SOA-C03 Exam Guide: What It Measures and How to Prepare
SOA-C03 is the AWS Certified CloudOps Engineer - Associate exam, formerly known as AWS Certified SysOps Administrator - Associate. It validates practical ability to deploy, manage, monitor, troubleshoot, secure, and operate AWS workloads. AWS targets candidates with approximately 1 year of relevant AWS experience and at least 1 year in an operations role. This guide helps you decide whether your current experience is sufficient, which domains deserve the most study time, and how to schedule preparation around hands-on practice rather than memorization.
What does SOA-C03 validate?
SOA-C03 validates the operational side of AWS work: supporting workloads according to the AWS Well-Architected Framework, using the AWS Management Console and AWS CLI, applying security controls, monitoring systems, troubleshooting failures, and remediating incidents. It also covers networking, high availability, performance, capacity, business continuity, and disaster recovery.
The exam is intended for CloudOps engineers. Its focus is not simply knowing what an AWS service does. A stronger candidate can connect a symptom to an appropriate metric or log, identify the likely control or configuration involved, and select an operationally suitable response.
AWS also identifies several activities as outside the target candidate’s expected scope, including designing distributed architectures, designing CI/CD pipelines, developing software, defining security or governance requirements, and assessing and planning resource capacity. Use those boundaries to prevent your study from expanding into an unfocused review of every AWS topic.
Who should take this exam?
The intended candidate has approximately 1 year of experience with AWS deployment, management, troubleshooting, networking, and security, plus at least 1 year in a related operations role such as system administration. That profile is a useful readiness test: certification study can organize knowledge, but it does not replace familiarity with operational decisions.
AWS recommends general IT knowledge in monitoring, logging, troubleshooting, DNS, TCP, IP, firewalls, high availability, performance, capacity, scripting, operating systems, cloud computing, containers, orchestration, CI/CD, and Git. It also recommends experience with the AWS Well-Architected Framework, storage and container solutions, monitoring tools, infrastructure as code, networking and security services, cloud financial management, hybrid and multi-VPC operations, databases, and compute services.
If your experience is mostly console navigation or introductory service reading, build an operating foundation before choosing a test date. If you already maintain AWS workloads, begin with the blueprint and use targeted labs to close gaps. Candidates changing from SOA-C02 should not assume that old notes cover the current exam.
What changed from SOA-C02?
SOA-C03 began being used on September 30, 2025, while SOA-C02 was listed as being in use through September 29, 2025. The certification name also changed: candidates who pass SOA-C03 receive the AWS Certified CloudOps Engineer - Associate name, and AWS says the change is not retroactive for holders of AWS Certified SysOps Administrator - Associate.
The current comparison identifies additions involving CloudFormation and the AWS CDK, compliance requirements such as Region and service selections, and CloudWatch network monitoring services. It also identifies movement of VPN content into networking and a shift of former SOA-C02 Tasks 6.1 and 6.2 into SOA-C03 Task 1.3.
AWS removed Configure S3 static website hosting from Task 5.2. Do not make that topic a major study objective merely because it appears in older SysOps materials. Review the current exam guide, comparison page, and revisions page before relying on an older course or question bank.
How is the exam structured?
SOA-C03 is an associate-level exam with 65 questions, a 130-minute testing period, and multiple-choice or multiple-response formats. AWS states that 50 questions affect your score and that the exam includes 15 unscored questions. Each scored multiple-choice or multiple-response question counts as a single scored opportunity.
Results are reported as a scaled score of 100–1,000, and the minimum passing score is 720. Treat the score as an AWS reporting scale rather than a simple percentage target. AWS also cautions candidates when interpreting section-level feedback, so avoid assuming that a domain result precisely represents your overall capability.
The listed exam languages are English, Japanese, Korean, and Simplified Chinese. SOA-C03 is offered through Pearson VUE testing centers or online proctoring. The listed exam price is 150 USD, while AWS directs candidates to its exam-pricing information for foreign-exchange and other cost details. Verify current delivery and pricing information before booking.
What should you confirm before booking?
Confirm that the exam version, language, delivery option, and scheduling information match your plan on the official AWS certification site. These details can change, and the official page is the proper source for current booking instructions.
Choose a testing-center appointment if a controlled physical environment is easier for you, or consider online proctoring if your equipment, room, and connectivity meet the provider’s requirements. This is a practical choice, not an exam-content requirement.
Do not schedule solely because you have completed a video course. Schedule when you can explain operational trade-offs, perform the core workflows in a practice account, and consistently diagnose unfamiliar scenarios without depending on recalled question wording.
