CLF-C02 Exam Guide and Cloud Concepts

Every AWS certification begins with one document: the official exam guide. For CLF-C02 that single document does two jobs at once; it tells you exactly how the exam is built (codes, minutes, questions, cut score, weights) and it defines the cloud vocabulary you must command in Domain 1 (Cloud Concepts, 24% of scored content). Candidates who skip the guide and jump straight to flashcards fall into the same three traps: they misjudge the time budget, they confuse a scaled score with a percentage, and they cannot separate what is in scope from what merely sounds like real cloud work.

In this lesson you will:

• read the exam guide the way an exam writer does: code, timing, cost, scoring, languages;
• turn domain weights into a concrete study-hour budget;
• map all 19 task statements across the four content domains;
• master the two question types and the compensatory scoring model;
• separate the in-scope service list from the out-of-scope list that feeds distractors;
• define cloud computing, and contrast on-premises versus cloud in the guide's own terms;
• recognise the three deployment models (cloud, hybrid, on-premises) and the service-model ladder;
• explain elasticity, scalability, high availability and agility without mixing them up;
• state the six advantages of cloud computing and the AWS value proposition;
• apply the Well-Architected Framework pillars and its six general design principles;
• work through cloud economics: fixed versus variable cost, rightsizing, BYOL, automation;
• study four real AWS customer case studies: services, numbers and sources;
• read the October 2026 update box; which guide facts moved, and which did not;
• practise with 12 exam-style questions plus three interactive checks.

---

1. What the CLF-C02 exam measures

1.1 The target candidate

CLF-C02 is a Foundational credential, and AWS's own material describes its audience as non-IT entrants, sales, product and project-management people, who need cloud literacy, not cloud engineering. Nobody expects you to build infrastructure. You are tested on judgment, vocabulary, service selection and cost awareness.

1.2 The out-of-scope job tasks: verbatim

The guide draws a hard line, and it is worth memorising word for word:

"The following list contains job tasks that the target candidate is not expected to be able to perform. This list is non-exhaustive. These tasks are out of scope for the exam: Coding / Designing cloud architecture / Troubleshooting / Implementation / Load and performance testing"

This list is not decoration. Items from it appear as distractors, because a distractor that sounds like "real cloud work" is the most tempting wrong answer of all.

1.3 Appendix A: the technologies and concepts list

Alongside the domains, the guide carries an appendix of 28 technologies and concepts that "might appear on the exam", explicitly with no indication of their relative weight. The Domain 1 entries are cloud migration benefits, AWS Cloud Adoption Framework (CAF), cost management, global infrastructure, infrastructure as code, migration and data transfer, AWS Prescriptive Guidance, AWS Pricing Calculator, AWS Professional Services, AWS Partner Network, solutions architects and the Well-Architected Framework. Read it as a vocabulary checklist, never as a weight.

• 📚 Did you know? The exam has two official guide artefacts that genuinely disagree: an HTML guide on docs.aws.amazon.com and an older PDF stamped "Version 1.0 CLF-C02" on d1.awsstatic.com. 13 of the 19 task statements differ between them (checked Oct 2026). The rule AWS's own structure implies: the HTML guide wins over any old PDF, blog post or third-party practice test.

---

2. Exam logistics: the numbers that change your strategy

2.1 The logistics table

Every row below comes from the exam guide, the certification page's "Exam overview", or the AWS Certification FAQs, all accessed October 2026.

2.2 Worked example E1: the per-question time budget

The single most useful number in the whole guide is arithmetic that nobody publishes:

$$t = \frac{90 \times 60}{65} = \frac{5{,}400}{65} \approx 83.1 \text{ seconds per question}$$

Reserve 10 minutes for review and the real budget becomes $80 \times 60 / 65 \approx$ 73.8 seconds per item. Budgeting rule of thumb: answer fast early, then spend the surplus on the slower multiple-response items.

