ISC · Information Systems and Data Management · 9 practice questions
Normal Change Management: Testing, Approval, and Migration
For a normal change, complete testing, obtain formal release approval, and document a rollback plan before independent production migration. Below: a billing calculation fix moves from the initial request to a controlled production release.
Try one first
Hint
Focus on the core preventive controls in normal IT change management, especially approval, testing, and who should move code into production.
Answer B. A sound change-management process requires formal documentation of the change request, testing in a nonproduction environment, obtaining appropriate approvals before deployment, and separation of duties so the developer does not migrate their own change to production. These preventive controls reduce the risk of untested or unauthorized changes causing billing errors. Because this is not an emergency change, the normal approval and implementation controls should be followed.
Why not A: This option relies on a detective control (post-implementation review) and allows the developer to both create and deploy the change, violating separation of duties. Detecting problems after they affect billing is too late compared with preventive testing and approval.
Why not C: Business users can participate in acceptance testing, but they should not implement code in production. Informal user confirmation is not a substitute for documented testing and IT-controlled deployment with proper approvals.
Why not D: Backups aid recovery but do not replace preventive change-management controls or separation of duties; allowing the developer to install directly in production still increases the risk of unauthorized or harmful changes.
The timeline
- Day 0Document and obtain initial approval of the billing calculation change request.
Identify the requested correction and its business impact. Initial approval authorizes development, not production release. This is a normal change, with no emergency exception.
- DevelopmentThe developer codes the billing fix and performs unit testing.
Successful unit tests do not make the fix ready for production. Required QA and user acceptance testing remain.
- TestingQA and billing users test the fix in a separate, production-like environment.
Document and retain results showing that the correction works and does not disrupt existing billing calculations.
- Before release approvalDocument the rollback plan for the billing fix.
Specify how to back out the change if implementation fails. A backup alone does not replace a documented rollback plan.
- Release approvalThe business owner or required change authority reviews test results and impact, then formally approves release.
Release approval follows completed testing. The initial approval to develop the fix does not satisfy this requirement.
- Before production migrationConfirm the release controls are complete and an independent operations person is assigned.
Postpone deployment if required testing, formal release approval, the rollback plan, or independent migration is missing.
- Production releaseAuthorized operations personnel migrate the approved billing fix, not the developer.
Restrict production deployment rights to authorized personnel independent of development, and retain the migration record.
Key points
- Remember: Approval to start development is not approval to release; release approval follows documented testing.
- Developer unit tests do not replace the required QA and user acceptance testing.
- If testing, release approval, rollback planning, or independent migration is missing, postpone deployment.
How the exam traps you
- Treating successful developer unit tests as sufficient for production release. Complete the required testing in a separate, production-like environment and retain the results.
- Approving production release when coding ends, before testing results are available. Obtain formal release approval after testing results and business impact have been reviewed.
- Letting the developer deploy because testing and business approval are complete. Use authorized personnel independent of development to migrate the approved change.
Question 2
Hint
Think preventive, pre-deployment controls (formal approvals, production-like testing, segregation of duties, rollback planning) rather than detective measures or relaxed emergency authorizations.
Answer C. Option C is best because payroll changes affect financial reporting and require preventive governance. Formal change requests and CAB approval provide oversight; testing in a production-like environment and segregation of duties reduce the chance of erroneous payroll calculations reaching production; and a pre-deployment rollback plan limits the impact of failures. Emergency exceptions should be narrowly defined, authorized in advance when possible, and reviewed afterward.
Why not A: Tempting because it preserves deployment speed and adds apparent controls (tests and checklists), but it fails to remove developer production access or require CAB approval, production-like acceptance testing, or a formal rollback plan. Those missing preventive controls mean erroneous payroll code can still be introduced before detection.
Why not B: Tempting because it centralizes expertise for rapid fixes, but it is wrong: granting a developer team persistent production write access and bypassing CAB concentrates risk and removes preventive governance. Post-deployment verification and audits are detective and cannot reliably prevent payroll errors from affecting runs and financial statements.
