ISC2 Certified in Cybersecurity (CC) Cheat Sheet: Key Facts
Every key takeaway and common trap from the ISC2 Certified in Cybersecurity (CC) study guide on one page, grouped by chapter. Tap a lesson link to read the full explanation.
65 facts · 10 chaptersFacts checked 2026-11-15
Chapter 1
Security Principles & Control Types
Key takeaways
- Confidentiality is about disclosure, integrity is about unauthorized change (including poisoned AI training data), and availability is about timely, reliable access for authorized users. → The CIA triad: confidentiality, integrity and availability
- The access sequence is always identification, authentication, authorization, then accounting, and a successful login is never by itself permission to use a resource. → Identification, authentication, authorization and accounting
- Multifactor authentication needs factors from different types, such as something you know plus something you have, not two items of the same type. → Authentication factors and true multifactor
- Classify a control twice: by category (technical, administrative, physical) and by function (preventive, detective, deterrent, corrective, recovery, compensating, directive). → Classifying controls by category and by function
- Insurance transfers financial impact but never accountability, guidelines are the only advisory document type, and due diligence is finding out while due care is acting on it. → Risk transfer, guidelines and due care in this domain
Common trap
- Distractors often swap sensitivity and criticality, or call a leak an integrity problem because the data left the system that controlled it. Leaving a system is disclosure, not modification. → The CIA triad: confidentiality, integrity and availability
- Watch for answers that say a phone prompt is a form of something you know, or that multifactor requires a biometric. Neither is true: any two different factor types qualify. → Authentication factors and true multifactor
Quick reference
| Scene | Pillar harmed |
|---|---|
| Customer list left on a public file share, unchanged | Confidentiality |
| Attacker alters the data an AI model is retrained on | Integrity |
| Storm cuts power, data intact but unreachable for a day | Availability |
| Login | Factor types | Multifactor? |
|---|---|---|
| Password plus approve a prompt on enrolled phone | Know plus have | Yes |
| Fingerprint plus code from a key fob | Are plus have | Yes |
| Password plus a security question | Know plus know | No |
Chapter 2
Risk, Policy Hierarchy & Ethics
Key takeaways
- A vulnerability is the weakness, a threat is what could exploit it, and risk exists only where a threat meets a matching vulnerability, judged by likelihood and impact. → Threats, vulnerabilities, likelihood and impact
- The four treatments are avoid, mitigate, transfer and accept, and residual risk must be formally accepted by senior management, not by the analyst. → The risk management lifecycle and the four treatments
- Laws and regulations come from governments, PCI DSS binds through contracts, and frameworks such as the NIST CSF are voluntary for private companies unless something imports them. → Laws, frameworks and the policy document hierarchy
- Policy says what and why, a standard names the specific measurable requirement, a procedure gives the steps, and only a guideline is advisory. → Laws, frameworks and the policy document hierarchy
- Due diligence is investigating and verifying, due care is acting as a prudent person would, and every ISC2 member must commit to the four-canon Code of Ethics. → Due care, due diligence and the ISC2 Code of Ethics
Common trap
- Missing patches are the vulnerability, not the threat and not the risk. The attacker is the threat, and the breach that could follow is the risk. → Threats, vulnerabilities, likelihood and impact
- Ignoring a flaw is not acceptance. Acceptance needs a weighed decision, a named owner and a record. Also, insurance transfers money, not accountability. → The risk management lifecycle and the four treatments
Quick reference
| Treatment | What happens |
|---|---|
| Avoid | Do not start, or stop, the activity that creates the exposure |
| Mitigate (reduce) | Keep the activity and add controls that lower likelihood, impact or both |
| Transfer (share) | Shift the financial impact to a third party, such as an insurer |
| Accept | A documented, approved decision to absorb a risk within tolerance |
Chapter 3
Governance, Risk & Compliance
Key takeaways
- Governance sets direction and accountability, risk management treats threats to the mission, and compliance shows that legal, regulatory and contractual obligations are met. → What GRC is for, and who is accountable