2.3 Worked example E2: scored versus unscored

$$65 = 50 + 15 \qquad \frac{15}{65} = 23.1\% \text{ of the exam does not affect your score}$$

Nearly a quarter of the paper is invisible pilot material, and you cannot tell which 15 they are. The only sane strategy is to treat every single item as scored.

2.4 Worked example E3: the cost of a retake

Three attempts cost 300 USD as of Oct 2026 plus at least 28 calendar days of waiting between them. A missed appointment is different again: it forfeits the fee but is not a failed attempt.

• 📚 Did you know? Validity of 3 years does not mean a mandatory retake. AWS lists three renewal routes, and one of them, Cloud Quest Recertify Cloud Practitioner, needs no exam at all; alternatively, passing any Associate- or Professional-level exam renews it. So a candidate who plans to continue on the AWS path can let CLF-C02 expire deliberately.

---

3. The four domains and their weights

3.1 Official weights

The guide states four content domains, and the percentages are shares of scored content only:

3.2 Worked example E4: derived item counts (arithmetic, not AWS's)

Multiply the 50 scored items by each weight:

3.3 From CLF-C01 to CLF-C02

The exam was updated in September 2023 (CLF-C01 ran until 18 September 2023; CLF-C02 from 19 September 2023), and the guide's own comparison appendix shows how the emphasis moved:

Three facts follow from that table: security grew by five points; billing shrank by four; and AWS states "No content was deleted from the exam": content from seven old task statements was recategorised into the new ones. The one genuine addition is Task 1.3, which introduced the AWS Cloud Adoption Framework (CAF).

3.4 Worked example E5: turning weights into study hours

Suppose you have 40 hours before exam day. The weights convert directly into hours:

Domains 2 + 3 are 64% of the score. A candidate who spends all their time on cloud concepts because "that's the Cloud Practitioner topic" has misallocated two-thirds of the paper.

---

4. The task-statement map: 4 domains → 19 tasks

The guide does not publish topics; it publishes task statements: each one a "Knowledge of" plus a "Skills in" pair. Everything examinable hangs off one of these 19.

4.1 Domain 1 in detail: the four tasks this course starts with

4.2 The other 15 tasks in one line each

---

5. Question types and scoring mechanics

5.1 Exactly two question types

There are no ordering, matching, drag-and-drop or case-study items in the CLF-C02 guide: if a practice test shows you those, it is not following this exam's guide.

5.2 The scoring pipeline

Compensatory means one thing: only the overall result must clear 700. You do not need to reach 700 inside each domain: a weak Domain 4 can be carried by a strong Domain 3, and dominating one domain cannot rescue an overall score below the cut.

5.3 Worked example E6: 700 is not 70%

700 out of 1,000 looks like 70%, and the naive conversion $0.70 \times 50 = 35$ correct answers is arithmetically clean, and wrong as an exam fact. The scale starts at 100, so the usable range is $1{,}000 - 100 = 900$ points:

$$\frac{700 - 100}{900} = \frac{600}{900} \approx 66.7\%$$

AWS scales scores across multiple exam forms that may differ in difficulty, so the raw number of correct answers required for 700 is not fixed and is not published. The only deterministic rule is behavioural: unanswered items are scored incorrect, so answer all 65.

---

6. In-scope versus out-of-scope services

6.1 Two lists, one rule

The guide publishes an In-Scope AWS Services list spanning 19 categories, and a separate Out-of-Scope AWS Services list. Both are labelled non-exhaustive and subject to change (as of Oct 2026). The in-scope list is what you must recognise; the out-of-scope list tells you which names are being used as baits.

