Claude Code Plugins

Community-maintained marketplace

Feedback

information-architecture

@tachyon-beep/skillpacks
1
0

Organize content and design navigation using mental models and discoverability

Install Skill

1Download skill
2Enable skills in Claude

Open claude.ai/settings/capabilities and find the "Skills" section

3Upload to Claude

Click "Upload skill" and select the downloaded ZIP file

Note: Please verify skill by going through its instructions before using it.

SKILL.md

name information-architecture
description Organize content and design navigation using mental models and discoverability

Information Architecture

Overview

This skill provides a systematic approach to organizing content and designing navigation systems. Use when users can't find features, navigation feels confusing, or content needs structural organization.

Core Principle: Good IA matches users' mental models and provides multiple paths to content with clear information scent.

When to Use

Load this skill when:

  • Users complain "I can't find [feature]"
  • Navigation structure needs design or reorganization
  • Content organization is unclear or inconsistent
  • Designing site maps, menus, or navigation systems
  • User requests mention: "navigation", "menu", "structure", "organize", "findability"

Don't use for: Visual styling (use visual-design-foundations), interaction mechanics (use interaction-design-patterns)


The Navigation & Discoverability Model

A systematic 4-layer framework for evaluating and designing information architecture:

Layer 1: MENTAL MODELS - Does structure match how users think? Layer 2: NAVIGATION SYSTEMS - Are there multiple paths to content? Layer 3: INFORMATION SCENT - Do labels lead users to their goal? Layer 4: DISCOVERABILITY - Can users find features beyond the basics?

Evaluate each layer systematically to identify IA gaps.


Layer 1: Mental Models

Purpose: Align information structure with users' existing mental models

Evaluation Questions

  • Does the IA match how users think about this domain?
  • Are we using terminology users understand?
  • Would users expect to find X under category Y?
  • Does the structure match users' tasks and goals?
  • Are we organizing by internal company structure (BAD) or user needs (GOOD)?

Research Methods

Card Sorting (Open)

  • Give users cards with content/features
  • Ask them to group cards into categories they create
  • Reveals natural mental model organization
  • Use for: New IA, understanding user categories

Card Sorting (Closed)

  • Give users cards and predefined categories
  • Ask them to place cards in categories
  • Validates existing IA structure
  • Use for: Testing proposed navigation

Tree Testing

  • Give users IA tree (text-only, no visual design)
  • Ask them to find specific items
  • Measures findability without visual interference
  • Use for: Validating IA before visual design

Mental Model Diagrams

  • Map user's understanding of domain (Indi Young method)
  • Compare to organization's mental model
  • Identify gaps between user expectations and implementation
  • Use for: Deep understanding, strategy

Patterns

Task-Based Organization

  • Structure around what users do, not what you sell
  • Example: "Send Money" not "Products > Wire Transfer"
  • Users think in goals, not features

Domain-Appropriate Terminology

  • Use industry terms users know (not internal jargon)
  • Example: "Shopping Cart" not "Basket" (e-commerce convention)
  • Test terminology with card sorting

Flat-ish Hierarchy

  • Aim for 3-5 levels deep maximum
  • Broad (5-9 top-level categories) > Deep (3 categories, 10 levels)
  • Users struggle with navigation deeper than 3-4 clicks

Mutually Exclusive Categories

  • Each item belongs in one obvious place
  • Avoid: "Accessories" and "Cables" (cables are accessories)
  • Reduces confusion, speeds navigation

Anti-Patterns

Organized by Internal Structure

  • "About Us > Divisions > Division 3 > Services"
  • Users don't care about org chart
  • Organize by user needs, not company hierarchy

Jargon Users Don't Understand

  • Internal product names ("Project Phoenix")
  • Acronyms without context ("SDLC Dashboard")
  • Test labels with users unfamiliar with domain

Categories That Overlap

  • Item could reasonably go in 2+ categories
  • Users waste time checking multiple places
  • Forces arbitrary decisions

Too Many Top-Level Categories

  • 15+ menu items = overwhelming
  • Users can't hold that many options in memory
  • Chunk into 5-9 logical groups

Practical Application

Step 1: Conduct card sorting with 15-20 users

  • Open card sort to discover categories
  • Closed card sort to validate proposed structure

Step 2: Analyze sorting patterns

  • What categories did users create?
  • What labels did they use?
  • Where was agreement? Disagreement?

