Well-Architected Framework Assessment Skill

Evaluate Azure architectures against Microsoft's Well-Architected Framework (WAF) five pillars to identify strengths, risks, and improvement opportunities.

When to Use

• Assess existing or proposed Azure architectures
• Validate designs meet WAF best practices
• Identify architectural risks and gaps
• Provide scored assessment with recommendations
• Review before production deployment

Five Pillars Overview

1. Reliability

Ability of the system to recover from failures and continue functioning.

• Focus: Availability, resiliency, disaster recovery, health monitoring

2. Security

Protecting applications and data from threats.

• Focus: Identity, network security, encryption, secrets management

3. Cost Optimization

Managing costs to maximize value delivered.

• Focus: Right-sizing, reserved instances, monitoring, waste elimination

4. Operational Excellence

Operations processes that keep a system running in production.

• Focus: IaC, CI/CD, monitoring, incident response, automation

5. Performance Efficiency

Ability of a system to adapt to changes in load.

• Focus: Scaling, caching, CDN, resource selection, optimization

Assessment Process

Step 1: Analyze Architecture

Review the architecture for each pillar:

Reliability Checklist:

• Availability Zones configured?
• Multi-region deployment for critical workloads?
• Health checks and monitoring configured?
• Auto-healing and circuit breakers implemented?
• Backup strategy defined with RPO/RTO?
• Disaster recovery plan documented?

Security Checklist:

• Managed identities used (no credentials in code)?
• Private endpoints for PaaS services?
• HTTPS only with TLS 1.2+?
• Network security groups with least privilege?
• Key Vault for secrets management?
• Azure AD authentication and RBAC configured?
• Data encrypted at rest and in transit?

Cost Optimization Checklist:

• Resources right-sized for actual usage?
• Auto-scaling configured?
• Reserved instances considered for predictable workloads?
• Storage tiering implemented (Hot/Cool/Archive)?
• Unused resources identified?
• Cost monitoring and alerts configured?

Operational Excellence Checklist:

• Infrastructure as Code (Bicep/Terraform)?
• CI/CD pipelines implemented?
• Application Insights for telemetry?
• Centralized logging (Log Analytics)?
• Alerts configured for critical scenarios?
• Deployment automation and rollback?

Performance Efficiency Checklist:

• CDN for static content?
• Caching strategy (Redis, CDN)?
• Asynchronous processing for long operations?
• Appropriate compute SKUs selected?
• Auto-scaling rules defined?
• Performance testing completed?

Step 2: Score Each Pillar

Use 0-100 scoring system:

Scoring Criteria:

• 80-100 (Excellent): Meets all best practices, production-ready
• 60-79 (Good): Meets most practices, minor gaps
• 40-59 (Fair): Some practices missing, moderate risk
• 20-39 (Poor): Many gaps, significant improvements needed
• 0-19 (Critical): Major gaps, not production-ready

Step 3: Provide Recommendations

For each identified gap:

• Finding: What's missing or problematic
• Risk: Impact if not addressed
• Recommendation: Specific action to take
• Priority: Critical / High / Medium / Low
• Effort: Hours or days to implement

Assessment Output Format

# Well-Architected Framework Assessment
**Architecture**: [Name]
**Assessment Date**: [Date]
**Overall Score**: [Average of 5 pillars]/100

## Executive Summary
[2-3 sentences on overall health, key strengths, top risks]

## Pillar Scores

| Pillar | Score | Status |
|--------|-------|--------|
| Reliability | 75/100 | 🟢 Good |
| Security | 65/100 | 🟡 Fair |
| Cost Optimization | 80/100 | 🟢 Good |
| Operational Excellence | 70/100 | 🟡 Fair |
| Performance Efficiency | 85/100 | 🟢 Excellent |
| **Overall** | **75/100** | **🟢 Good** |

---

## 1. Reliability (75/100) - 🟢 Good

### Strengths
Availability Zones configured for App Service and Azure SQL
Health checks implemented with automatic failover
Backup strategy defined (RPO: 1 hour, RTO: 4 hours)

### Gaps & Recommendations

#### Finding #1: No Multi-Region Deployment
**Risk**: Regional outage causes complete service unavailability
**Recommendation**: Implement active-passive multi-region with Azure Front Door
**Priority**: High
**Effort**: 3-5 days
**Implementation**: Deploy secondary region (West US), configure Azure Front Door with priority routing

#### Finding #2: Missing Circuit Breaker Pattern
**Risk**: Cascading failures when dependencies are degraded
**Recommendation**: Implement circuit breaker using Polly library
**Priority**: Medium
**Effort**: 1-2 days

---

## 2. Security (65/100) - 🟡 Fair

### Strengths
Azure AD authentication configured
HTTPS enforced with TLS 1.2
Key Vault used for connection strings

### Gaps & Recommendations

#### Finding #1: Service Principal Used Instead of Managed Identity
**Risk**: Credential rotation required, potential secret exposure
**Recommendation**: Replace service principal with system-assigned managed identity
**Priority**: Critical
**Effort**: 4 hours
**Implementation**: 
1. Enable managed identity on App Service
2. Grant RBAC permissions to SQL and Key Vault
3. Remove service principal credentials

#### Finding #2: No Private Endpoints
**Risk**: PaaS services exposed to public internet
**Recommendation**: Implement private endpoints for SQL, Storage, Key Vault
**Priority**: High
**Effort**: 1 day

---

## 3-5. [Remaining Pillars Follow Same Structure]

---

## Priority Roadmap

### Critical (Fix Immediately)
1. Replace service principal with managed identity
2. Implement private endpoints for PaaS services

### High (Next 30 Days)
3. Multi-region deployment (active-passive)
4. Infrastructure as Code implementation
5. Implement comprehensive alerting

### Medium (Next 90 Days)
6. Circuit breaker pattern
7. Reserved instances for predictable workloads
8. Performance testing automation

### Low (Future Enhancements)
9. Chaos engineering tests
10. Additional caching layers

---

## Cost Impact Summary
- **Savings Opportunities**: ~$480/month (right-sizing, reserved instances)
- **Security Enhancements**: +$200/month (private endpoints)
- **Multi-Region**: +$850/month (passive region infrastructure)
- **Net Impact**: +$570/month for significantly improved resilience and security

---

## Conclusion
[Summary of assessment with key takeaways and prioritized next steps]

Tips for Effective Assessments

Be Specific: Reference exact resources and configurations Quantify Risk: Use concrete examples of potential impact Actionable Recommendations: Provide implementation steps, not just principles Prioritize Ruthlessly: Help teams focus on what matters most Show Business Impact: Connect technical gaps to business risks Include Quick Wins: Balance strategic improvements with fast fixes Cost-Aware: Show ROI for recommendations (cost vs benefit)

Avoid: Generic advice, overwhelming lists, missing priorities, theoretical recommendations