GCP Professional Cloud DevOps Engineer Syllabus 2026
Half this exam is CI/CD and observability. Here are Google's five sections with their published weightings, and the SRE material that is smaller than everyone assumes.
- 50-60Questions
- 2 hoursTime
- 5Sections
- YesWeights published
- Not publishedPass mark
- $200Fee
Table of Contents
The whole paper in one bar
The Professional Cloud DevOps Engineer exam guide is the exam guide Google Cloud publishes for Google Cloud Professional Cloud DevOps Engineer, restated so you can see the shape of the paper before you read a word of the detail. The bar below is the entire syllabus: 5 sections, in the order the certifying body lists them.
- §1 Bootstrapping and maintaining a Google Cloud organization — ~20%
- §2 Building and implementing CI/CD pipelines, including continuous testing — ~25%
- §3 Applying site reliability engineering practices — ~18%
- §4 Implementing observability practices and troubleshooting issues — ~25%
- §5 Optimizing performance and cost — ~12%
Those percentages are Google Cloud’s own, not an estimate. That matters more than it sounds: most pages ranking for this query quote weightings the vendor never published.
The syllabus, section by section
Each block below is one section of the official outline. The bullets are the sub-objectives Google Cloud publishes underneath it — the actual scope statement, not a summary of it.
Bootstrapping and maintaining a Google Cloud organization
- Designing the overall resource hierarchy
- Managing infrastructure as code
- Designing a CI/CD architecture stack across cloud, hybrid and multi-cloud
- Managing multiple environments such as staging and production
- Enabling secure cloud development environments
Building and implementing CI/CD pipelines, including continuous testing
- Designing pipelines for application, infrastructure and ML workloads
- Implementing and managing pipelines
- Managing pipeline configuration and secrets
- Securing the deployment pipeline
- Cloud Deploy, Kustomize, Skaffold, Argo CD and kpt
- Binary Authorization and SLSA
Applying site reliability engineering practices
- Balancing change, velocity and reliability
- Managing service lifecycle
- Mitigating incident impact on users
- SLIs, SLOs and error budgets
Implementing observability practices and troubleshooting issues
- Instrumenting and collecting telemetry
- Managing and analysing logs
- Managing metrics, dashboards and alerts
- Capturing and analysing distributed traces
- Troubleshooting issues
- OpenTelemetry and Managed Service for Prometheus
Optimizing performance and cost
- Collecting performance information in Google Cloud
- Implementing FinOps practices
- Optimising resource utilisation and cost
Watch the ordering. The certification landing page lists these competency areas in a different sequence from the exam guide PDF - it promotes SRE practices to second place. The PDF ordering above is the authoritative one, and it matches the weightings.
What the weightings really mean
A percentage in an exam guide is a promise about how many questions get drawn from a section, not about how hard those questions are. On a paper of roughly 55 items, Building and implementing CI/CD pipelines, including continuous testing at ~25% is worth about 14 questions and Optimizing performance and cost at ~12% is worth about 7. Missing the whole of the smallest section still leaves you a pass on paper; missing the largest one does not.
The trap is treating that arithmetic as permission to skip. Weightings are the floor of what a section costs you, because the low-weight sections on almost every blueprint are the ones that supply context for questions filed under a different heading. You lose those marks in the heavy section and never find out why.
Reading the verbs in the objectives
The verb at the front of each sub-objective is the part candidates skim and examiners take literally. It sets the depth you are tested at, and it is the difference between recognising a service name and being asked to choose between two that both look right.
| Verb | What a question at that depth looks like |
|---|---|
| Designing | Architecture-level judgement. A scenario with constraints and four workable answers, one of which survives the constraint you were told about. Section one and section two open with this verb. |
| Implementing / Managing | Operational depth. Which product, which setting, which order. This is where the exam stops being conceptual and starts assuming you have run something in production. |
| Troubleshooting | The verb that carries the observability section. Symptom first, and the correct answer is usually the diagnostic step rather than the fix. |
Google recommends three or more years of industry experience with at least one year on Google Cloud, and the verb mix is why. Almost nothing here is Describe. If you are studying this from reading alone, the troubleshooting objectives are where that shows up.