Which SOA-C03 domains deserve study time?
The blueprint gives three domains 22% each, one domain 16%, and one domain 18% of scored content. AWS lists the five domains as monitoring and performance operations; reliability and business continuity; deployment, provisioning, and automation; security and compliance; and networking and content delivery. Use the domain labels with every weight when planning study time.
Monitoring, Logging, Analysis, Remediation, and Performance Optimization — 22% of scored content
Content Domain 1: Monitoring, Logging, Analysis, Remediation, and Performance Optimization represents 22% of scored content. Prepare to reason from metrics, logs, traces, alarms, events, and operational symptoms toward a corrective action.
Study how Amazon CloudWatch, AWS CloudTrail, AWS Systems Manager, AWS Lambda, auto scaling, AWS X-Ray, Amazon Managed Service for Prometheus, and related services fit into an observability and remediation workflow. The current skills include configuring monitoring and logging for workloads, analyzing performance metrics, automating remediation, and selecting shared storage for specific requirements.
A useful lab is to create a workload, collect its logs and metrics, define an alarm, and connect an operational response. Then deliberately introduce a failure or performance issue and record which evidence distinguishes a capacity problem from an application, permission, network, or dependency problem.
Reliability and Business Continuity — 22% of scored content
Content Domain 2: Reliability and Business Continuity represents 22% of scored content. Prepare for decisions involving resilient architecture, backup and restore, recovery procedures, scaling, and the operational consequences of a regional or component failure.
The exam guide includes high availability, performance, and capacity requirements, along with business continuity and disaster recovery procedures. The revised skill for disaster recovery explicitly includes backup and restore, pilot light, warm standby, and active/active approaches.
Study these approaches as operational choices tied to recovery needs, not as a vocabulary list. For each one, describe what must already exist, what is activated during recovery, what data or configuration must be restored, and which verification steps would demonstrate that the workload is usable.
Deployment, Provisioning, and Automation — 22% of scored content
Content Domain 3: Deployment, Provisioning, and Automation represents 22% of scored content. Focus on repeatable resource creation, safe changes, event-driven responses, and the operational management of infrastructure as code.
The current outline includes creating and managing stacks with AWS CloudFormation and the AWS CDK, as well as implementing event-driven automation with services and features such as Lambda, Amazon S3 Event Notifications, and Amazon EventBridge. AWS also lists AWS CLI use and infrastructure as code among recommended knowledge areas.
Build a small stack, update it, inspect the change set or resulting resources, and remove it cleanly. Separately, create an event-driven workflow and trace the event, permissions, target, logs, and failure handling. The goal is to understand dependencies and rollback implications rather than memorize template syntax.
Security and Compliance — 16% of scored content
Content Domain 4: Security and Compliance represents 16% of scored content. Although it has the smallest listed weighting, it affects decisions across nearly every AWS operation, so do not treat it as an optional final review.
Study IAM, security groups, network ACLs, AWS KMS, AWS Secrets Manager, AWS Config, AWS Security Hub, Amazon GuardDuty, Amazon Inspector, Amazon Macie, AWS WAF, AWS Shield, and related services in terms of control purpose and remediation. Current skills include securely implementing multi-account strategies and enforcing compliance requirements, including Region and service selections.
Practice mapping a finding to an owner, evidence source, corrective action, and verification step. Include AWS Organizations, service control policies, and IAM Identity Center in your multi-account review. Also distinguish identity permissions, resource policies, network controls, encryption controls, and detective controls; confusing these categories is a common source of wrong answers.
Networking and Content Delivery — 18% of scored content
Content Domain 5: Networking and Content Delivery represents 18% of scored content. Prepare to troubleshoot connectivity and choose private or public paths using routing, DNS, filtering, endpoints, load balancing, and monitoring evidence.
The in-scope list includes Amazon VPC, VPC endpoints, VPC peering, AWS PrivateLink, AWS Transit Gateway, AWS Site-to-Site VPN, AWS Client VPN, Amazon Route 53, Route 53 Resolver, VPC Flow Logs, VPC Reachability Analyzer, Elastic Load Balancing, Amazon CloudFront, and AWS Global Accelerator. The current outline also includes configuring private networking connectivity and analyzing CloudWatch network monitoring services.
Draw traffic paths instead of studying network services in isolation. For each path, identify the source and destination, route tables, security groups, network ACLs, DNS resolution, endpoint or gateway choice, and logging evidence. A diagram that explains why traffic fails is more useful than a list of product definitions.
