---

name: design-information-architecture description: > Use when the user asks about information architecture, sitemaps, user flows, content hierarchy, navigation design, content organization, labeling, or taxonomy. Do NOT use for: UX research (design-ux-research), visual design (design-visual-design), or prototyping (design-prototyping). version: "1.0.0" author: "j4flmao" license: "MIT" compatibility: claude-code: true cursor: true codex: true windsurf: true tags: [design, information-architecture, phase-3]

Information Architecture

Purpose

Design information architecture: organize content, create navigation systems, define taxonomies, design user flows, and ensure findability across digital products. The goal is to help users find information and complete tasks with minimum cognitive load by creating intuitive, user-centered content structures.

Agent Protocol

Trigger

Exact user phrases: "information architecture" "sitemap" "user flow" "content hierarchy" "navigation" "content organization" "labeling" "taxonomy" "card sorting" "tree testing" "findability" "IA audit".

Input Context

• What type of product or website is being designed?
• Who are the primary user segments and their information needs?
• What content exists and what content is planned?
• What are the key user tasks and goals?
• What are the business objectives (conversion, engagement, retention)?
• What is the current IA (if redesigning) and its known issues?
• What competitive or reference IAs exist?

Output Artifact

Information architecture specification with content audit, sitemap, navigation design, taxonomy, and user flows.

Response Format

## Information Architecture Plan
### Content Audit Summary
Total pages: {N} | Gaps: {list} | Duplicates: {list}

### Sitemap Structure
Level 1: {category} → Level 2: {subcategory} → Level 3: {page}

### Navigation System
Primary: {top / side / hub-and-spoke}
Secondary: {utility / breadcrumb / footer}
Contextual: {cross-links / related content / tags}

### Taxonomy
Type: {flat / hierarchical / faceted / tags}
Categories: {list}
Labels: {list}

No preamble. No postamble. No explanations.

Completion Criteria

• Content audit completed with inventory and gap analysis
• User research conducted to understand mental models
• Card sorting conducted (open or closed) with results analyzed
• Tree testing conducted with findability rate measured
• Sitemap created showing full content hierarchy
• Navigation system designed (primary, secondary, utility)
• Taxonomy defined with labels and categorization rules
• User flows documented for key tasks
• IA validated with usability testing
• Responsive/adaptive navigation behavior defined

Max Response Length

150 lines of spec and structure.

Framework/Methodology

IA Design Framework

The IA design process follows the Four Circles model:

Content Context
  ↓
User Needs & Behaviors  →  IA Design  →  Navigation & Labeling
  ↑
Business Context & Constraints

The intersection of these three circles (content, users, context) defines the optimal IA.

Findability Principles

Principle	Definition	Application
Predictability	Users can guess where content lives	Clear labels, consistent patterns
Findability	Content can be located by browsing or search	Multiple access paths to same content
Understandability	Content organization makes sense to users	Mental model alignment
Consistency	Similar content is organized similarly	Same patterns across sections
Accessibility	Content is available to all users regardless of ability	Semantic structure, keyboard nav

Cognitive Load in IA

IA Factor	Cognitive Load Impact	Mitigation
Navigation depth (>3 levels)	Users lose orientation	Breadcrumbs, visual hierarchy
Navigation breadth (>7 items per level)	Choice overload	Prioritize top items, group secondary
Unclear labels	Users waste time clicking	User-tested labeling, familiar terms
Multiple competing navigation systems	Confusion	Clear primary navigation, distinct secondary
Inconsistent organization across sections	Mental effort to re-learn	Apply same IA principles site-wide

Information Architecture Maturity

Level	Name	Characteristics
1	Organic	Content organized by team structure, no intentional IA
2	Hierarchical	Basic category tree, top-down organization
3	User-centered	IA informed by user research (card sorting, tree testing)
4	Adaptive	IA adapts to user behavior and context
5	Predictive	Anticipates user needs, proactive content delivery

Workflow

Step 1: Content Audit

Inventory all content, identify gaps and duplicates. Categorize content by type, owner, format, and status. Quantify volume to understand scale of IA challenge.

