- Home
- Skills
- Research & Analysis
- Landing Page Conversion Auditor
Works with the AI Agents you already use
Landing Page Conversion Auditor
Evidence-based conversion and technical audit skill that diagnoses friction and generates implementation-ready tickets.
Free
Landing Page Conversion Auditor
Example session with this skill installed
Audit the provided HTML and screenshot for a B2B SaaS signup page. The goal is trial starts. Traffic comes from LinkedIn ads promising a '14-day free trial'.
- Read your context and instructions
- Compiled the landing page conversion
Audit complete for B2B SaaS signup page. I have mapped the decision path and identified three P1 blockers related to message mismatch and form friction.
Conversion Audit: SaaS Trial Funnel
| ID | Dimension | Severity | Evidence Status | Finding | Recommended Change |
|---|---|---|---|---|---|
| 001 | Message Match | CRITICAL | OBSERVED | LinkedIn ad promises "14-day trial," but page hero text and CTA specify "Start Free Demo." | Update hero H1 and CTA to "Start 14-Day Free Trial" to maintain intent. |
| 002 | Choice Architecture | HIGH | OBSERVED | Three different CTA styles (Outline, Solid Blue, Ghost) above the fold with equal visual weight. | Standardize primary CTA to Solid Blue; move secondary "View Pricing" to ghost style. |
| 003 | Form Friction | MEDIUM | INFERRED | Form requires Phone Number field without a "Why" tooltip; high likelihood of drop-off. | Make Phone Number optional or add micro-copy explaining it is for 2FA only. |
| 004 | Accessibility | MEDIUM | OBSERVED | Submit button background (#E0E0E0) and text (#FFFFFF) contrast ratio is 1.2:1 (fails WCAG). | Update button CSS to #2563EB for 4.5:1+ contrast ratio. |
Experiment Backlog
- Hypothesis: Clarifying the "Free" nature of the trial in the CTA will increase click-through.
- Control: "Start Trial"
- Variant: "Start My Free 14-Day Trial"
- Metric: CTA Click-through Rate (Primary), Trial Activation Rate (Guardrail).
Next steps
- Update the CSS color tokens for button contrast to meet AA standards.
- Synchronize the H1 headline with the LinkedIn campaign copy to reduce immediate bounce.
- Remove the mandatory phone field from the
auth-formcomponent to reduce friction.
Connects securely to your tools. The creator never sees your data.
About this skill
The problem
Vague "best practice" audits often ignore real evidence, leading to wasted engineering effort on low-impact tweaks. Developers and growth teams struggle to separate qualitative guesses from observed user friction and technical blockers.
What it does
- Analyzes URLs, screenshots, and HTML/CSS to map the visitor decision path from entry to conversion.
- Classifies every claim as OBSERVED, INFERRED, HYPOTHESIS, or CALCULATED to maintain rigorous evidence standards.
- Identifies technical friction in CTAs, form hierarchy, mobile responsiveness, and accessibility barriers.
- Generates structured implementation tickets with clear acceptance criteria and instrumentation requirements.
- Builds an experiment backlog with defined control/variant mechanics and primary/guardrail metrics.
Frameworks & tools
Compatible with any web stack including React, Next.js, Webflow, and Shopify. Supports analysis of analytics exports (GA4/Mixpanel) and structured HTML/CSS evidence.
Why this beats prompting it yourself
Generic prompts result in shallow feedback like "make the button bigger." This skill enforces a strict evidence graph, preventing the AI from fabricating metrics or promising fake conversion lifts. It forces a distinction between visual observations and code-level technical verification.
Use cases
- Pre-launch technical and conversion audit for new landing pages or funnels.
- Translating qualitative user session recordings into actionable Jira or Linear tickets.
- Validating message match between ad copy and destination page architecture.
- Auditing mobile-specific layout shifts and tap-target accessibility risks.
Known limitations
Cannot prove DOM semantics or screen-reader behavior from screenshots alone. Performance claims are limited to visible risk signals unless raw Core Web Vitals data is provided.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Research & Analysis.
Cannot verify screen-reader output or keyboard focus order from visual evidence alone. Requires raw data for precise performance metrics. Scope is limited to the provided evidence assets.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before buying a skill.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
4 uses
Times used to date
Free to use
No account required to browse
Trust & safety
Security scan
Security scan passed 2 months ago
- Free with an Agensi account
Permissions
No special permissions declared or detected