- Being compliant is a point-in-time minimum, not proof of being secure. → What GRC is for, and who is accountable
- Responsibility for the work can be delegated, but accountability for risk stays with senior leadership and the accountable business owner. → What GRC is for, and who is accountable
- ISO/IEC 27001 is the certifiable management-system standard, and the NIST CSF Govern function holds risk strategy, roles and policy. → Frameworks, standards and where obligations come from
- A KPI reports performance against a target, a KRI warns of rising exposure against a threshold, and every metric needs a baseline, a trend and the right audience. → Key metrics, KPIs and key risk indicators
Common trap
- The audience does not decide whether a figure is a KPI or a KRI. Ask whether it reports past performance against a target or warns of future exposure against a threshold. → Key metrics, KPIs and key risk indicators
Quick reference
| Vehicle | What it is | Typical reader |
|---|---|---|
| Dashboard | Visual, near-real-time monitoring | Operations team during a shift |
| Scorecard | Periodic rating of performance against targets | Managers reviewing control areas |
| Report | Periodic written narrative with detail and context | Leadership needing the why behind the numbers |
Chapter 4
Continuity, Disaster Recovery & Awareness
Key takeaways
- Redundancy removes single points of failure, business continuity keeps critical functions running during a disruption, and disaster recovery restores the technology. → Redundancy, business continuity and disaster recovery
- The business impact analysis comes first and produces the recovery objectives: RPO caps data loss, RTO caps restore time, and restore plus work recovery must fit inside the maximum tolerable downtime. → Business impact analysis, RPO, RTO and alternate sites
- A differential restore needs the full backup plus the latest differential, while an incremental restore needs the full backup plus every incremental since. → Backup types, offline copies and test restores
- Keep at least one backup copy offline or immutable, and treat a backup as real only after a test restore succeeds. → Backup types, offline copies and test restores
- Awareness is short recurring reinforcement for everyone, and because AI now writes fluent lures and clones voices, staff should verify unexpected requests through a channel they already know. → Security awareness, culture and AI-powered phishing
Common trap
- RPO and RTO are swapped in many distractors. RPO is about data, looking back; RTO is about time to restore, looking forward. Also, cost never overrides a recovery objective the business already approved. → Business impact analysis, RPO, RTO and alternate sites
Chapter 5
Identity Lifecycle Management
Key takeaways
- Permissions belong to roles that describe job functions, and people get access by being assigned to a role. → Roles definition and provisioning new accounts
- Provisioning runs in the order verify identity, create the account, assign the role, then issue credentials, ideally triggered from the HR record. → Roles definition and provisioning new accounts
- A mover must lose the old role's access when the new access is added, or privilege creep builds up. → Access reviews, movers and privilege creep
- On departure, disable first and delete later under the retention schedule, and revoke tokens, keys and badges too. → Deprovisioning leavers and non-human identities
- Service accounts, bots and AI agents follow the same life cycle, with a named human owner and removal when the workload ends. → Deprovisioning leavers and non-human identities
Common trap
- Multifactor authentication, shorter password expiry and session timeouts do not fix privilege creep. They check who signs in, not what that person is allowed to do once inside. Only review and removal of the excess entitlements fixes it. → Access reviews, movers and privilege creep
- Do not confuse SAML with SCIM. A user can sign in through a federated SAML login and still have no account in the target app; creating and removing that account is SCIM's job. → Identity frameworks and tools: SSO, federation, SCIM, PAM and IGA
Quick reference
| Departure | When access is cut |
|---|---|
| Amicable resignation with notice | Limited access can taper during handover, ending in full deprovisioning |
| Dismissal or hostile termination | At or just before notification, planned in advance by HR and security |
Chapter 6
Logical Access Controls & Models
Key takeaways
- Least privilege limits both how broad a permission set is and how long it is held, and it applies to bots and AI agents as well as people. → Least privilege and need-to-know