Content audit template:

#	URL/Path	Title	Content Type	Format	Owner	Status	Notes
1	/products/widget	Widget Pro	Product page	HTML	Product team	Published	Needs IA review
2	/help/faq	FAQ	Support	HTML	Support team	Published	Multiple duplicates

Content audit analysis outputs:

• Total content volume (pages, documents, media files)
• Content type distribution (product, support, marketing, legal)
• Orphaned content (no parent, no navigation path)
• Duplicate content (same information in multiple locations)
• Content gaps (missing content for known user needs)
• Stale content (not updated in >12 months)

Step 2: User Research for IA

Understand mental models and findability patterns. Methods specific to IA research:

Method	Purpose	Participants	Timeline	Output
Card sorting (open)	Understand how users naturally group content	20-50	1 week	Category clusters, label suggestions
Card sorting (closed)	Validate proposed categories	20-50	1 week	Category assignment accuracy
Tree testing	Test findability in proposed navigation	20-50	1 week	Findability rate, directness score
First-click testing	Test if users click correct navigation item	20-50	3-5 days	First-click accuracy
Reverse card sorting	Test if users can place content in correct categories	20-50	1 week	Categorization accuracy
Content needs survey	Identify what content users need	50-200	1-2 weeks	Content priority ranking

Step 3: Card Sorting

Group content into logical categories based on user mental models.

Open card sorting process:

1. Prepare content cards (40-80 items for manageable sessions)
2. Use tools: OptimalSort, Miro, UserZoom, or physical cards
3. Instruct participants to group cards and label groups
4. Analyze using similarity matrix to identify clusters
5. Identify: high-agreement groups (strong categories), ambiguous cards (don't fit anywhere), labels that resonate

Closed card sorting process:

1. Provide predefined categories
2. Ask participants to assign each content item to a category
3. Calculate: % correct assignment, categories with confusion
4. Identify: categories that need renaming, content that needs recategorizing

Analysis outputs:

• Dendrogram: visual tree of content clusters
• Similarity matrix: % of participants who grouped each pair together
• Category labels: the most common labels suggested by participants
• Mismatches: content that consistently gets placed in unexpected categories

Step 4: Tree Testing

Validate navigation structure by testing findability of specific content items.

Tree testing process:

1. Build text-only tree from proposed navigation
2. Define test tasks: "Find information about X"
3. Participants navigate tree to find each item
4. Measure: success rate, directness (did they go straight there), time, path
5. Identify: items with low findability, mislabeled categories, wrong placement

Targets:

• Findability rate: >80% for critical content, >70% for secondary content
• Directness: >60% direct paths
• Time: users should find most items within 10-20 seconds

Iteration: After first round of tree testing, revise IA and re-test. Each round typically improves findability by 10-20%.

Step 5: Sitemap Creation

Visual hierarchy of pages and sections. Sitemap formats:

• Visual sitemap: Diagram showing page relationships (spreadsheet, diagram tool)
• XML sitemap: Search engine crawlable listing (developers)
• Content matrix: Table showing content by category and page

Sitemap conventions:

Level 1          Level 2         Level 3          Level 4
Home
  Products
    Widget Pro
      Features   (features list page)
      Pricing    (pricing page)
      FAQ        (product FAQ)
    Widget Lite
      Features
      Pricing
  Support
    Documentation
      Getting Started
      Advanced Guide
      API Reference
    Contact
      Email Support
      Live Chat
  About
    Team
    Careers
    Press

Step 6: Navigation Design

Primary, secondary, utility, breadcrumb, and contextual navigation systems.

Navigation design principles:

• Expose 80% of user tasks in top two levels of navigation
• Highlight current location (visual indicator, breadcrumb)
• Use progressive disclosure for deep hierarchies
• Provide consistent navigation across all pages
• Minimize number of clicks to reach any content (3-click rule is a guideline, not law)
• Offer multiple paths to same content when appropriate

Navigation patterns detailed:

Pattern	Implementation	Max Items	Best For
Top horizontal nav	Categories across top, dropdown for sub-items	5-7 top-level items	Marketing sites, broad content
Side navigation (expandable)	Expandable/collapsible tree in sidebar	Unlimited nested levels	SaaS applications, documentation
Hub-and-spoke	Central hub, each spoke is a self-contained flow	3-5 spokes	Task-focused workflows, checkout
Search-dominant	Search as primary access, browsing secondary	N/A (search handles depth)	Large content volumes, knowledge bases
Faceted navigation	Filters and attributes for content refinement	Unlimited filters	E-commerce, catalogs, databases
Mega menu	Large dropdown showing multiple levels at once	20-50 items visible	Large product catalogs, complex sites
Progressive disclosure	Show more options on interaction	Any number	Settings, account management

Step 7: Taxonomy Design

Define content classification system.