Step 3: Map to mental model

  • Create IA structure matching user categories
  • Use user terminology for labels
  • Keep hierarchy flat (3-4 levels max)

Step 4: Validate with tree testing

  • Test findability before implementing
  • Iterate based on task success rates

Layer 2: Navigation Systems

Purpose: Provide multiple paths to content so users can navigate their preferred way

Evaluation Questions

  • Can users reach content via global navigation?
  • Is search available and functional?
  • Are contextual links provided within content?
  • Can users get back to where they were (breadcrumbs)?
  • Are there shortcuts for common tasks?
  • Is navigation consistent across screens?

Navigation Types

Global Navigation

  • Persistent across entire site/app (top bar, sidebar)
  • Contains primary categories (5-7 items)
  • Always visible or easily accessible
  • Example: Main menu, app navigation drawer

Local Navigation

  • Specific to current section (sub-menu)
  • Shows related pages within category
  • Example: Sidebar in documentation, tabs within settings

Contextual Navigation

  • Links embedded in content
  • "See also", "Related articles", inline links
  • Helps users discover related content

Search

  • Direct path to specific content
  • Essential when >50 pages/items
  • Requires: autocomplete, filters, clear results

Utility Navigation

  • Secondary functions (login, settings, help)
  • Usually top-right corner or under menu
  • Less prominent than global nav

Breadcrumbs

  • Show path from home to current page
  • Allow users to backtrack easily
  • Example: Home > Products > Electronics > Laptops

Patterns

3-Click Rule (Modified)

  • Any content reachable within 3 clicks
  • More accurately: 3 CLEAR clicks (with good information scent)
  • Better to have 5 obvious clicks than 2 confusing ones

Consistent Navigation Placement

  • Same location across all pages
  • Same labels, same order
  • Users learn once, applies everywhere

Primary + Secondary Actions

  • One primary path (main menu, prominent)
  • Secondary paths (search, contextual links)
  • Redundancy helps different user styles

Mobile Navigation Patterns

  • Bottom tab bar (iOS standard, 3-5 items)
  • Hamburger menu (for 6+ items)
  • Bottom sheet menus (Android)
  • Prioritize: Most important actions at bottom (thumb zone)

Web Navigation Patterns

  • Horizontal top nav (desktop)
  • Sticky header (nav always accessible)
  • Mega-menus (for complex sites, show 2+ levels)
  • Sidebar nav (for deep hierarchies)

Desktop Navigation Patterns

  • Menu bar (File, Edit, View, Help)
  • Toolbar (icon buttons for common actions)
  • Keyboard shortcuts (Cmd+N, Cmd+O, Cmd+S)

Anti-Patterns

Single Path to Content

  • Only one way to reach feature (menu > submenu > page)
  • Users with different mental models get lost
  • Provide: Global nav + search + contextual links

Mega-Menu Overload

  • Showing 50+ links in dropdown
  • Overwhelming, hard to scan
  • Chunk into logical groups, use hierarchy

Mystery Meat Navigation

  • Icons without labels
  • Users guess what icon means
  • Use: Icon + label (or tooltip on hover)

Inconsistent Navigation

  • Menu moves or changes on different pages
  • Different labels for same destination
  • Forces users to relearn on each page

Buried Search

  • Search hidden in menu or obscure icon
  • Essential for large sites (>50 pages)
  • Make prominent: Top-right is convention

No Way Back

  • Dead-end pages with no navigation
  • Modal with no close/back button
  • Always provide: Back, breadcrumbs, or close

Practical Application

Step 1: Audit all navigation paths

  • List all ways to reach each major page/feature
  • Identify single-path content (problem)

Step 2: Add redundant paths

  • Global nav for primary content
  • Search for direct access
  • Contextual links within content
  • Breadcrumbs for backtracking

Step 3: Test with first-click tests

  • Show users page/screen
  • Ask "Where would you click to [task]?"
  • Measure if first click is correct path

Step 4: Verify consistency

  • Same navigation on all pages
  • Same labels across contexts
  • Test on 10+ different pages

Layer 3: Information Scent

Purpose: Labels and clues accurately predict what users will find

Evaluation Questions

  • Do link labels accurately describe destination?
  • Are category names specific (not vague)?
  • Would users know where to look for their task?
  • Do labels use user terminology?
  • Are categories mutually exclusive?
  • Does "scent" get stronger as user gets closer (progressive refinement)?