- Need-to-know can separate two people with identical privileges, because it asks whether a specific item is needed for the current duty. → Least privilege and need-to-know
- Separation of duties stops one person from both committing and hiding a fraud; its leftover risk is collusion. → Separation of duties, dual control and job rotation
- Mandatory vacation and job rotation are detective controls that let a stand-in notice what the regular holder was hiding. → Separation of duties, dual control and job rotation
- Owner decides means DAC, labels and clearances mean MAC, job function means RBAC, conditions for everyone mean rule-based, and many attributes mean ABAC. → MAC, RBAC, rule-based and attribute-based access control
Common trap
- If a restriction applies to every user, even executives, it is rule-based, not role-based. And a clearance is not a role: it expresses a trust level for classified material, which points to MAC. → MAC, RBAC, rule-based and attribute-based access control
Quick reference
| Control | How it works |
|---|---|
| Separation of duties | Different people do different steps of one process |
| Dual control | Two people act together on the very same step, like two keys for a vault |
| Mandatory vacation, job rotation | Detective: a stand-in may notice what the regular holder hid |
Chapter 7
Network Security Concepts
Key takeaways
- Layer 2 moves frames by MAC address, Layer 3 routes packets by IP address, and Layer 4 tracks conversations with ports. → The OSI and TCP/IP models
- The private IPv4 blocks are 10.0.0.0/8, 172.16.0.0/12 and 192.168.0.0/16, and NAT saves addresses but enforces no policy. → IPv4 private addresses, NAT and IPv6
- A VPN protects data in transit, not an infected endpoint, and split tunneling sends some traffic around corporate inspection. → VPNs: what they protect and what they do not
- A firewall rule base should end in default deny, and application awareness is needed when many programs share one port. → Firewalls, ports and applications
- Hidden SSIDs and MAC filtering are obscurity, WPA3 SAE resists offline guessing, and OT puts safety and availability first. → Wireless, Bluetooth, ICS and IoT devices
Common trap
- Ping has no port. ICMP is carried directly by IP at Layer 3 and uses message types and codes, so a firewall rule permits ICMP echo request and echo reply rather than opening a port number. → The OSI and TCP/IP models
Quick reference
| OSI layer | Job and data unit | TCP/IP layer |
|---|---|---|
| 1 Physical, 2 Data Link | Bits on the wire; frames by MAC address on one segment | Network Access (Link) |
| 3 Network | Packets by IP address; routing between networks | Internet |
| 4 Transport | Segments; TCP and UDP ports identify conversations | Transport |
| 5 Session, 6 Presentation, 7 Application | Services such as web and mail | Application |
Chapter 8
Network Architecture & Cloud Security
Key takeaways
- Segmentation limits how far an intruder can move after compromising one host, while a flat network lets one infection reach everything. → Network segmentation, firewall zones and VLANs
- Defense in depth means independent layers of different control types, administrative, technical and physical, not two copies of the same device. → Defense in depth and zero trust
- Zero trust gives no trust for being inside the network and checks every request for every resource against policy. → Defense in depth and zero trust
- NIST SP 800-145 names five essential cloud characteristics, three service models and four deployment models, and a private cloud is defined by single-organization use, not by who owns the hardware. → Cloud characteristics and deployment models
- In every service model a cloud provider offers, the customer keeps its data, identities and configuration choices, the provider secures the facility, hardware and virtualization layer, and accountability for the data never transfers. → Service models and the shared security model
Common trap
- Do not treat zero trust and micro-segmentation as the same thing. Micro-segmentation is a network technique that adds internal perimeters. Zero trust is broader: it centers protection on identities, assets and data and verifies each request. The two are often deployed together. → Defense in depth and zero trust
- Watch for answers saying a breach in the cloud is entirely the provider's problem, or that a private cloud must be self-hosted. Both are wrong: accountability stays with the customer, and private means single-organization use. → Service models and the shared security model
Quick reference
| Characteristic | Clue in the scenario |
|---|---|
| On-demand self-service | Capacity added through a portal with no human contact at the provider |