Taxonomy types:

• Flat taxonomy: Simple list of categories (best for <20 categories)

  • Pros: Simple to implement, easy to understand
  • Cons: Limited scalability, single perspective
  • Use when: Small content volume, single user type

• Hierarchical taxonomy: Parent-child category relationships

  • Pros: Intuitive, supports depth, well-understood
  • Cons: Rigid, can force unnatural groupings
  • Use when: Broad content scope, multiple levels of detail

• Faceted taxonomy: Multi-dimensional classification

  • Pros: Flexible, supports multiple user needs
  • Cons: Complex to implement, requires metadata
  • Use when: Large content volumes, diverse user needs

• Tags: Flexible, user-generated categorization

  • Pros: Adaptive, low-maintenance
  • Cons: Inconsistent, no hierarchy
  • Use when: User-generated content, blogs

Step 8: User Flows

Document key task paths through the IA. Each flow shows:

• Entry point (how user arrives)
• Decision points
• Navigation path through content
• Completion (success or failure)
• Alternative paths (search, direct navigation)

User flow template:

Task: {user goal}
Entry: {entry point}
Primary Path:
  Step 1: {action} → {destination}
  Step 2: {action} → {destination}
  Step 3: {action} → {destination}
Alternative Paths:
  Path B: {alternative route}
  Search: {search query} → {results} → {selection}
Failure: {what happens if user can't complete}

Common Pitfalls

Pitfall	Description	Prevention
Mirroring org chart	Content organized by company structure, not user needs	Research user mental models via card sorting
Overly deep hierarchy	Content buried under too many levels	Keep maximum 3-4 levels; use mega menus for breadth
Ambiguous labels	Labels that don't clearly describe content	Test labels in tree testing; use familiar terms
One-size-fits-all IA	Same structure for different user segments	Consider persona-specific navigation or personalization
Ignoring search behavior	Designing for browsing when users search	Include search prominently; use search analytics to inform IA
Inconsistent labeling	Different terms for same concept across the site	Create label taxonomy with approved terms and variations
Content migration without IA review	Moving content to new design without restructuring	Always conduct content audit and IA review during redesign
Too many navigation systems	Users overwhelmed by choice	Distinguish primary, secondary, and contextual navigation
Skipping mobile IA	Desktop IA that doesn't adapt to mobile	Design mobile-first IA with progressive disclosure
No IA governance	IA degrades over time without ownership	Assign IA owner; establish content publishing guidelines

Best Practices

Practice	Rationale
Test IA with real users before building	IA changes after build are expensive; validate early
Use user language for labels	Industry jargon confuses users; use terms they use
Provide multiple access paths	Users arrive differently; offer browse, search, and related links
Keep navigation consistent across pages	Users should not have to re-learn navigation on each page
Use breadcrumbs for deep hierarchies	Shows location and enables navigation up the hierarchy
Design for scanning, not reading	Users scan; use descriptive headings, clear categories, visual hierarchy
Prioritize content by user needs	Most-needed content should be easiest to find
Plan for content growth	IA should accommodate 2-3x content volume without restructuring
Use analytics to inform IA decisions	Search queries, page views, and navigation patterns reveal IA issues
Establish IA governance	Assign ownership, create content standards, review IA annually

Templates & Tools

Content Audit Template (Spreadsheet)

URL | Page Title | Content Type | Category | Format | Owner | Last Updated | Status | Notes
/help/getting-started | Getting Started Guide | Help | Onboarding | Article | Docs Team | 2025-06-01 | Published |
/products/pro/features | Pro Features | Product | Products | HTML | Product | 2025-05-15 | Needs Update |

Card Sorting Results Analysis

Task: {sorting task description}
Participants: {N}
Method: {Open / Closed / Hybrid}

Top Category Clusters (by agreement %):
1. {Category Label} — {X%} agreement — {N items assigned}
2. {Category Label} — {X%} agreement — {N items assigned}

Ambiguous Items (no clear category):
- {Item}: equally distributed across {categories}
- {Item}: {X%} categorized differently from expected

Suggested Labels:
- {Original term} → {Suggested term} ({X%} of participants)
- {Original term} → {Suggested term} ({X%} of participants)

