CISM domain - 20% of the exam

Information Security Risk Management

Information Security Risk Management is 20% of the CISM exam. These are the objectives it covers, each with practice questions, with every answer explained.

The domain in numbers

59
Practice questions
6
Objectives
40
Medium
19
Hard

Where people go wrong

  • Moving risk accountability to the provider

    The wrong answers treat outsourcing, delegation or a contract clause as moving ownership of residual risk to the provider: recording the provider as risk owner, sharing ownership, or relying on an indemnity. Outsourcing transfers the operation of controls and some financial liability, not accountability. The organisation's own business leader stays risk owner for its data, the provider is a control owner under the contract, and regulatory consequences remain with the organisation.

    See it in a sample question
  • Treating insurance as mitigation or a full transfer

    The wrong answers treat cyber insurance as mitigation, or as a way to close out a risk. A policy pays for some financial loss after an event, so it is risk transfer, not mitigation, and it does nothing to reduce the likelihood of a breach, fraud or production halt. It also cannot transfer reputational harm or regulatory accountability. Insurance complements mitigation and cannot by itself bring an exposure within risk appetite.

    See it in a sample question
  • Reading ordinal ratings and guessed inputs as precise

    The wrong answers credit risk scoring with more precision than its inputs hold. Likelihood and impact ratings on a five-point scale are ordinal labels, so multiplying them assumes equal gaps between points and the ranking can mislead; a shared scale or a ten-point scale does not change that. A quantitative annual loss expectancy given to the nearest pound has the same problem when there is no incident history or loss data: guessed inputs give false precision, and rounding hides it. A ranged or qualitative estimate is more honest.

    See it in a sample question
  • Treating compensating controls as a fix

    The wrong answers misjudge what surrounding controls do to a missing one: dismissing the finding as a false positive, calling the other measures redundant, or closing the deficiency as remediated because a proxy and a web application firewall sit in front. Segmentation, access control and signature blocking are compensating controls. They reduce likelihood or impact, so residual risk is lower, but the missing encryption or input validation stays open and is recorded alongside them.

    See it in a sample question
  • Ranking remediation by CVSS base score alone

    The wrong answers rank remediation by a generic technical score: sorting 4,200 scan findings by CVSS base score, or ordering by age or ease of fix. The base score ignores asset value, exposure and compensating controls already in place. Prioritisation starts by correlating each vulnerability with the criticality and exposure of the affected asset, which turns a raw scan into a risk-ranked order.

    See it in a sample question

Objectives in this domain

What to study

The official documents this domain's practice questions cite most.

Sample questions from this domain

Free sampleInformation Security Risk Managementmedium

A logistics firm migrates its customer billing platform to a cloud provider that now operates the technical security controls under contract. A risk analyst proposes recording the provider as the owner of the residual confidentiality risk because the provider runs the controls. How should the information security manager respond on the question of residual risk ownership?

  • ARecord the cloud provider as the risk owner, since the provider now operates the controls and is best placed to keep the residual risk at the agreed level
  • BRetain a senior internal business leader as the risk owner, because accountability for residual risk to the firm's own data cannot be transferred by outsourcing Correct
  • CSplit ownership equally between the firm and the provider, so both parties share accountability for any residual confidentiality loss in proportion
  • DLeave the residual risk unowned until the contract's assurance reports confirm the provider's controls are operating as the agreement requires
Outsourcing transfers control operation but never accountability, so a senior internal leader must remain the owner of residual risk to the organisation's own data. A third party can be contracted to operate controls and may be held to service levels, but the organisation stays accountable to regulators and customers for its data, so residual risk ownership must remain internal.

Why A is wrong: Tempting because the provider runs the controls, but operating controls is not the same as owning the residual risk, which stays with the data's organisation.

Why B is correct: Correct because outsourcing transfers control operation but not accountability; the organisation remains answerable for its own data, so an internal leader must own the residual risk.