Why not D: Tempting because monitoring improves detection and can speed remediation, but it is wrong: monitoring is a detective control and does not prevent incorrect code from being executed in payroll runs. Without pre-deployment approvals, production-like testing, segregation of duties, and rollback planning, the primary risks remain unmitigated.
Question 3
Hint
Focus on which control prevents a developer from both creating and deploying a change without independent oversight.
Answer D. A key change management control is separating development from production migration so that no single person can create and deploy a change without independent oversight. Even when approvals and testing are documented, allowing developers to perform the production migration creates a risk that unapproved or altered code will be implemented. That gap weakens the intended change-management controls.
Why not A: Post-implementation documentation does not replace preventive controls. Recording that a developer moved code after the fact doesn't prevent unauthorized or unreviewed changes from reaching production.
Why not B: This is tempting because testing is an important control. However, testing does not prevent a developer from bypassing the approved change or deploying a different version; migration authority and segregation of duties are complementary preventive controls.
Why not C: This distractor points to access rights, which are relevant, but migration authority is part of the change process. The problem affects change management directly because who can perform the migration determines whether changes are independently implemented.
Question 4
Hint
Focus on what makes a production change acceptable under control standards, even when the business says the change is urgent.
Answer A. Formal change management is governed by whether the change has gone through appropriate authorization, testing, and approval before promotion to production. If the situation is urgent, an emergency change process may be used, but the change still should be documented and subject to required review. Urgency does not replace control over production changes.
Why not B: This is tempting because business urgency often drives the timing of a change. However, a user request and a deadline do not by themselves justify bypassing change controls. A valid emergency can accelerate the process, but it does not eliminate the need for authorization, testing, and approval.
Why not C: This distractor appeals to the common shortcut of prioritizing speed over control. In a sound change management process, the ability of one person to act quickly is not the governing factor, and direct developer access to production may create segregation-of-duties concerns. The key issue is whether the change was properly controlled, not who can do it fastest.
Why not D: This can seem plausible because candidates may think table updates are less risky than code changes. But configuration and master-data changes can still materially affect processing and financial reporting, so they also require change management. The nature of the item changed does not override the need for proper authorization, testing, and approval.
Question 5
Hint
Focus on which change management steps are expected before a routine change reaches production, not on management's deadline pressure.
Answer A. For a routine (nonemergency) change, key change-management controls, formal approval, user acceptance testing, and a documented back-out plan, should be completed before production deployment. These controls reduce the risk of unauthorized or defective changes and ensure recoverability if something goes wrong. Deadline pressure does not eliminate these required preventive controls.
Why not B: Tempting because unit testing is a legitimate step and management pressure may push for speed. However, developer unit testing alone does not replace UAT, formal approval should not be retroactive, and documenting approval after the fact weakens preventive controls.
Why not C: Tempting because business-owner involvement matters and verbal approval may seem fast. But verbal approval does not substitute for the formal approval process, and it does not address missing UAT or the absent back-out plan.
Why not D: Tempting because a partial release may appear to reduce immediate impact. However, splitting the deployment without completing approval, UAT, and a rollback plan still leaves operations exposed and can increase complexity and risk.
Question 6
Hint
Identify which control was explicitly bypassed during implementation (who performed the deployment and how that maps to segregation of duties and deployment controls).
Answer C. The developer's direct edit of production code breached segregation of duties and circumvented the policy-required implementation controls (operations-led automated pipeline). While the emergency authorization and delayed documentation are issues, the immediate and most serious control failure is the implementer performing changes they were not authorized to perform, increasing risk of unauthorized or incorrect changes affecting financial reporting.
Why not A: CAB oversight is important, but the policy explicitly allows emergency authorization by the IT operations manager with later CAB review. The bigger failure was bypassing the operations implementation requirement, not the CAB timing in this emergency context.
Why not B: Insufficient testing is a valid concern, especially for revenue-affecting code, but the stem emphasizes that required implementation controls and segregation of duties were bypassed (developer edited production directly), which is the more immediate and core control deficiency.
Why not D: Delayed documentation and after-the-fact approvals reduce auditability and are weaknesses, but they are secondary to the primary control breach: the developer directly modified production, circumventing the designated operations deployment process and segregation of duties.