Tree Testing Report Template

Navigation Structure Tested: {version}
Participants: {N}
Tasks: {N}

Overall Findability Rate: {X%}
Average Directness: {X%}
Average Time: {X}s

Tasks with Findability <80%:
1. {Task} — {X%} findability — Issue: {common wrong path}
   Action: {IA change needed}

Tasks with Best Findability:
1. {Task} — {X%} findability — {path used}

IA Tools

Tool	Purpose	Pricing
OptimalSort	Card sorting (online, remote)	Paid/free tier
Treejack	Tree testing (online, remote)	Paid/free tier
Miro/Mural	Collaborative diagramming, card sorting	Freemium
UserZoom	Integrated IA research platform	Paid
LucidChart	Sitemap and flow diagramming	Freemium
FlowMapp	Sitemap creation, user flows	Paid/free tier
Dynalist	Content inventory and hierarchy	Freemium
Slickplan	Visual sitemap builder	Paid

Case Studies

Case Study 1: E-commerce IA Redesign Increases Revenue 18%

An online retailer with 50,000+ products organized them by manufacturer (the company's org structure). Card sorting with 35 shoppers revealed users thought in terms of use case (home office, gaming, professional) and price range, not brand. The redesigned IA used use-case categories as primary navigation with brand as a filter. Findability in tree testing improved from 55% to 85%, and revenue increased 18% after launch as users found products more easily.

Method: Open card sorting (35 participants) + tree testing (40 participants) Key insight: User mental model was use-case based, contrary to org chart structure Impact: Findability 55% to 85%, revenue +18%

Case Study 2: SaaS Documentation IA Reduces Support Tickets 30%

A B2B SaaS company had support documentation organized by feature name (matching internal engineering names). Tree testing showed only 40% findability for common tasks. After restructuring around user goals (getting started, troubleshooting, integrations, account management), findability improved to 82%. Support tickets for "how to" questions dropped 30% as users found self-serve answers.

Method: Tree testing (50 participants) before and after IA restructure Key insight: Feature-name organization meant nothing to users; goal-based organization was intuitive Impact: Support tickets -30%, documentation findability 40% to 82%

Case Study 3: Government Website IA with Universal Design

A government services website was organized by department (35+ departments, each with their own navigation). User research showed citizens don't know which department handles their issue. The redesign used a life-event IA (birth, marriage, moving, death, business) as primary, with department as secondary. Findability for common tasks increased from 35% to 78%, and task completion time decreased by 60%.

Method: Large-scale card sorting (200 participants across demographics) + tree testing Key insight: Life-event IA maps to citizen mental models better than government structure Impact: Findability 35% to 78%, task time -60%

Rules

• IA must be based on user research, not team structure assumptions.
• Labels must use user language, not internal terminology.
• Navigation depth should not exceed 4 levels without strong justification.
• Every content item must have a single canonical location in the IA.
• Critical content must be findable within 2 clicks of home page.
• Navigation must be consistent across all pages (same labels, same order).
• Breadcrumbs must be provided for hierarchies deeper than 2 levels.
• Search must be available on every page for content-heavy sites.
• IA must be validated with tree testing before implementation.
• 80% findability rate is the minimum acceptable threshold.
• Card sorting requires minimum 20 participants for statistically meaningful results.
• Content audit must be conducted before any IA redesign of existing products.
• Sitemap must show all pages, including utility and support pages.
• Mobile IA must be designed independently, not adapted from desktop.
• IA governance must be established with a designated owner and review cadence.

References

• references/content-modeling.md — Content Modeling Reference
• references/ia-methods.md — Information Architecture Methods Reference
• references/information-architecture-advanced.md — Information Architecture Advanced Topics
• references/information-architecture-fundamentals.md — Information Architecture Fundamentals
• references/navigation-design.md — Navigation Design Reference
• references/sitemap-taxonomy.md — Sitemaps and Taxonomy Reference
• references/ia-card-sorting-tree-testing.md — IA Card Sorting and Tree Testing
• references/ia-navigation-design.md — IA Navigation Design

Implementation Patterns

Observer Pattern for Event Handling

` interface EventObserver { onEvent(event: T): Promise; }

class EventBus { private observers: Set<EventObserver> = new Set(); subscribe(observer: EventObserver): void { this.observers.add(observer); } unsubscribe(observer: EventObserver): void { this.observers.delete(observer); } async emit(event: T): Promise { const results = Array.from(this.observers).map(o => o.onEvent(event)); await Promise.allSettled(results); } } `