Why C is wrong: Tempting as it seems balanced, but shared ownership diffuses accountability so that no single party is answerable, defeating the purpose of assigning an owner.

Why D is wrong: Tempting because assurance evidence matters, but a risk must always have an accountable owner, and deferring assignment leaves the exposure unmanaged in the interim.

Free sampleInformation Security Risk Managementmedium

An information security manager is explaining the four risk treatment options to a new governance committee. A member asks which option is being applied when the organisation buys cyber insurance to cover the financial loss from a data breach it cannot fully prevent. Which treatment option does purchasing that insurance represent?

  • ARisk transfer, because the financial consequence of a loss is shifted to a third party while the organisation continues the activity. Correct
  • BRisk avoidance, because the organisation removes the threat by withdrawing entirely from the exposed activity or technology.
  • CRisk mitigation, because the policy reduces the likelihood that an attacker can successfully breach the organisation's systems.
  • DRisk acceptance, because the organisation knowingly retains the breach exposure rather than acting to address it at all.
Identify that purchasing insurance is risk transfer or sharing, shifting financial impact to a third party while the activity continues. Transfer (sharing) moves the financial consequence of a loss to another party, typically through insurance or a contractual clause, without reducing the likelihood of the event or letting the organisation walk away from the activity. It is distinct from mitigation, avoidance, and acceptance.

Why A is correct: Insurance is the classic example of transfer (also called sharing): the financial impact of a realised loss is passed to an insurer, yet the organisation keeps running the activity and retains accountability for the data.

Why B is wrong: Avoidance is tempting because insurance feels protective, but avoidance means ceasing the risky activity altogether; buying cover lets the activity continue, so this misnames the option being used.

Why C is wrong: Mitigation is tempting because insurance is bought to deal with the risk, but a policy pays out after a loss and does nothing to reduce the likelihood of the breach occurring, so it is not mitigation.

Why D is wrong: Acceptance is tempting because some exposure remains insured, but acceptance means taking no treatment action; buying a policy is an active treatment that shifts the loss elsewhere, not pure retention.

Free sampleInformation Security Risk Managementhard

A risk team rates each threat for likelihood and impact on a five-point scale and multiplies the two numbers to rank threats on a colour-coded matrix. An information security manager must warn the risk committee about a flaw specific to treating these ordinal ratings as numbers that can be multiplied. What is the most accurate caution to give?

  • AMultiplying the ratings is sound provided every assessor uses the same five-point scale, since a shared scale removes the subjectivity from the likelihood estimates.
  • BThe main concern is that a five-point scale is too coarse, and moving to a ten-point scale would make multiplying the likelihood and impact ratings mathematically valid.
  • CMultiplying ordinal ratings implies the gaps between scale points are equal and the scores are true quantities, which they are not, so the resulting rankings can mislead. Correct
  • DThe flaw is that colour coding the matrix hides the underlying numbers, so reverting to plain numeric scores would resolve the limitation the committee should worry about.
Multiplying ordinal risk-matrix ratings wrongly assumes equal intervals and true quantities, which can distort relative risk rankings. Ordinal scales rank categories without guaranteeing equal distances between points, so multiplying likelihood by impact treats labels as measured quantities they are not; the product can rank one risk above another even when the underlying severity does not justify it.

Why A is wrong: Tempting because consistency helps, but a shared scale does not make ordinal points equal-interval, so the arithmetic flaw and the subjectivity both remain.

Why B is wrong: Tempting as a fix, but adding points does not turn an ordinal scale into an interval one, so multiplication stays invalid regardless of granularity.

Why C is correct: Correct because ordinal labels carry order but not equal intervals, so arithmetic on them assumes measurement properties the scale lacks and can distort the ranking.

Why D is wrong: Tempting because presentation matters, but exposing the numbers does not cure multiplying ordinal ratings, which is the actual limitation here.