Which AWS services should be part of your review?
The official in-scope list is non-exhaustive and subject to change, so use it as a boundary and cross-check rather than as a promise that every listed service receives equal attention. Start with services that recur in the domains and in the tasks you perform at work.
AWS identifies Amazon EC2, Amazon ECS, Amazon EKS, AWS Lambda, Amazon RDS, Amazon DynamoDB, Amazon CloudWatch, AWS CloudTrail, AWS CloudFormation, AWS CDK, IAM, Amazon VPC, Elastic Load Balancing, Amazon S3, and AWS Backup among the services or features in scope.
Extend that core with the services named in the domain skills: AWS Systems Manager, AWS Config, AWS Organizations, service control policies, IAM Identity Center, AWS Security Hub, Amazon GuardDuty, Amazon Inspector, AWS PrivateLink, VPC endpoints, Amazon Route 53, VPC Flow Logs, AWS Transit Gateway, and AWS X-Ray. The right question for each service is: what operational problem does it solve, what evidence does it produce, and what permissions or dependencies does it require?
AWS periodically reviews the exam guide. Its revisions page says changes are published at least 1 month before they are reflected on the exam. Check the revisions and in-scope services pages near your booking date, especially if your preparation spans a revision announcement.
How should you study each topic?
Use a four-part loop: read the task statement, perform the workflow, explain the failure modes, and test yourself with a new scenario. This approach turns service familiarity into operational judgment and exposes gaps that passive reading hides.
For every service or feature, make a compact study record with five fields: purpose, configuration path, telemetry, security boundary, and recovery or remediation action. For example, a monitoring record should say what is collected, where it appears, what alarm or event can be created, which identity performs the action, and how you would verify the result.
Prefer small, disposable labs over one large project that becomes difficult to reset. A lab might combine EC2, CloudWatch, IAM, Systems Manager, a VPC, and CloudFormation. Change one variable at a time, capture the symptom, and write the shortest reliable diagnostic sequence.
Use the AWS Management Console and AWS CLI during practice because both are explicitly relevant to the exam. You do not need to memorize every command flag, but you should recognize the workflow, the required permissions, the resource state, and the kind of output that confirms success.
Study cost awareness without turning it into unsupported formula memorization. AWS recommends cloud financial management, and the in-scope list includes AWS Cost and Usage Reports, AWS Cost Explorer, and Savings Plans. Learn how an operator investigates usage and selects a cost-management action when the scenario supplies the necessary context.
What is a practical study roadmap?
A flexible roadmap should begin with diagnosis, move through the three 22% domains, then integrate security and networking across realistic workloads. The sequence below is a recommendation, not an AWS scheduling requirement; adjust it to your experience and available lab time.
Start by downloading the current exam guide and marking every task statement as strong, familiar, or unknown. Do not begin with random practice questions. Your first objective is to identify whether the weakness is service knowledge, troubleshooting method, networking fundamentals, or reading the scenario precisely.
Next, build a baseline workload with compute, storage, identity, networking, and monitoring. Use it to practice logs, metrics, alarms, access control, deployment, and recovery. Keep a written change log. When a lab fails, record the first observable symptom and the evidence that led to the fix.
Then study Domain 1 and Domain 2 together. Monitoring evidence informs reliability decisions, while recovery planning depends on knowing what state, data, and configuration are observable and restorable. Follow with Domain 3 so that deployment and automation can reproduce the environment and respond to events.
Study Domain 4 through the workload you already built rather than as a detached security catalog. Add least-privilege permissions, encryption, findings, compliance checks, and multi-account boundaries. Finish the first pass with Domain 5 by tracing traffic through the same environment.
In the final review phase, use scenario-based practice that is not based on leaked or recalled exam content. For each missed item, write why the selected option failed, what evidence the correct choice uses, and which task statement it represents. Revisit only the underlying gap instead of rereading every service.
Before booking, repeat the blueprint audit. You should be able to explain your diagnostic sequence aloud, perform representative configurations, and identify trade-offs involving reliability, security, automation, and network connectivity. If one domain remains entirely theoretical, postpone the appointment and create a lab for that domain.
A four-pass weekly pattern
Pass 1 is orientation: read the official domain tasks and map them to services. Pass 2 is implementation: configure a small workflow in the console and CLI. Pass 3 is failure analysis: remove a permission, route, metric, dependency, or resource and diagnose the result. Pass 4 is retrieval: answer scenario questions and explain the reasoning without looking at notes.
