Create a prioritized action plan, three feasible experiments and a practical 30-day roadmap. Score the quality/completeness of the roadmap—not the page.
4. Score modules consistently
Score applicable criteria from 0–5: 0 missing/broken/critically deficient; 1 very poor; 2 below expectations; 3 acceptable; 4 very good; 5 excellent. Use N/A when a criterion cannot reasonably be evaluated and exclude it from the denominator.
Module score = (points earned ÷ maximum applicable points) × 100
Round to a whole number and state the numerator/denominator or explain the scored criteria. If no meaningful criteria can be scored, report Not Evaluated, not 0. Calculate the provisional overall score as the equal-weight average of evaluated module scores. Exclude Not Evaluated modules and disclose how many of 12 were included. Do not imply the diagnostic score is an industry benchmark or predicts conversion performance.
Interpret overall score: 90–100 Exceptional; 80–89 Very Good; 70–79 Good; 60–69 Needs Improvement; 40–59 Poor; 0–39 Critical.
Assign evidence confidence: High = strong directly observable/measured evidence; Medium = reasonable evidence with some missing information; Low = limited evidence or substantial assumptions. Assign priorities: P0 Critical serious issue requiring immediate attention; P1 High important potential impact; P2 Medium valuable after higher priorities; P3 Low minor refinement/longer-term opportunity.
5. Build the report
Deliver one readable Markdown report in this order:
- Executive Summary: business/page overview; primary conversion objective; provisional overall score and number of modules included; strongest findings; most serious weaknesses; biggest opportunities; overall evidence confidence.
- 12-Module Scorecard: module, score/rating or Not Evaluated, main finding, confidence. Include all 12.
- Detailed Module Findings: for each module, overview, what works, problems/evidence, why they matter, practical recommendations, priority and score. Provide enough detail to substantiate conclusions without repeating boilerplate.
- Copy & Conversion Improvements: where relevant, existing copy (or labeled paraphrase), issue, proposed replacement and rationale.
- Competitor Benchmarking: comparison against supplied competitors; explain if none were supplied.
- Prioritized Recommendations: consolidate duplicate actions and rank by P0–P3; indicate owner/effort/dependency when reasonably inferable, otherwise label as suggested.
- A/B Testing Plan: propose three experiments, each with name, problem, falsifiable hypothesis, control, variation, primary metric, guardrail metric, expected learning and implementation considerations. Do not promise uplift. If traffic volume is low or unknown, state that statistical feasibility is unknown and suggest qualitative/usability research as an alternative.
- 30-Day Optimization Roadmap: Week 1 critical fixes/quick wins; Week 2 copy, offer and CTA; Week 3 UX, mobile and accessibility; Week 4 measurement, validation and experimentation. Reorder when dependencies or business constraints require it.
- Final Consultant Verdict: current page condition; key conversion barriers; strongest opportunities; immediate next steps; data/tests needed to validate hypotheses.
Avoid fake precision. Do not assign false confidence or numeric impact estimates without supporting data. Where evidence is limited, say so plainly and make the next evidence-gathering step concrete.
Final quality check
Before responding, verify that all 12 modules are represented; no unavailable information was invented; observed facts, reports, inferences and recommendations are distinguishable; mobile, performance, competitor and accessibility claims match actual evidence; scores use the correct denominator; the overall average excludes and discloses unevaluated modules; roadmap score measures roadmap quality; and the experiments/roadmap are feasible and do not promise results.