Question 7
Hint
Focus on the key objectives of change management: authorization, testing, documentation, and separation between development and production.
Answer A. This is the strongest change-management control because it provides formal authorization and ensures changes are properly tested before release. Documented testing and approval reduce the risk of defects or functional regressions reaching production. Requiring someone independent of development to perform the production migration enforces segregation of duties and reduces the chance of unauthorized or untested code being implemented.
Why not B: Manager approval by email seems like a control but does not eliminate the conflict of interest when the developer both creates and deploys the change. It also fails to address the lack of documented testing and formal change approval procedures.
Why not C: Delaying changes to month-end is a scheduling tactic, not a substitute for proper testing and authorization. It doesn't ensure changes are validated before release and can still allow untested or unauthorized changes to reach production.
Why not D: Backups help with recovery but are corrective rather than preventive controls. They don't prevent unauthorized or insufficiently tested changes from being implemented and therefore don't strengthen change-management controls.
Question 8
Hint
Focus on who can bypass the normal change process and whether any compensating independent control catches that risk.
Answer B. Change management is strengthened when developers do not have direct access to production and approved changes are migrated by someone independent. Here, developer production access creates a path around the formal ticket, approval, testing, and migration process. Because there is no independent review of those direct changes, the control weakness is significant.
Why not A: This is tempting because emergencies do sometimes require faster handling. However, urgent fixes do not remove the risk created when developers can change production directly without independent review. The concern is the bypass of controls, not whether emergencies occur.
Why not C: This distractor misstates the fact pattern. Business or user approval of a completed change can be a valid part of ensuring requirements are met; the stronger control problem here is developers' direct production access that bypasses migration and review controls.
Why not D: Documentation is important, but it is a compensating or detective control, not a replacement for preventive controls like restricting direct production access and requiring independent migration or review. Post-hoc documentation does not prevent or detect all risks from unauthorized or untested changes.
Question 9
Hint
Focus on which fact most directly affects control over migration into production, not just whether every step is ideal.
Answer C. The shared deployment credential undermines accountability and traceability because the deployment tool only logs the shared account, not the individual who initiated the deployment. That gap makes it difficult to determine who actually promoted code to production and increases the risk that unauthorized or unapproved changes could occur without reliable attribution. The other controls described (peer review, QA testing, and documented post-implementation review for emergencies) are reasonable when operating as stated.
Why not A: This choice is tempting because preapproval is ideal, but many organizations permit emergency fixes to be implemented before final business-owner sign-off provided there is prompt documentation and a timely post-implementation review, which the fact pattern specifies. That makes the shared deployment credential a broader, more persistent control gap.
Why not B: This distractor conflates technical code review with business authorization. Developer peer review is an appropriate control for technical quality, while the business owner's role is to approve the business rationale or release. Requiring the business owner to perform code review would be impractical and is not indicated by the facts.
Why not D: Testing before final business-owner approval is common and often desirable because the business owner should approve a release that has already been validated. The inability to identify the individual who performed the deployment poses a greater control and accountability risk.
Common questions
When should a normal change receive formal approval for production?
After testing results are documented and reviewed, but before production migration. Initial request approval authorizes development, not production release.
Are developer unit tests enough before deploying a normal change?
No. For this billing fix, complete QA and user acceptance testing in a separate, production-like environment and retain evidence before release approval.
Can a normal change be deployed without a rollback plan?
No. Document how to back out the billing fix before release. Postpone deployment if the rollback plan or another required release control is incomplete.
Watch it solved
A full CPA ISC task-based simulation on Change management, worked step by step.
Change Management: Approval & Segregation | CPA ISC TBS on YouTube
Practice ISC like the real exam
The free ChatCPA simulator: real exam layout, timed testlets, starting with a question on this topic. No account needed to start.
Open the free simulator →More on Change management
Questions from the ChatCPA bank of 17,658 CPA exam questions, each with a written reason for every wrong answer. ChatCPA is built by Nicholas Miller, CPA (Oregon #14907). How these pages are made. Spot an error? Tell us.