Keep an error ledger with separate columns for misunderstood requirement, incorrect service choice, missing prerequisite, and careless reading. These categories lead to different fixes. A missing prerequisite requires a dependency diagram; an incorrect service choice requires comparison notes; careless reading requires slower extraction of constraints such as Region, recovery objective, or private connectivity.
Which mistakes waste preparation time?
The most expensive mistake is treating SOA-C03 as a product glossary. Operational scenarios combine services and constraints, so memorize less and practice cause-and-effect more. A correct definition of CloudWatch does not by itself show that you can select a useful metric, alarm, log source, or remediation path.
Do not use old SOA-C02 material without checking the comparison and revisions pages. Older notes may overemphasize S3 static website hosting, omit CloudFormation and AWS CDK work, or place VPN topics in the wrong domain. They can still provide background, but they should not define your current study priorities.
Do not turn the blueprint into a bare percentage ranking. Domain 1: Monitoring, Logging, Analysis, Remediation, and Performance Optimization is 22% of scored content; Domain 2: Reliability and Business Continuity is 22%; Domain 3: Deployment, Provisioning, and Automation is 22%; Domain 4: Security and Compliance is 16%; and Domain 5: Networking and Content Delivery is 18%. The labels matter because each percentage describes a different body of work.
Avoid studying only successful configurations. Operators are tested on symptoms, constraints, and remediation. Practice failures such as insufficient IAM permissions, missing routes, incorrect security-group rules, absent log collection, unhealthy targets, failed stack updates, and incomplete backup recovery.
Finally, do not rely on dumps, leaked questions, or memorized answer patterns. They do not establish that you understand the task, can handle a changed scenario, or are using current content. Use legitimate learning resources, the official exam guide, AWS documentation, and hands-on exercises instead.
How should you approach questions?
Read the requested outcome before examining the answer choices. Extract the workload, failure symptom, constraints, and operational objective. Then eliminate choices that solve a different problem, require unnecessary redesign, weaken security, or ignore the stated deployment or network boundary.
For troubleshooting scenarios, ask four questions: what is failing, where is the first evidence, which layer owns that evidence, and what change is least disruptive? For architecture or recovery scenarios, ask what must remain available, what data state is acceptable, and what the operator must already have configured.
Multiple-response questions require every selected choice to satisfy the stated requirement. Do not select an answer merely because it is generally useful. Multiple-choice questions often contain plausible services that address the symptom indirectly; choose the option whose mechanism matches the constraint.
Use the allotted testing time deliberately. Mark a question only after recording why it is uncertain, then return with the scenario’s constraints in view. Because unscored questions are mixed into the exam and are not identified, give each question the same professional attention and do not try to detect which questions affect the result.
What should you do next?
Start with the official SOA-C03 exam guide, comparison page, revisions page, and in-scope services list. Confirm the current version before purchasing a course, booking an appointment, or building a study checklist. Then compare the task statements with your real AWS responsibilities and select the first lab that addresses your weakest operational skill.
If you are moving from SOA-C02, make a change-focused checklist: CloudFormation and AWS CDK, compliance and Region or service selection, CloudWatch network monitoring, revised private-connectivity examples, and the reclassified tasks. Remove S3 static website hosting from your core SOA-C03 objectives unless you are studying it for broader AWS knowledge.
After each lab, write a short incident record containing the symptom, evidence, root cause, correction, and verification. After each practice session, update the error ledger. Book only when your results are supported by repeatable reasoning and hands-on understanding, not by recognition of familiar wording.
Use the official AWS certification page for the current price, languages, delivery choices, and scheduling process. Use the official revisions page for changes that may affect your exam. That final verification protects your preparation plan from relying on stale administrative or content information.
Sources and currency checks
This guide uses AWS certification and AWS documentation sources only. AWS describes the in-scope services list as non-exhaustive and subject to change, and it periodically reviews the exam guide. Recheck the linked official pages before final revision and before scheduling.
The name change applies to candidates who pass SOA-C03 and is not retroactive for existing AWS Certified SysOps Administrator - Associate holders. That distinction matters when deciding how to describe a current or previous certification on a résumé or internal skills record.
Conclusion
SOA-C03 preparation is strongest when it mirrors the work the exam validates: observe a workload, understand its dependencies, apply a controlled change, protect it with appropriate controls, and verify recovery or normal operation. Use the official blueprint to set priorities, the current service and revision pages to keep the scope current, and small failure-focused labs to turn reading into judgment. Your next step is to complete a domain-by-domain gap review, build the first targeted lab, and verify current scheduling details before selecting an exam date.
