Claude Certified Associate – Foundations
The Associate – Foundations exam is for people who get real work done with Claude every day: analysts, project leads, marketers, operations and HR teams, consultants. It tests whether you can brief Claude well, check what comes back, set up Projects and knowledge sources properly, and know where the risks are. No code is involved.
Intended audience: Professionals who use Claude as a productivity tool and build Claude Projects in functions such as operations, marketing, project management, education, communications and general knowledge work. Internal staff and external consultants both qualify.
Exam content update: Pearson VUE lists updated content for this exam effective 7 December 2026, with a revised exam guide due 17 November 2026. The facts on this page reflect the current guide, version 1.0 (effective July 2026), and will be re-verified when the new guide is published. Source
- Exam code
- CCAO-F
- Level
- Foundations
- Official fee
- $99
- Paid to Anthropic, not AI7x7
- Duration
- 120 minutes
- Questions
- 60 items
- Format
- Multiple choice & multiple response
- Passing score
- 720 scaled
- Scale 100–1,000
- Validity
- 12 months
Verified against official exam guide version 1.0 (effective July 2026) on 4 November 2026.
Eligibility, prerequisites and registration
Official exam registration is through the Anthropic Partner Academy and is described by Anthropic as available to members of the Claude Partner Network (membership is free; firms join at claude.com/partners). AI7x7 resources may be used by anyone for learning and preparation.
Prerequisites: None. No mandatory prerequisites or required courses. No software development or API experience needed. Credential awarded on exam performance alone.
Partner Network: Does not count toward Claude Partner Network tier eligibility.
Delivery: Proctored, online (Pearson OnVUE) or at a Pearson VUE test centre. Waiting periods of 14 days after the first failed attempt, 30 days after the second, 90 days after the third. Maximum four attempts per exam in any rolling 12-month period. Fee applies to each attempt.
Who should take this
Built for people who do this work
- Business analysts, project and programme managers who use Claude for research, planning and documentation
- Marketing, communications, HR, operations and education professionals who draft, summarise and analyse with Claude
- Consultants and delivery leads who need to show clients they can scope and run Claude-assisted work responsibly
- Anyone building Claude Projects with instructions and knowledge sources for a team, without writing code
Not intended for: Software developers building against APIs or designing agentic systems; ML or advanced AI system specialists (those escalate to the Developer and Architect credentials).
Recommended experience
- — Regular hands-on professional use of Claude
- — Foundation in structured problem-solving and workflow design
- — Experience in roles such as business analyst, project manager, operations lead, or marketing/communications/HR/education professional
- — Practical understanding of AI limitations (hallucinations, context constraints, data sensitivity)
What it validates
A pass shows that
- You can turn a vague business request into a prompt, or a sequence of prompts, that produces usable output
- You can tell when an answer is wrong, incomplete, biased or made up, and you know what to verify and how
- You can pick the right Claude feature and model for a task and manage context limits sensibly
- You can configure and maintain a Project — instructions, uploaded knowledge, connectors — so a team gets consistent results
- You understand data sensitivity, organisational AI policy and the ethics of putting AI output in front of other people
Minimally qualified candidate (official guide): Uses Claude as a core productivity tool, has applied knowledge of its capabilities (prompting, task orchestration, Projects, Artifacts, workflows) and can independently complete common productivity tasks.
Exam blueprint
Official domains and weights
7 domains, reproduced from the official exam guide 1.0 (effective July 2026).
1.Prompting and Task Execution
14%- Create effective prompts for business and technical tasks
- Decompose complex requests into structured steps
- Iterate on prompts to improve output quality
- Adjust prompting strategy by task type (analysis, research, drafting, brainstorming)
2.Output Evaluation and Validation
21%- Evaluate outputs for accuracy and completeness
- Spot hallucinations, inconsistencies and biases
- Apply fact-checking and validation techniques
- Decide when human review or further verification is needed
- Edit, adapt, refine and compare outputs for the intended audience
- Organise information and choose output formats (artifacts, inline, structured data)
3.Product and Model Selection
12%- Select Claude product features (Projects, research mode, chat, Artifacts)
- Differentiate Claude model types (Haiku, Sonnet, Opus)
- Match model choice to task needs for cost, speed and quality
- Manage context limits and memory (when to restart, summarise or persist)
4.Workflow Integration and Solution Design
16%- Analyse requirements and use cases with Claude
- Use Claude for research, planning and process optimisation
- Support solution design, development and iteration
- Integrate Claude into existing workflows to augment or redesign them
- Communicate Claude's value and limitations to stakeholders
5.Configuration and Knowledge Management
12%- Configure Projects with instructions and knowledge sources
- Manage uploaded knowledge and connectors (e.g. Google Drive, Gmail)
- Write effective system-level instructions
- Maintain and update configurations, knowledge sources and instructions
6.Governance, Risk, and Responsible Use
15%- Identify appropriate and inappropriate use cases
- Apply data sensitivity, regulatory and privacy considerations
- Follow organisational AI policies and governance standards
- Understand the ethical implications of AI use
7.Troubleshooting and Optimization
10%- Diagnose and resolve underperforming prompts and poor outputs
- Adjust approach based on feedback and results
- Optimise workflows for efficiency and effectiveness
Weights are reproduced from the official exam guide. Domain percentages are reported on the official score report but do not individually determine pass/fail.