6.2 The out-of-scope list: services that exist but are not examinable

• Amazon AppFlow
• AWS Clean Rooms
• AWS Data Exchange
• Amazon DataZone
• Amazon MSK
• AWS AppFabric
• Amazon WorkDocs
• AWS Wavelength
• AWS Copilot
• Amazon GameLift
• AWS CloudShell
• AWS CodeDeploy
• AWS CodeArtifact
• AWS Device Farm
• Amazon CodeGuru
• AWS IoT Greengrass
• Amazon Fraud Detector
• Amazon Personalize
• Amazon Lookout for Metrics
• AWS Chatbot
• AWS Billing Conductor
• all AWS Elemental media services
• Amazon MemoryDB
• AWS Network Firewall
• AWS RoboMaker
• Amazon FSx for Lustre
• AWS Transfer Family
• AWS IQ
• AWS Activate
• AWS Managed Services (AMS)
• Amazon Keyspaces
• AWS Launch Wizard
• AWS AppConfig
• Amazon Data Lifecycle Manager
• Amazon Elastic Transcoder
• AWS Cloud Map
• AWS Ground Station
• Amazon VPC Lattice
• Amazon Cloud Directory
• AWS Migration Hub Refactor Spaces
• AWS Network Access Analyzer
• AWS IoT Device Defender
• Amazon Monitron
• AWS Application Composer
• Amazon Interactive Video Service (Amazon IVS)

6.3 Naming traps in the in-scope list

• The guide spells it "Amazon Quick Sight" (two words): the product page says QuickSight.
• Task 4.2 says "AWS Cost and Usage Report" while the service list says "AWS Cost and Usage Reports" (plural).
• It is "AWS IAM Identity Center": the old "(AWS Single Sign-On)" wording no longer appears.
• AWS VPN, AWS Site-to-Site VPN and AWS Client VPN are listed as three separate bullets, not one service with three modes.
• The service is Amazon SageMaker AI: the older "SageMaker" name without "AI" is what older practice tests use.

6.4 Which guide wins

The HTML version dropped items such as "AWS Snowball" and "AWS Audit Manager" from task statements, and changed the service lists substantially: the out-of-scope list went from 11 names in 3 categories in the PDF to a far longer list in the HTML.

• 📚 Did you know? AWS Snowball is a trap in two directions at once: the older PDF exam guide still names it inside Task 1.3, yet it appears on neither the current in-scope list nor the out-of-scope list, and the Snowball Edge documentation carries a banner stating the device is no longer available to new customers (as of Oct 2026). Both service lists are non-exhaustive, so AWS never reconciled the contradiction; do not build your study plan around it.

2026 Updates (as of October 2026)

• 📚 Did you know? The current exam guide PDF carries no version number at all; it prints only "Copyright © 2026", and the single stamped artefact still circulating is the launch-era "Version 1.0 CLF-C02". The uiVersion=2024.10 you can see in the docs URL is the AWS Docs site-template version, not a guide revision date (checked Oct 2026). There is also no AWS changelog page for this guide, so "the guide was refreshed in March 2026" style claims are unverifiable.

---

7. Cloud basics: what the cloud is, and what changes when you move

7.1 What cloud computing is

On this exam, cloud computing is defined by its consequence rather than by hardware: you stop owning and maintaining infrastructure and instead consume a metered service, paying only while it runs. AWS's own advantages page carries the whole economic argument in one line:

"Trade fixed expense for variable expense … pay only when you consume computing resources"

AWS describes the platform as offering over 200 fully featured services (as of Oct 2026), and its origins page states the founding intent was to let anyone "access the same powerful technology as the world's largest and most sophisticated companies". The six advantages in Section 8 are what Task 1.1 means by the "value proposition of the AWS Cloud".

7.2 On-premises versus cloud

The four on-premises figures above are AWS-published industry estimates (2022 and 2024 posts), current as of Oct 2026: verify current before use.

7.3 Worked example E7: the on-premises costs a price list never shows

A hypothetical small deployment of 10 servers over a five-year horizon. The dollar inputs are illustrative assumptions, not AWS prices (as of Oct 2026: verify current before use); the point of the example is which lines are normally forgotten:

The lesson is not the total; it is that four of those five lines are invisible to anyone comparing a server sticker price with an hourly rate. AWS's guidance is explicit that flawed comparisons omit labour, refresh and facilities.