Study it in this order, not the syllabus order
Exam guides are written to describe a job role, not to teach one. The published order is almost never the order that builds knowledge fastest. This is the sequence that front-loads the material everything else depends on.
| Order | Why here |
|---|---|
| 1. Bootstrapping the organization | ~20%, and it establishes the resource hierarchy, IaC and environment model that every later section assumes. |
| 2. CI/CD pipelines | ~25%, the joint-largest section. Cloud Build, Cloud Deploy, Artifact Registry, progressive delivery and supply-chain security including Binary Authorization and SLSA. |
| 3. Observability and troubleshooting | The other ~25%. Do it straight after pipelines - you cannot troubleshoot a deployment strategy you have not learned yet. |
| 4. SRE practices | ~18%, and smaller than its reputation. SLIs, SLOs, error budgets, toil and blameless postmortems are quick to learn and heavily quoted. |
| 5. Performance and cost | ~12% and last. FinOps framing, resource utilisation and performance data collection. |
The two 25% sections together are half the paper. If you have a fixed number of study hours, split them 50/50 between CI/CD and observability first, then divide what is left across the other three.
What changed, and how to spot the next change
The guide carries no dated changelog, unlike the Cloud Digital Leader one, but the content is clearly refreshed. Machine learning workloads were added to the CI/CD section title. A new sub-objective covers secure cloud development environments including Cloud Workstations, Cloud Shell and Gemini Code Assist. Cloud Deploy, Kustomize, Skaffold, Argo CD and kpt appear in the pipeline architecture objective; Parameter Manager and Workload Identity Federation in secrets; Binary Authorization and SLSA in pipeline security; OpenTelemetry, Managed Service for Prometheus and synthetic monitors across observability; and the final section is now explicitly framed as FinOps.
Two habits keep you from studying a retired outline. First, open the official exam guide and look for its revision date before you buy anything — courses and question banks lag a syllabus change by months, and the cheapest ones never catch up. Second, re-check it the week you book. A revision announced after you started studying is still a revision you sit.
Turn the syllabus into questions
Reading an outline tells you the shape of the paper. Answering questions tells you which sections you would actually lose marks on. Start free, then $5.99 unlocks the full Professional Cloud DevOps Engineer bank.
Free practice testProfessional Cloud DevOps Engineer exam pageFAQ
What is the passing score for the Professional Cloud DevOps Engineer exam?
Google does not publish a passing score for any of its Cloud certifications. You receive a pass or fail result with no numeric score, so any specific percentage quoted online is a guess.
What are the Professional Cloud DevOps Engineer exam sections and weightings?
Five: Bootstrapping and maintaining a Google Cloud organization ~20%, Building and implementing CI/CD pipelines ~25%, Applying site reliability engineering practices ~18%, Implementing observability practices and troubleshooting ~25%, and Optimizing performance and cost ~12%. Those figures are printed in Google's exam guide PDF.
How much does the exam cost and how long is it?
US$200 plus tax, two hours, 50 to 60 multiple-choice and multiple-select questions, delivered online-proctored or at a test centre. It is offered in English and Japanese.
Do I need experience to pass this exam?
There is no formal prerequisite, but Google recommends three or more years of industry experience including at least one year designing and managing production systems on Google Cloud. The troubleshooting objectives in particular are hard to pass from reading alone.
Sources
Every section title, sub-objective and number on this page comes from the certifying body:
- https://cloud.google.com/learn/certification/cloud-devops-engineer
- https://services.google.com/fh/files/misc/professional_cloud_devops_engineer_exam_guide_english.pdf
- https://support.google.com/cloud-certification/answer/9907853
Checked September 2, 2026. Exam guides are revised without notice — confirm against Google Cloud before you build a study plan around this page.