Core Concept: Scent Strength

Strong Scent

  • Label clearly predicts content
  • Example: "Download Receipt" → PDF receipt downloads
  • Users confident they're on right path

Weak Scent

  • Label vague, ambiguous
  • Example: "Resources" → Could be anything
  • Users unsure if this is right path

False Scent

  • Label misleads users
  • Example: "Settings" → Only shows account info, not all settings
  • Users frustrated, waste time

Patterns

Specific, Action-Oriented Labels

  • Good: "Download Invoice", "Change Password", "View Order History"
  • Bad: "Account", "Options", "More"
  • Users scan for action verbs matching their goal

Front-Loading Keywords

  • Good: "Projects > Project Alpha > Documents"
  • Bad: "All > Alpha > Files and Documents for Projects"
  • Users scan left-to-right, see important words first

Descriptive Category Names

  • Good: "User Settings", "Billing Information", "Security Preferences"
  • Bad: "Preferences", "Options", "More"
  • Specific > generic

Progressive Refinement

  • Each level narrows scope
  • Example: Electronics > Laptops > Gaming Laptops > 17-inch Gaming Laptops
  • Scent gets stronger at each level

Preview/Descriptions for Ambiguous Items

  • When label can't be specific, add description
  • Example: "Advanced" (label) + "For power users: API keys, webhooks, custom scripts" (description)
  • Clarifies before user clicks

Avoid Overlapping Categories

  • Bad: "Accessories" + "Cables" (cables are accessories)
  • Users check both, waste time
  • Make mutually exclusive or nest hierarchically

Anti-Patterns

Vague Labels

  • "Stuff", "Other", "More", "Miscellaneous"
  • Tells users nothing about content
  • If necessary, add description

Marketing Speak

  • "Solutions", "Innovative Tools", "Next-Gen Platform"
  • Sounds impressive, means nothing
  • Use plain language users understand

Icon-Only Labels (No Text)

  • Users guess what icon means
  • Different cultures interpret icons differently
  • Use icon + text (or tooltip)

Jargon Users Don't Know

  • Internal product names ("Project Phoenix Dashboard")
  • Industry acronyms ("SDLC", "KPIs") without context
  • Test with users unfamiliar with domain

Same Label for Different Things

  • "Settings" in menu → All settings
  • "Settings" in modal → Only this feature's settings
  • Creates confusion, breaks trust

Label Doesn't Match Content

  • Link says "View Profile", page shows "Account Settings"
  • Breaks trust, wastes time
  • Ensure label = page title/content

Practical Application

Step 1: Audit all labels

  • List every menu item, button, link
  • Ask: "Does this label clearly predict destination?"

Step 2: Test with users (5-second test)

  • Show label for 5 seconds
  • Ask: "What do you expect to find here?"
  • Compare to actual content

Step 3: Refine labels

  • Replace vague with specific
  • Front-load keywords
  • Use user terminology (from card sorting)

Step 4: Validate with tree testing

  • Users navigate text-only IA tree
  • Measure task success rate
  • Iterate on failing labels

Layer 4: Discoverability

Purpose: Users can find features beyond the basics, progressively

Evaluation Questions

  • Can users discover advanced features?
  • Is there progressive disclosure (simple first, advanced later)?
  • Are there tooltips/contextual help?
  • Do features have logical placement (near related functionality)?
  • Is there onboarding for core features?
  • Can users learn without documentation?