Skills and practical examples
What each domain looks like at work
1.Prompting and Task Execution
14%Skills
- — Write prompts with role, goal, inputs, constraints and output format
- — Break a multi-part request into ordered steps
- — Iterate: diagnose a weak output and change one thing at a time
- — Change strategy by task type: analysis vs drafting vs brainstorming
In practice
- A regional manager pastes six store reports and asks for 'insights'. The strong prompt names the audience, the decision the insights support, the metrics that matter and the format of the answer.
- A first draft of a policy summary is too long. The fix is a follow-up with a word limit and a reading level, not a rewrite of the whole prompt.
2.Output Evaluation and Validation
21%Skills
- — Check facts, figures and citations before they leave your desk
- — Spot hallucinated references, invented statistics and silent assumptions
- — Decide when a human expert must review
- — Choose the output format: inline text, an artifact, a table or structured data
In practice
- Claude produces a competitor table with revenue figures. Each number is a claim to verify against a filing or a named source before it goes in a board deck.
- A legal-sounding paragraph in a customer email is exactly the kind of output that needs a human sign-off regardless of how confident it reads.
3.Product and Model Selection
12%Skills
- — Choose between chat, Projects, research mode and Artifacts
- — Distinguish Haiku, Sonnet and Opus by cost, speed and depth
- — Know when to start a fresh conversation, summarise, or persist knowledge in a Project
In practice
- Daily summarisation of support tickets is a Haiku-class task; a one-off strategy memo synthesising twenty sources is where a stronger model earns its cost.
- A 200-page tender document should live in a Project's knowledge base, not be pasted into chat each morning.
4.Workflow Integration and Solution Design
16%Skills
- — Map a workflow and find the steps Claude can augment or replace
- — Use Claude for requirements analysis, research and process design
- — Explain Claude's value and its limits to stakeholders in their language
In practice
- An onboarding process has eleven manual steps. Four are drafting and summarising (good candidates); two involve approvals that must stay with people.
- Telling a sceptical director that Claude 'can be wrong, so here is the review step we added' builds more trust than overselling accuracy.
5.Configuration and Knowledge Management
12%Skills
- — Write Project instructions that encode house style, audience and guardrails
- — Curate uploaded knowledge: current, relevant, not contradictory
- — Use connectors (Drive, Gmail) deliberately and know what they expose
- — Review and update configurations as sources change
In practice
- A Project for a sales team includes the current price list, the approved messaging and an instruction never to quote discounts. When prices change, the old file is removed, not left alongside the new one.
- Project instructions saying 'always answer in British English, cite the source document by name' remove a class of recurring corrections.
6.Governance, Risk, and Responsible Use
15%Skills
- — Recognise inappropriate use cases: medical, legal, high-stakes decisions without review
- — Apply data sensitivity and privacy rules to what you paste and upload
- — Follow organisational AI policy and know who owns exceptions
- — Reason about the ethics of AI-generated content in front of customers and colleagues
In practice
- Pasting a customer's full support history including card details into a chat to 'summarise the issue' is a data-handling failure even if the summary is excellent.
- Using Claude to draft performance reviews is a governance question (policy, fairness, disclosure) before it is a prompting question.
7.Troubleshooting and Optimization
10%Skills
- — Diagnose why an output is poor: unclear task, missing context, wrong model, overloaded context
- — Adjust based on feedback rather than re-rolling
- — Reduce effort per task: templates, Projects, reusable instructions
In practice
- Answers drift halfway through a long session. Starting a new conversation with a tight summary fixes it faster than more instructions.
- A weekly report prompt that needs ten tweaks each time belongs in a Project with a saved instruction, not in someone's clipboard.
Example scenario
Rolling out Claude to a 40-person operations team
You lead operations at a logistics company. Your team spends hours each week turning supplier emails and spreadsheets into a status report for the COO. Leadership has approved Claude for the team and asked you to make it 'actually useful' within a month, while the data protection officer has flagged that supplier contracts contain commercially sensitive terms.
Your task: Decide how to set this up: what goes into a Project and what stays out, what the instructions should say, which model is appropriate for the weekly report versus ad-hoc analysis, how the team should verify figures before the report reaches the COO, and how you would explain the limits to the COO.
An original AI7x7 scenario illustrating the type of professional problem the exam tests. It is not an official exam item.
What this scenario tests
- Workflow integration and solution design (Domain 4): identifying which steps Claude augments and which stay manual
- Configuration and knowledge management (Domain 5): what belongs in a Project and how instructions encode house rules
- Governance, risk and responsible use (Domain 6): handling the sensitive contract terms the DPO raised
- Output evaluation and validation (Domain 2): the verification step before anything reaches leadership
- Product and model selection (Domain 3): matching model and feature to a recurring versus a one-off task
How to prepare
A 7-day study plan
Seven focused sessions of 60–90 minutes. The exam rewards judgement built from use, so every day includes hands-on work in Claude, not just reading.