Configuration-Driven Approach

config: defaults: timeout: 30s retryCount: 3 overrides: production: timeout: 60s retryCount: 5 development: timeout: 300s retryCount: 1

Production Considerations

Deployment Checklist

• Configuration validated against schema before startup
• Health check endpoints registered and monitored
• Graceful shutdown with draining period (30s timeout)
• Resource limits configured (CPU, memory, file descriptors)
• Log level set appropriate for environment
• Metrics endpoint secured and exposed
• Rate limiting configured per-tier
• TLS certificates valid and auto-renewing
• Database migrations run as separate deployment step
• Feature flags ready for gradual rollout

Monitoring and Alerting

Metric	Threshold	Severity	Action
Error rate	> 1% over 5min	Critical	Page on-call
p99 latency	> 2s over 5min	Warning	Investigate
Throughput drop	> 50% over 1min	Critical	Check upstream
Queue depth	> 1000 over 1min	Warning	Scale consumers
Disk usage	> 85%	Warning	Clean or expand
Memory usage	> 90% heap	Critical	Restart or scale

Anti-Patterns

Anti-Pattern	Symptom	Root Cause	Solution
Premature optimization	Complex code for no measured benefit	Guessing instead of profiling	Measure first, optimize based on data
Copy-paste reuse	Duplicate code across codebase	Lack of abstraction	Extract shared logic into libraries
Gold-plating	Features with no current requirement	Over-engineering	YAGNI — build what's needed now
Magical thinking	Assumptions without validation	Skipping error handling	Handle all failure modes explicitly

Performance Optimization

Caching Strategy

Cache hierarchy: L1 (in-memory local) → L2 (distributed Redis/Memcached) → L3 (CDN/Edge). Cache invalidation: TTL-based (simple, stale), event-based (complex, fresh), write-through (consistent, higher write latency), write-behind (fast writes, eventual consistency).

Resource Pooling

• Database connections: Pool of reusable connections (HikariCP, pgBouncer)
• HTTP connections: Keep-alive + connection pooling for external calls
• Thread pool: Bounded thread pools for async task execution

Profiling Methodology

1. Establish baseline with production traffic profile
2. Profile CPU with sampling profiler (pprof, perf, async-profiler)
3. Profile memory with heap dumps and allocation tracking
4. Profile I/O with strace/perf trace for syscall analysis
5. Profile latency with distributed tracing (OpenTelemetry)
6. Identify bottleneck, formulate hypothesis, implement fix
7. Re-profile to verify improvement, repeat

Security Considerations

Threat Modeling (STRIDE)

• Spoofing: Identity validation, authentication
• Tampering: Integrity checks, digital signatures
• Repudiation: Audit logs, non-repudiation
• Information disclosure: Encryption, access control
• Denial of service: Rate limiting, resource quotas
• Elevation of privilege: Principle of least privilege

Supply Chain Security

• Dependency scanning: Snyk, Dependabot, Trivy
• SBOM generation: CycloneDX or SPDX format
• Signed commits: GPG or SSH commit signing
• Artifact verification: Checksum validation, signature verification

Secrets Management

• Secrets never in code — always in secrets manager (Vault, AWS Secrets Manager)
• Rotation policy: Rotate database credentials every 90 days
• Access audit: Log every secrets access, alert on anomalies
• Encryption at rest and in transit for all secrets
• Principle of least privilege: each service gets only its own secrets

Architecture Decision Trees

Navigation Structure Decision Tree

How many top-level sections? ├── 3-5 → Flat navigation (all sections accessible from top) ├── 5-8 → Mega menu or hub-and-spoke with sub-navigation └── 8+ → Search-dominant with faceted navigation Is content primarily task-oriented or exploratory? ├── Task → Goal-focused navigation, shortcut paths └── Exploratory → Browse-focused, category-based discovery Mobile vs desktop usage ratio? ├── Mobile-first → Bottom navigation, hamburger menu, progressive disclosure └── Desktop-first → Sidebar or top navigation with visible labels

Content Organization Decision Tree

Is the content structured or unstructured? ├── Structured → Database-driven, taxonomy essential, metadata critical └── Unstructured → Full-text search primary, tagging secondary Does the content need versioning? ├── Yes → Content management system with revision history └── No → Static content with periodic updates Multi-language support needed? ├── Yes → Language-independent IA with locale-specific adaptations └── No → Single-language IA, no i18n considerations