Patterns

Progressive Disclosure

  • Show basics by default
  • Hide advanced features behind "Advanced" button/section
  • Example: Simple form with 3 fields + "More options" reveals 10 more
  • Prevents overwhelming new users, empowers advanced users

Onboarding for Core Features

  • First-time users see guided tour
  • Highlights 3-5 essential features
  • Optional skip (don't force)
  • Example: "Here's how to create your first project"

Contextual Help

  • "?" icon next to complex fields
  • Tooltips on hover (desktop) or tap (mobile)
  • Inline explanations (not separate docs)
  • Example: "What's this?" next to "API Key"

Logical Feature Placement

  • Related features grouped together
  • Example: "Export as PDF" near "Print"
  • Users discover by exploring related functionality

Empty States with Guidance

  • When no content exists, show how to create content
  • Example: "No projects yet. Create your first project."
  • Provides clear next action

Search for Hidden Features

  • Power users use search to find advanced features
  • Ensure all features indexed by search
  • Example: Search for "API" finds API settings

"Learn More" Links

  • Link to docs/help for complex features
  • Non-intrusive (users opt-in)
  • Example: "Learn about webhooks" link in webhook settings

Feature Announcements

  • Toast/modal when new feature ships
  • "Dismiss" option (one-time)
  • Example: "New: Dark mode in settings"

Anti-Patterns

Power Features Buried 5+ Levels Deep

  • Users never discover them
  • Balance: Progressive disclosure vs discoverability
  • Solution: Provide search, contextual hints

No Onboarding (Users Never Discover Features)

  • Blank screen with no guidance
  • Users churn because they don't understand value
  • Solution: Quick tour highlighting core features

Everything Exposed at Once (Overwhelming)

  • 50 buttons on screen
  • Analysis paralysis, cognitive overload
  • Solution: Progressive disclosure, hide advanced

No Contextual Help

  • Complex features with no explanation
  • Users trial-and-error or leave
  • Solution: Tooltips, inline help, "?" icons

Inconsistent Placement

  • "Settings" in different places on different screens
  • Users can't predict where to look
  • Solution: Consistent placement (top-right for settings is convention)

Dead-End Experiences

  • Users reach feature but don't know what to do
  • No placeholder text, no examples, no guidance
  • Solution: Empty states with next steps

Features Without Labels

  • Icon-only buttons (users guess)
  • Hover-only discoverability (fails on mobile)
  • Solution: Icons + labels or clear tooltips

Practical Application

Step 1: Identify core vs advanced features

  • Core: 80% of users need these
  • Advanced: 20% of users, power features

Step 2: Design progressive disclosure

  • Core features: Visible by default
  • Advanced features: Behind "Advanced" or "More"
  • Ensure advanced still discoverable (search, contextual hints)

Step 3: Add contextual help

  • Complex fields get tooltips
  • First-time experience gets onboarding
  • Empty states explain next actions

Step 4: Test with new users

  • Give users tasks (don't tell them how)
  • Observe: Do they discover features?
  • Iterate: Add hints where users get stuck

User Journey Integration

Information architecture must support complete user journeys:

Journey Mapping for IA

Step 1: Map Primary User Journeys

  • List top 5-10 tasks users do
  • Example: "Find product", "Check order status", "Contact support"

Step 2: Identify Touchpoints

  • Where does user enter? (Google, home page, deep link)
  • What pages do they visit? (in what order)
  • Where do they exit? (goal complete or frustrated)

Step 3: Evaluate IA Against Journeys

  • Can users complete journey with current IA?
  • Where do they get stuck? (missing scent, dead ends)
  • Are there unnecessary steps? (simplify)

Step 4: Optimize IA for Common Journeys

  • Put common endpoints in global nav
  • Add shortcuts for frequent paths
  • Remove obstacles (reduce clicks, clarify labels)

Example: E-Commerce Journey

Journey: Find and purchase product

Touchpoints:

  1. Enter via Google search → Product page (deep link)
  2. View product details
  3. Add to cart
  4. View cart
  5. Checkout
  6. Confirmation

IA Optimizations:

  • Breadcrumbs on product page (Layer 2: Navigation)
  • Related products (Layer 2: Contextual nav)
  • Clear "Add to Cart" button (Layer 3: Information scent)
  • Persistent cart icon (Layer 4: Discoverability)
  • Guest checkout option (Layer 1: Mental model - not everyone wants account)

Cross-Platform Navigation Considerations

Navigation patterns adapt to platform context:

Mobile Navigation

Constraints:

  • Small screen (375px width typical)
  • Thumb-based interaction
  • One screen at a time (no multi-window)

Patterns:

  • Bottom tab bar (3-5 items, iOS standard)
  • Hamburger menu (for 6+ items)
  • Bottom sheet menus (Android)
  • Swipe gestures (swipe back, swipe between tabs)

Prioritization:

  • Most important actions at bottom (thumb zone)
  • Secondary actions in hamburger menu
  • Search prominent (magnifying glass icon)

Reference: See mobile-design-patterns for platform specifics

Web Application Navigation

Affordances:

  • Large screen (1024px+ width)
  • Keyboard + mouse
  • Multi-tab workflows

Patterns:

  • Horizontal top nav (persistent)
  • Sidebar nav (for deep hierarchies)
  • Breadcrumbs (show location in hierarchy)
  • Mega-menus (reveal 2+ levels on hover)
  • Keyboard shortcuts (Cmd+K command palette)

Considerations:

  • Responsive: Collapses to mobile patterns on narrow screens
  • Dense information: Can show more hierarchy on desktop

Reference: See web-application-design for responsive patterns

Desktop Software Navigation

Affordances:

  • Multi-window support
  • Extensive keyboard shortcuts
  • Menu bars

Patterns:

  • Menu bar (File, Edit, View, Help)
  • Toolbars (icon buttons, customizable)
  • Panels/palettes (floating or docked)
  • Context menus (right-click)

Considerations:

  • Keyboard-first: Every menu item has shortcut
  • Customization: Users rearrange toolbars/panels

Reference: See desktop-software-design for window management

Game UI Navigation

Constraints:

  • Gamepad/controller input (limited buttons)
  • Immersion (minimize UI)
  • Performance (UI can't tank frame rate)

Patterns:

  • Radial menus (analog stick selection)
  • D-pad for menus (up/down/left/right)
  • Pause menu (Start button)
  • Context-sensitive prompts ("Press X to interact")

Considerations:

  • Minimize navigation: Players want to play, not navigate menus
  • Contextual: Show only relevant options for current situation

Reference: See game-ui-design for immersion vs visibility tradeoffs


Navigation Pattern Library

Common navigation solutions with rationale:

Pattern: Bottom Tab Bar (Mobile)

Use When:

  • Mobile app with 3-5 primary sections
  • Users frequently switch between sections
  • Following iOS Human Interface Guidelines

Structure:

  • 3-5 tabs at bottom
  • Always visible
  • Icon + label (or icon only if space constrained)
  • Active tab highlighted

Example: Instagram (Home, Search, Create, Reels, Profile)


Pattern: Hamburger Menu (Mobile)

Use When:

  • 6+ navigation items (too many for tabs)
  • Secondary functions (not primary navigation)
  • Android or web mobile

Structure:

  • ☰ icon (top-left or top-right)
  • Slides out drawer with nav items
  • Overlay or push content

Warning: Reduces discoverability ("out of sight, out of mind")


Pattern: Breadcrumbs (Web/Desktop)

Use When:

  • Deep hierarchies (3+ levels)
  • Users need to understand location
  • E-commerce, documentation, file systems

Structure:

  • Home > Category > Subcategory > Current Page
  • Each level clickable (except current)
  • Shows path from root to current

Example: Amazon product pages


Pattern: Mega-Menu (Web)

Use When:

  • Complex site with many categories
  • Need to show 2-3 levels of hierarchy at once
  • Desktop web (requires hover)

Structure:

  • Hover top nav item
  • Reveals large dropdown with subcategories
  • Grid or column layout

Warning: Doesn't work on mobile (no hover), requires desktop


Pattern: Sidebar Navigation (Web/Desktop)

Use When:

  • Deep hierarchies (documentation, admin panels)
  • Always need to show navigation context
  • Desktop-first or responsive web

Structure:

  • Left sidebar (persistent or collapsible)
  • Hierarchical tree (expand/collapse sections)
  • Active page highlighted

Example: Documentation sites, admin dashboards


Pattern: Command Palette (Web/Desktop)

Use When:

  • Power users need fast access to all commands
  • Many features (50+ actions)
  • Keyboard-first workflows

Structure:

  • Keyboard shortcut (Cmd+K common)
  • Fuzzy search all actions
  • Jump to any page/feature instantly

Example: VS Code, Slack, Figma


Red Flags & Anti-Patterns

Red Flag: Users Can't Find Primary Features

Symptom: Support requests "Where is [feature]?"

Root Cause: Poor information scent or buried navigation

Fix:

  • Audit Layer 3 (Information Scent): Are labels clear?
  • Add to global navigation (Layer 2)
  • Provide search (Layer 2)

Red Flag: Navigation Inconsistent Across Screens

Symptom: Menu changes or moves on different pages

Root Cause: Lack of navigation system standards

Fix:

  • Define persistent global navigation
  • Same placement, same labels everywhere
  • Test on 10+ different screens

Red Flag: One Way to Reach Content

Symptom: Users lost if they don't know exact path

Root Cause: Single navigation path (Layer 2 failure)

Fix:

  • Add redundant paths (global nav + search + contextual links)
  • Provide breadcrumbs for backtracking
  • Test with users who enter via deep links

Red Flag: Users Complete Task But Say "That Was Confusing"

Symptom: Low satisfaction despite task completion

Root Cause: Weak information scent (Layer 3) - users succeeded through trial-and-error

Fix:

  • Improve label specificity
  • Add descriptions to ambiguous categories
  • Test with 5-second test (label → expected content)

Red Flag: Power Users Can't Find Advanced Features

Symptom: "Does this have [feature]?" (feature exists but hidden)

Root Cause: Poor discoverability (Layer 4)

Fix:

  • Add search (all features indexed)
  • Contextual hints ("Looking for advanced options? Click here")
  • Feature announcements when new features ship

Red Flag: New Users Overwhelmed, Churn Quickly

Symptom: High bounce rate, low activation

Root Cause: No progressive disclosure, everything shown at once

Fix:

  • Hide advanced features behind "Advanced" section
  • Add onboarding (highlight core features)
  • Empty states with guidance

Red Flag: Users Click Wrong Category, Backtrack Repeatedly

Symptom: Analytics show high back-button usage, category exploration

Root Cause: Labels don't match user mental model (Layer 1) or weak scent (Layer 3)

Fix:

  • Card sorting to understand user categories
  • Relabel based on user terminology
  • Tree testing to validate

Related Skills

Cross-references to other Lyra skills:

  • lyra/ux-designer/user-research-and-validation - Card sorting, tree testing, mental model research methods for IA validation
  • lyra/ux-designer/interaction-design-patterns - Navigation interactions (menus, dropdowns, tabs) and micro-interactions for nav elements
  • lyra/ux-designer/visual-design-foundations - Visual hierarchy to emphasize navigation, typography for readable labels
  • lyra/ux-designer/accessibility-and-inclusive-design - Keyboard navigation, screen reader compatibility for nav systems
  • lyra/ux-designer/ux-fundamentals - Core IA principles and terminology

Platform-specific IA:

  • lyra/ux-designer/mobile-design-patterns - Mobile navigation patterns (bottom tabs, hamburger menus, gestures)
  • lyra/ux-designer/web-application-design - Complex web nav systems (mega-menus, breadcrumbs, responsive patterns)
  • lyra/ux-designer/desktop-software-design - Desktop navigation (menu bars, toolbars, keyboard shortcuts)
  • lyra/ux-designer/game-ui-design - Game navigation (radial menus, contextual prompts, minimal UI)

Cross-faction references:

  • muna/technical-writer/documentation-structure - IA for documentation sites (organizing docs for findability)
  • muna/technical-writer/clarity-and-style - Writing clear labels and navigation copy