- 1
Blueprint and prompting (Domain 1)
- — Read the official exam guide end to end and note the seven domains and weights
- — Rewrite three of your own recent prompts using role, goal, inputs, constraints, format
- — Practise decomposing one complex request into ordered steps
- 2
Evaluation and validation (Domain 2, the heaviest domain)
- — Take a Claude-generated summary of something you know well and mark every claim: verified, unverifiable, wrong
- — Practise asking for sources and checking them
- — Decide, for five outputs, whether a human review is required and why
- 3
Products and models (Domain 3)
- — Run the same task on Haiku, Sonnet and Opus; compare speed, cost and depth
- — Use Projects, research mode and Artifacts for one task each
- — Hit a context limit deliberately and recover by summarising
- 4
Workflow integration (Domain 4)
- — Map one real process in your team step by step
- — Mark which steps Claude can augment, replace or must not touch
- — Write a one-paragraph explanation of value and limits for a stakeholder
- 5
Configuration and knowledge (Domain 5)
- — Build a Project with instructions and two knowledge files
- — Add a connector and note exactly what it can reach
- — Change a source file and update the Project; observe what breaks
- 6
Governance and troubleshooting (Domains 6 and 7)
- — Read your organisation's AI policy (or a published one) and list three things it forbids
- — Classify ten tasks as appropriate, appropriate-with-review, or inappropriate
- — Take three bad outputs and diagnose the cause before fixing
- 7
Practice exam and review
- — Sit the AI7x7 20-question practice exam under the timer
- — Review every explanation, including the distractors
- — Re-read guide sections for any domain under 70% and note the gaps
Preparation tips
- — Treat Domain 2 (21%) as the centre of gravity: most scenario questions hinge on what you would check and when you would escalate
- — Know the product surface: what Projects, Artifacts, research mode and connectors do, and what they don't
- — Read every option. Multiple-response items tell you how many to pick; the distractors are often 'almost right' practices
- — Think like the minimally qualified candidate in the guide: a capable, careful daily user, not an engineer
Official documentation to read
- Claude Certified Associate – Foundations Exam Guide (official PDF)The blueprint: seven domains, weights, policies and the minimally qualified candidate profile
- Anthropic Partner Academy — Associate preparation pathOfficial self-paced courses (available to Claude Partner Network members)
- Claude Help CenterFeature-level detail for Projects, Artifacts, connectors and model availability
Practice with AI7x7
Free practice exam, free guide, $19 full-length package
Practice percentages are not the official 720 scaled score and do not predict your result. They tell you which domains need work.
20-question practice exam
Timed 40-minute session with a domain breakdown and full explanations on completion. AI7x7 practice timing, not the official 120-minute duration.
Preparation guide (PDF)
A 12–20 page study guide: blueprint, domain-by-domain notes, worked examples, common mistakes and a 7-day and 14-day plan. Delivered by email.
5 full-length practice exams
5 distinct 60-question timed simulations matching the official format, explanations for every option, domain reports and score history. Unofficial AI7x7 simulation.
FAQ
CCAO-F questions, answered
Do I need to be a developer to take CCAO-F?
No. The official exam guide states that no software development or API experience is needed. The credential targets professionals who use Claude as a productivity tool — operations, marketing, project management, education, communications and similar roles.
How much does the Associate – Foundations exam cost?
The official exam fee is USD 99, paid to Anthropic via the Partner Academy. AI7x7 preparation materials are priced separately and are not an exam fee.
Who can register for the official exam?
Anthropic describes the certifications as available to members of the Claude Partner Network, with registration through the Anthropic Partner Academy. Membership of the network is free for organisations. AI7x7 practice resources are open to anyone.
Does this certification count toward Claude Partner Network tier status?
No. The Partner Academy catalogue states that the Associate certification does not count toward Partner Network tier eligibility. The Developer and Architect certifications do.
How many questions are on the exam and how long is it?
60 multiple-choice and multiple-response items in 120 minutes, with a passing scaled score of 720 on a 100–1,000 scale.
Is the exam content changing?
Yes. Pearson VUE lists updated exam content effective 7 December 2026, with a revised exam guide due 17 November 2026. If you test on or after 7 December, prepare from the updated guide. AI7x7 will re-verify this page when the new guide is published.
Related certifications
Other Claude tracks
Claude Certified Developer – Foundations CCDV-F
If you also build against the API or with Claude Code, the Developer track is the natural next step
Claude Certified Architect – Foundations CCAR-F
For those moving from using Claude to designing agentic systems with it
Compare all four certifications side by side →Official certification page: https://anthropic-partners.skilljar.com/claude-certified-associate-foundations-certification · Exam guide: PDF (1.0 (effective July 2026))
AI7x7 is an independent exam preparation platform and is not affiliated with, sponsored by, or endorsed by Anthropic. Claude and related trademarks belong to their respective owners. AI7x7 practice questions are original and are not official exam questions.