The AWS Well-Architected Framework
The exam guide states that SAA-C03 validates your ability to design solutions based on the AWS Well-Architected Framework. The framework is AWS's collection of best practices for cloud workloads, organised into six pillars. The exam's four domains map directly onto four of them.
The six pillars
| Pillar | Core question | Exam domain |
|---|---|---|
| Operational excellence | Can we run, monitor and improve the system safely? | Throughout (automation, monitoring) |
| Security | Are data and systems protected? | Domain 1 — Secure (30%) |
| Reliability | Does it work correctly and recover from failure? | Domain 2 — Resilient (26%) |
| Performance efficiency | Does it use resources efficiently as demand changes? | Domain 3 — High-performing (24%) |
| Cost optimization | Does it deliver value at the lowest price? | Domain 4 — Cost-optimized (20%) |
| Sustainability | Does it minimise environmental impact? | Indirectly (efficient resource use) |
Key design principles
General
- Stop guessing capacity — scale with demand.
- Test at production scale — environments on demand, torn down afterwards.
- Automate — infrastructure as code, automated recovery.
- Allow architectures to evolve — small, reversible changes.
- Drive decisions with data — metrics, not opinions.
Operational excellence
- Perform operations as code (CloudFormation, Systems Manager).
- Make frequent, small, reversible changes.
- Anticipate failure and learn from operational events.
Security
- Implement a strong identity foundation — least privilege, centralised identity, no long-lived credentials.
- Enable traceability — log and monitor everything (CloudTrail, CloudWatch, Config).
- Apply security at all layers — edge, VPC, subnet, instance, application, data.
- Automate security best practices.
- Protect data in transit and at rest.
- Keep people away from data — reduce manual access.
- Prepare for security events.
Reliability
- Automatically recover from failure — health checks, Auto Scaling, Multi-AZ.
- Test recovery procedures.
- Scale horizontally to reduce the impact of a single failure.
- Stop guessing capacity.
- Manage change through automation.
Performance efficiency
- Democratise advanced technologies — use managed services (e.g. managed databases, ML services).
- Go global in minutes — Regions, CloudFront.
- Use serverless architectures.
- Experiment more often.
- Consider mechanical sympathy — use the technology that fits the access pattern.
Cost optimization
- Implement cloud financial management.
- Adopt a consumption model — pay only for what you use.
- Measure overall efficiency.
- Stop spending on undifferentiated heavy lifting — managed services.
- Analyse and attribute expenditure — tags, Cost Explorer.
Sustainability
- Understand your impact; maximise utilisation; use managed services and efficient hardware; reduce downstream impact.
The Well-Architected Tool
The AWS Well-Architected Tool (in the console) walks you through questions for each pillar against a workload and produces a list of risks and improvement plans. Lenses add specialised guidance (e.g. serverless, SaaS).
When two answers both work, ask which one better follows these principles: managed over self-managed, automated over manual, multi-AZ over single, least privilege over broad access, pay-per-use over over-provisioned.
The four exam domains are four pillars in disguise. Learning the pillars' design principles gives you a reliable way to break ties between plausible answers.
Senior Full-Stack Engineer · AI & AWS. I design and run production systems on AWS — serverless, data and AI.