Free sampleInformation Security Risk Managementhard

A control deficiency analysis of a payment platform finds that the database lacks encryption at rest, which on its own would expose stored card data. However, the database resides on an isolated network segment with strict access control, file-level integrity monitoring and full backups. When characterising the residual risk from the missing encryption, how should the security manager treat these other measures?

  • AAs evidence that the encryption deficiency is a false positive, since the surrounding controls show the data is not genuinely exposed.
  • BAs redundant controls that duplicate the protection encryption would provide, allowing the encryption requirement to be formally retired.
  • CAs detective controls that will identify any breach after it occurs, which removes the need to remediate the missing encryption.
  • DAs compensating controls that reduce the likelihood or impact of the deficiency and therefore lower, but do not eliminate, the residual risk. Correct
Other controls that mitigate a deficiency without removing it are compensating controls that lower residual risk rather than eliminate the gap. When a required control is absent, surrounding measures that reduce the likelihood or impact of exploitation are compensating controls. They lower residual risk but the underlying deficiency persists, so the manager records reduced residual risk while still tracking the missing encryption for remediation or formal acceptance.

Why A is wrong: It is tempting to dismiss the finding given the strong surrounding controls, but the deficiency is real: the data remains unencrypted at rest, so the surrounding measures mitigate the risk without making the absent control a false positive.

Why B is wrong: Calling them redundant is tempting because they protect the same data, but access control and segmentation defend confidentiality by a different mechanism than encryption at rest, so they reduce risk without making the requirement obsolete.

Why C is wrong: Integrity monitoring is indeed detective, but detection after the fact does not prevent disclosure of unencrypted data, so treating detection as a substitute for remediation understates the residual risk that remains.

Why D is correct: Correct: the segmentation and access controls do not fix the missing encryption but they reduce the chance and impact of unauthorised data exposure, which is the definition of compensating controls, so residual risk is reduced rather than removed.

Free sampleInformation Security Risk Managementhard

An information security manager receives a vulnerability scan report listing 4,200 findings across the production estate, each tagged with a CVSS base score. The board has asked which weaknesses most threaten the organisation. Before recommending a remediation order, what should the manager do FIRST to make the report meaningful for prioritisation?

  • ACorrelate each vulnerability with the business value and exposure of the affected asset and any compensating controls already in place. Correct
  • BSort the findings by CVSS base score and schedule remediation of every item rated critical or high within the next maintenance window.
  • CForward the full report to the system owners and ask each to remediate the findings on the hosts they are accountable for.
  • DCommission a penetration test to confirm which of the 4,200 findings can actually be exploited in the live environment.
Vulnerability findings must be contextualised against asset value, exposure and existing controls before they can be prioritised for remediation. Raw vulnerability data describes technical weakness in isolation. Risk to the organisation is a function of that weakness combined with the criticality and exposure of the affected asset and the controls already reducing the likelihood or impact, so contextualisation, not raw scoring, drives meaningful prioritisation.

Why A is correct: Correct: a vulnerability only matters in proportion to the value and exposure of the asset it sits on and the controls already mitigating it, so contextualising the findings against asset criticality and existing controls is what turns a raw scan into a risk-ranked basis for prioritisation.

Why B is wrong: Sorting by base score alone is tempting because CVSS appears to rank severity, but the base score ignores asset value, exposure and existing compensating controls, so it cannot identify which weaknesses most threaten this organisation.

Why C is wrong: Delegating to system owners feels efficient and respects accountability, but distributing an unprioritised list pushes the prioritisation decision down without business context, producing inconsistent local choices rather than an enterprise risk view.

Why D is wrong: A penetration test does validate exploitability and is valuable, but commissioning one before any business-context triage is slow and costly, and it still does not rank confirmed findings by the value of the assets they affect.

Other domains in this exam

See also the CISM cert hub, the study guide, and the cheat sheet.

Examworthy is not affiliated with or endorsed by ISACA. Original, blueprint-aligned practice material only.