| Broad network access | Reached over the network from laptops, tablets and phones |
| Resource pooling | Many tenants share hosts, and capacity is reassigned by demand |
| Rapid elasticity | Capacity grows and shrinks to match load |
| Measured service | Use is metered and reported |
Chapter 9
Data Security & Security Operations
Key takeaways
- Classification is the decision about sensitivity, labeling is the marking that communicates it, and every other handling rule follows from the classification. → Data classification, labeling, data states and masking
- Deleting files leaves recoverable data, degaussing does not work on SSDs, and reusable flash media are typically purged with a cryptographic erase or the vendor's sanitize command. → Retention and media sanitization
- Encrypt with the recipient's public key for confidentiality, sign with your own private key for integrity, authentication and non-repudiation, and hash passwords with a salt. → Symmetric and asymmetric encryption, hashing and signatures
- Logs are a detective control that only work for timelines when clocks are synchronized and copies are sent to a central, write-once store. → Logging, monitoring and event triage
- An event is anything observable, an incident harms confidentiality, integrity or availability, and real alerts are prioritized by business impact, data sensitivity and recoverability. → Logging, monitoring and event triage
Common trap
- Degaussing is a common wrong answer for solid-state drives. It works only on magnetic media; SSDs store data in flash cells. If the SSDs will be reused, destruction is out too, so the usual Purge choice is a cryptographic erase or the manufacturer's sanitize command. Degaussing is also not an approved Destroy technique under SP 800-88 Rev. 2. → Retention and media sanitization
- Wrong answers often pair a script kiddie with a long espionage campaign, or a nation-state with quick cash. Match the actor to its motive and resources, and remember that threat intelligence informs decisions; it does not prove attribution for a court case. → Threat actors, threat intelligence and threat frameworks
Quick reference
| Goal | What to use |
|---|---|
| Only the recipient can read it | Encrypt with the recipient's public key |
| Prove who sent it and that it was not changed | Digital signature: hash signed with the sender's private key |
| Detect any change to a file | Hash: one-way, fixed length |
| Store passwords | Salted hash from a password key-derivation function |
| Actor | Main motivation |
|---|---|
| Nation-state | Espionage and strategic advantage for a government; well funded and patient |
| Organized crime | Profit |
| Hacktivist | An ideological or political message |
| Script kiddie | Curiosity or notoriety, using other people's tools |
Chapter 10
Incident Response, Assets & Testing
Key takeaways
- The incident response plan is written, approved and shared before anything goes wrong, and it names roles, authority, classification criteria, escalation paths and contacts. → The incident response plan, team and exercises
- In the NIST SP 800-61 Rev. 2 life cycle, containment comes before eradication, and recovery comes after the threat is removed. → The incident response life cycle
- A complete asset inventory comes first, because controls can only be applied to assets the organization knows it has. → Asset lifecycle, configuration and change management
- Configuration management governs the approved state of a system, and change management governs how that state is allowed to change. → Asset lifecycle, configuration and change management
- A vulnerability scan reports known weaknesses, a penetration test exploits them to show impact, and no test starts without written authorization. → Security testing: teams, scans, application tests and physical tests
Common trap
- Reformatting or reimaging an infected machine right away sounds decisive, but it is eradication too early. Isolate first: containment limits the spread and preserves the evidence needed to learn the scope and whether data was taken. → The incident response life cycle
Quick reference
| Method | What it does |
|---|---|
| Vulnerability scan | Automated and broad; reports known weaknesses that may exist, which must be checked for false positives |
| Penetration test | Expert-led; exploits and chains weaknesses to show real impact |
| Static analysis | Reads code that is not running; finds source-level defects |
| Dynamic analysis | Drives the running app; finds server misconfiguration, session faults and exposed test paths |
| Threat modeling | Design-time review of likely threats, such as the STRIDE categories, with a countermeasure for each |