How to answer MOST and GREATEST questions on CISA
GREATEST concern. MOST likely. MOST significant risk. GREATEST impact.
These questions give you four problems, and every one of them is a real problem.
Your job is to rank them. And most people rank them the way they would at work, not the way an auditor does.
The rule: pick the problem that causes the others
When four options are all genuine weaknesses, ask:
“If I could fix only one, which would make the others matter less?”
That’s almost always the answer.
A missing risk assessment is worse than an out-of-date document, because without the risk assessment you don’t even know which documents matter.
No policy is worse than a policy that isn’t followed in one case, because without the policy there’s nothing to follow.
Three tests for ranking problems
1. Root cause beats symptom. An option describing why things go wrong (no ownership, no risk assessment, no approval process) beats one describing a single thing that went wrong.
2. Whole objective beats part of it. Ask which problem could make the entire audit, project or control fail, not just slow it down or make it untidy.
3. Prevention gap beats detection gap, when the risk is the same. A control that can’t stop a problem is usually a bigger concern than a report that finds it late. But read carefully: a detective control that’s the only safeguard over a critical risk can be the bigger issue.
An example
An IS auditor is reviewing a project to replace the company’s finance system, due to go live in three months. Which of the following should be of GREATEST concern?
- A. The project is two weeks behind schedule
- B. User acceptance testing has not yet been planned
- C. The business case has not been reviewed since the project started, and costs have doubled
- D. Two developers have left the project team
Answer: C.
- A, B and D are real issues, and each can be managed: schedules are recovered, testing gets planned, staff are replaced.
- C questions whether the project should continue at all. If the benefits no longer justify the cost, everything else is detail.
Exam reality: When one option questions whether the whole thing is still valid, and the others are delivery problems, the validity question usually wins.
MOST likely is a different question
Careful: MOST likely asks about probability, not impact.
“Which is MOST likely to have caused…?” or “What is the MOST likely result of…?” wants the most direct, common cause or effect, not the most dramatic one.
- Users sharing passwords → the most likely result is that actions can’t be traced to individuals (loss of accountability), not “a major data breach”.
- No change management → the most likely result is unauthorised or untested changes reaching production.
Pick the consequence that follows directly from the weakness described.
The traps
The dramatic option. “Complete loss of data” or “the company could fail” sounds like the greatest concern. It rarely is, unless the scenario actually leads there.
The technical option. An IT person’s instinct is to rank the technical gap highest. CISA often ranks the governance gap higher: no owner, no policy, no risk assessment.
The small sample. One exception in a sample of 50 matters less than a control that doesn’t exist at all.
Practice drill
For your next 15 GREATEST or MOST questions, before choosing, ask of each option:
- Is this a cause or a symptom?
- Could it make the whole objective fail?
- Is the question about impact (GREATEST) or probability (MOST likely)?
Note any you miss and which test would have caught it.
Tomorrow, Part 4: BEST and PRIMARY questions, and how to tell the best choice from a good one.