7.4 Deployment models

Task 3.1 names exactly three deployment models:

A hybrid model is the answer whenever part of the estate must stay on premises while new capability is built on AWS; on-premises is the answer when nothing moves; cloud is the answer when everything moves.

7.5 The service-model ladder: how much AWS takes over

The guide does not enumerate IaaS/PaaS/SaaS in a task statement, but it does test the underlying idea through the shared responsibility shift: the more managed the service, the smaller your share.

Constants regardless of ladder position: physical security of the cloud is always AWS; data, IAM and encryption decisions are always the customer.

7.6 Four words the exam keeps apart

---

8. AWS value proposition: the six advantages

8.1 The six advantages, in AWS's wording

Task 1.1's "value proposition" is examined through six advantages, and questions frequently quote a phrase verbatim:

Task 1.1 adds the four headline benefits: benefits of global infrastructure (speed of deployment, global reach) and the advantages of high availability, elasticity and agility.

8.2 Worked example E8: why economies of scale are structural, not a promotion

Two customers each need a peak of 10 units of capacity, but their peaks occur at different times (anti-correlated):

Because AWS aggregates demand, it buys and runs ~40% less hardware for the same total customer peak, and passes the structural saving on as a lower variable rate. That is what "massive economies of scale" means; it is not a limited-time discount.

8.3 Worked example E9: twenty years of the same price unit

AWS's own 20-year S3 retrospective gives both ends of the price curve: Amazon S3 launched on 14 March 2006 at 15 cents per gigabyte, and AWS's March 2026 post puts the current figure at slightly over 2 cents per gigabyte: roughly an 85% reduction (as of Oct 2026; verify current pricing before use). On 100 TB ($102{,}400$ GB):

$$\text{2006: } 102{,}400 \times 0.15 \approx \$15{,}360/\text{mo} \qquad \text{2026: } 102{,}400 \times 0.02 \approx \$2{,}048/\text{mo}$$

Same unit, same service, same API: the price fell because scale and efficiency compound. The blog also notes the code written for S3 in 2006 still works unchanged, which is a durability argument as much as a price argument.

8.4 The Well-Architected Framework: Task 1.2

The framework is AWS's "best practices for designing and operating secure, reliable, efficient, cost-effective, and sustainable workloads". A review is explicitly "a constructive conversation … not an audit mechanism", and the AWS Well-Architected Tool that performs it is free and on the in-scope list.

The framework's six general design principles all do the same underlying trick; they convert fixed, upfront, idle spending into metered, on-demand spending: stop guessing your capacity needs; test systems at production scale; automate with architectural experimentation in mind; consider evolutionary architectures; drive architectures using data; improve through game days.

• 📚 Did you know? Three different "six/five lists" get mixed up on this exam, and they are not variants of one another: the Well-Architected Framework's 6 pillars (workload quality), the AWS Cloud Adoption Framework's 6 perspectives (organisational readiness: Business, People, Governance, Platform, Security, Operations) and the Cloud Value Framework's 5 pillars (business value, of which cost savings is only pillar 1). "Sustainability" and "Cost" appear in more than one list with different meanings: read the stem to learn which list is live.

---

9. Cloud economics: fixed, variable, and where the money actually goes

9.1 The core distinction

• Fixed cost exists whether or not the service runs: owned hardware, floor space, salaried staff.
• Variable cost tracks consumption: metered cloud resources.
• Capex (capital expenditure) buys an asset you own and depreciate; opex (operational expenditure) pays for what you use. The cloud turns capex into opex; that is advantage #1 restated as an accounting fact.

