Overview
This is a skill for UX design analysis and generating recommendations. It uses a Fan-out/Fan-in pattern to analyze three dimensions (heuristic evaluation, accessibility review, user flow analysis) in parallel, then integrates the results to present a prioritized improvement roadmap.
It analyzes a product's user experience (UX) from multiple dimensions and provides actionable improvement recommendations.
Trigger Keywords
UX design, user experience, UX analysis, heuristic evaluation, heuristic evaluation, accessibility, accessibility, usability, user flow, wireframe, UX research, UI/UX review, design improvement
Workflow
Step 1: Product Context Analysis
Check the product type (Web App, Mobile App, Desktop Software, Physical Product), target users (age, technical level, disability status, primary language), design goals (improve conversion rate, reduce churn rate, improve satisfaction, onboard new users), and existing screens (Wireframe, Mockup, actual screen captures).
Step 2: Fan-out - Parallel Analysis
Three dimensions are analyzed independently in parallel:
Dimension 1: Heuristic Evaluation Evaluate based on Nielsen's 10 Usability Heuristics: visibility of system status, match between system and the real world, user control and freedom, consistency and standards, error prevention, recognition rather than recall, flexibility and efficiency of use, aesthetic and minimalist design, help users recognize, diagnose, and recover from errors, help and documentation
Dimension 2: Accessibility Review Evaluate based on WCAG 2.1 (Level AA) criteria: Perceivable (text alternatives, captions, distinguishable, resize text), Operable (keyboard accessible, enough time, seizures, navigable), Understandable (readable, predictable, input assistance), Robust (compatible, markup)
Dimension 3: User Flow Analysis Analyze the user's journey (User Journey): entry paths, core paths, identifying barriers (Friction Points), exception paths (Edge Cases), exit paths (Exit Points)
Step 3: Fan-in - Result Integration
Perform correlation analysis (identify duplicate issues and interdependencies between dimensions), severity assessment (Critical, Serious, Minor, Cosmetic), impact evaluation (business impact), and priority determination (severity × impact × resolution cost).
Step 4: Prioritization
Determine priorities using the RICE framework or MoSCoW method:
- RICE Score = (Reach × Impact × Confidence) / Effort
- MoSCoW: Must Have, Should Have, Could Have, Won't Have
Step 5: Improvement Roadmap Proposal
Create an actionable roadmap by priority: short-term (1-2 weeks), mid-term (1-3 months), long-term (3-6 months)
Usage Examples
Example 1: Improving the UX of an e-commerce app payment flow
- Product: Mobile e-commerce app (Android/iOS)
- Target users: ages 20-40, mobile native
- Design goal: reduce payment abandonment rate (currently 68%)
- Existing screens: Cart → Shipping info → Payment method → Payment complete (4 steps)
Example 2: SaaS dashboard onboarding UX improvement
- Product: B2B SaaS dashboard (web app)
- Target users: non-technical marketers, ages 30-50
- Design goal: improve new user first-week achievement rate from 30% → 60%
- Existing screens: Sign-up → Email verification → Dashboard entry (empty state)
Output format
The UX analysis report includes the following sections:
- Product context
- Fan-out: analysis results by dimension (heuristic evaluation, accessibility review, user flow analysis)
- Fan-in: integrated analysis (correlation analysis, prioritization)
- Integrated UX recommendations (P0/P1/P2/P3)
- Improvement roadmap (short-term/mid-term/long-term)
Precautions
Precautions for UX analysis
- User-centered: define problems from the user's perspective (excluding developer/business perspectives)
- Evidence-based: claims must be supported by data or user testing
- Actionability: present actionable improvements rather than ideal solutions
- Prioritization: do not try to solve everything at once; approach by order of impact
- Accessibility by default: consider accessibility as a baseline for all UX improvements
Avoid common mistakes
- Focusing only on aesthetic design: UX is not aesthetic design but usability and experience
- Waterfall approach: experiment quickly rather than following analysis → design → implementation
- Skipping user testing: analysis alone is not enough; validate with actual user testing
- Generic recommendations: avoid vague suggestions like "UX improvement"
- Treating accessibility as secondary: accessibility is not an add-on feature but a basic requirement
Related skills
| Skill | When to use |
|---|---|
moai-product:ux-researcher |
Prerequisite: when user research is needed |
moai-product:spec-writer |
Follow-up: when a UX specification document is needed |
moai-content:landing-page |
Alternative: when landing page UX improvement is needed |
Post-processing chain
The UX analysis report's narrative recommendation section(Integrated UX recommendations P0-P3 descriptions, improvement roadmap narrative) only are finalized through the following chain. Tabular and score data such as severity, RICE, and priority matrix are not subject to refinement:
moai-product:ux-designer → moai-core:ai-slop-reviewer → moai-content:humanize-korean
moai-core:ai-slop-reviewer: check and polish AI-typical clichés, exaggeration, and redundancy.moai-content:humanize-korean: final polish into natural Korean sentences.