Introduction
The Claude Certified Associate – Foundations (CCAO-F) exam is organised into seven domains, each carrying a published weighting. Knowing those weightings tells you where the marks are and where to spend your study time. This post walks through all seven, explains what each tests, and points to where you can practise it.
CCA Prep is an independent practice platform and is not affiliated with, authorised by, or endorsed by Anthropic. Claude and Anthropic are trademarks of Anthropic PBC. The domain names and weightings below reflect the publicly published exam information.
The Seven Domains at a Glance
| # | Domain | Weight |
|---|---|---|
| 1 | Prompting and Task Execution | 14% |
| 2 | Output Evaluation and Validation | 21% |
| 3 | Product and Model Selection | 12% |
| 4 | Workflow Integration and Solution Design | 16% |
| 5 | Configuration and Knowledge Management | 12% |
| 6 | Governance, Risk, and Responsible Use | 15% |
| 7 | Troubleshooting and Optimization | 10% |
With 60 questions on the exam, a domain's weighting is roughly how many questions to expect from it. Domain 2 alone is about 13 questions; Domain 7 is about 6.
One pattern is worth noticing before we go further. Domains 2, 4, and 6 add up to about 52% of the exam, and all three are about judgement rather than recall. More than half the exam asks you to make a sensible call, not to remember a fact.
Domain 1: Prompting and Task Execution (14%)
This domain covers the mechanics of asking Claude to do a task well: giving a role, task, context, format, and examples; supplying source material; controlling structure, tone, and length; breaking work into steps; and avoiding common pitfalls such as ambiguity or overloading a single prompt.
It is the natural starting point, and on our platform it is the free domain. Practise Prompting and Task Execution for free.
Domain 2: Output Evaluation and Validation (21%)
The largest domain, and the heart of the exam. It tests whether you can tell a good output from a plausible-looking bad one: assessing factual accuracy and hallucination risk, verifying claims against sources, spotting bias and overconfidence, setting acceptance criteria, running human-in-the-loop review, and deciding when to trust, edit, or reject what Claude produces.
If you have limited study time, this is where to spend it. Practise Output Evaluation and Validation.
Domain 3: Product and Model Selection (12%)
This domain is about choosing the right tool for the job: understanding the Claude product family, matching a model tier to a task based on capability, speed, and cost, knowing when Claude is the right choice at all, and working within context-window limits. Practise Product and Model Selection.
Domain 4: Workflow Integration and Solution Design (16%)
The second-largest domain. It covers identifying which tasks are good candidates for AI, designing human and AI workflows with sensible hand-offs and checkpoints, integrating Claude with existing tools and data, standardising reusable prompts across a team, measuring impact, and scaling from individual to team use. Practise Workflow Integration and Solution Design.
Domain 5: Configuration and Knowledge Management (12%)
This domain looks at setting things up for consistency: Projects and custom instructions, managing reference material and project knowledge, building reusable assets such as saved prompts and templates, curating context, and keeping knowledge current and accurate. Practise Configuration and Knowledge Management.
Domain 6: Governance, Risk, and Responsible Use (15%)
A substantial judgement domain and one that matters a great deal to employers. It covers data privacy and confidentiality, responsible and acceptable use, mitigating harmful or biased outputs, transparency and disclosure of AI use, regulatory and compliance considerations, security basics, and intellectual property. Practise Governance, Risk, and Responsible Use.
Domain 7: Troubleshooting and Optimization (10%)
The smallest domain, focused on fixing things when they go wrong: diagnosing poor or off-target outputs, refining prompts to address common failure modes, managing long contexts and truncation, improving speed and cost, and iterating toward reliable, repeatable results. Practise Troubleshooting and Optimization.
Where to Focus Your Study
If you rank the domains by weight, a clear priority order emerges: Output Evaluation (21%), Workflow Integration (16%), Governance (15%), then Prompting (14%). Those four cover two-thirds of the exam, and three of them are judgement-led. Get comfortable making the kind of call those questions ask for and you have covered most of the marks.
For a full study plan built around these weightings, see How to Pass the Claude Certified Associate (CCAO-F).
Start Practising
The best way to learn the domains is to answer questions in them and read the explanations. Start with the free Prompting domain, then work through the rest from the CCAO-F practice hub.