9.2 On-premises cost drivers you must be able to name

• Hardware + annual maintenance
• operating-system and application licences
• space, power and cooling
• IT labour
• the ~5-year refresh cycle
• 6 to 9 month procurement
• 20 to 50% over-provisioning
• cost of capital
• bandwidth

A comparison that counts only server purchase price against hourly rates is the flawed model AWS warns about.

9.3 Worked example E10: rightsizing

A server is provisioned at 12 vCPU / 64 GB but measures 25% CPU and 25% memory utilisation over a representative month:

Rightsizing is not "pick the cheapest instance." It is matching size and type to measured utilisation while still meeting performance, and the order matters: right-size first, then choose a commitment or discount model.

9.4 Licensing: BYOL versus included

AWS License Manager is the in-scope tool that tracks licences across AWS, on-premises and other clouds.

9.5 Automation and managed services

• Automation / infrastructure as code: templates are versioned, reviewable, repeatable and rollback-able. The guide's named automation example is AWS CloudFormation (the PDF rendering of Task 1.4 names it explicitly; the HTML guide says only "benefits of automation": accept both framings). Note the distinction: automation ≠ auto scaling. CloudFormation makes provisioning repeatable; Auto Scaling provides elasticity. Both are in scope, and they answer different questions.

• Managed services absorb undifferentiated work: Amazon RDS takes engine patching, backups and failover; ECS and EKS take the platform; DynamoDB takes capacity and operation. RDS, ECS, EKS and DynamoDB are the named managed-service examples (again, named in the PDF version of Task 1.4).

• 📚 Did you know? AWS deliberately frames cost as only pillar 1 of 5 in its Cloud Value Framework: total cost of ownership is the first pillar, not the whole argument, and AWS pairs it with third-party research claiming 50% lower five-year cost of operations and a 10-month payback (IDC for AWS, 2022, as cited by AWS, as of Oct 2026). Those are research findings across a customer base, never a guarantee to any individual customer.

---

Real-World Case Studies

AWS publishes what these abstractions look like in production. Every figure below is customer- or AWS-claimed and unaudited, with the source named so you can check it: the examinable point is the pattern (which principle was applied, which services did the work, which number moved), not the marketing.

Case A, Capital One: exiting the data centre (migration and agility)

"We are truly all in on the cloud, and AWS has been instrumental in enabling us to take full advantage of the benefits of being in the cloud.": Chris Nims, SVP cloud and productivity engineering, Capital One

Case B, Box: Well-Architected cost optimization (cost pillar in practice)

"Our use of AWS best practices led to savings of over 2 million dollars, setting a new baseline…": Clay Alvord, director of FinOps and SRE, Box

Note the pattern: none of Box's four levers was a negotiation with AWS. Three are architectural (where traffic goes, which storage class, how much is logged) and one is a review process; which is precisely why the Well-Architected Cost Optimization pillar, not a discount code, is the Domain 4 concept.

Case C, Philip Morris International: 400 applications in two years (migration value)

The examinable point is not the count of applications; it is that a regulated company bought the migration window by disposing of the hardware first, and that AWS's own text records the initial aggressive lift-and-shift attempt "yielded suboptimal results" before the programme leaned on Professional Services, Well-Architected reviews and automation. Every figure above is customer/AWS-claimed and unaudited (as of Oct 2026).

Case D, Canva: a portfolio of pricing models (cloud economics in practice)

Read Canva as the counter-example to "find the one cheapest option": the answer the exam wants is a mix matched to workload shape, interruptible capacity for interruptible work, commitment for steady work, On-Demand for whatever is left, which is exactly the ordering rightsizing → commitment → Spot logic from Section 9.

• 📚 Did you know? AWS's own Philip Morris case study is unusually candid: it states that the company's initial aggressive lift-and-shift "yielded suboptimal results": a published AWS admission that migration strategy beats migration speed. That is why Task 1.3 asks you to identify migration strategies (for example, database replication), not merely to name a rehosting tool, and why "just move it all, fast" is a classic distractor (source accessed Oct 2026).

---

Practice Questions