X · Exam Strategy
Everything else in this course teaches you to answer questions. This chapter teaches you to take an exam, which is a different skill and — for you specifically — probably the one costing you more points.
Here is the situation this chapter is written for. You are somewhere between eight and twenty years into a career. You are good at your job. You have somewhere between two and six hours a week for this, and they are the hours after everything else is done. You will sit this exam once, maybe twice. That profile changes what matters. A twenty-four-year-old with six months and twenty hours a week can afford to be inefficient; volume covers it. You cannot. Every hour you spend has to buy something, and on exam day every minute has to buy a point.
Most preparation material ignores this entirely. It assumes the reason you got a question wrong is that you did not know the mathematics. Often you did know it. You ran out of time, or you spent four minutes on a question you should have abandoned in ninety seconds, or you second-guessed a correct answer with eight minutes left. None of that is a knowledge problem, and no amount of extra content fixes it.
The Executive Assessment is module-adaptive. Each section is delivered in two modules, and your performance on the first determines the difficulty of the second.
That is different from a question-adaptive exam, where each individual item is selected based on how you did on the previous one — and the difference has one consequence that changes how you should work.
Within a module, you can move freely. You can skip a question and come back. You can change an answer you have already given. You can review the whole module before you commit to it. The exam is not watching you item by item and adjusting; it is waiting until the module is done.
Most candidates never use this. They answer each question in order, commit, and move on — as though the exam would punish them for going back. It will not. What you cannot do is return to a module once you have submitted it. That boundary is real, and it is one-way.
- First pass. Answer everything you can answer cleanly. Flag anything that is fighting you and move on immediately.
- Second pass. Return to the flagged items with whatever time is left, working from the most tractable to the least.
- Final check. Make sure nothing is blank. A blank is a guaranteed zero; a guess is not.
That structure protects you from the single worst outcome in a timed exam: not reaching questions you could have answered because you spent your time on ones you could not.
Since module one determines what module two looks like, it carries weight beyond its own questions. That is not a reason to slow down and be careful — being slow in module one is how you fail to finish module one. It is a reason to be disciplined early, when discipline is easiest and the temptation to linger is strongest.
Most candidates have met a question-adaptive exam somewhere, or heard it described, and carry the assumption forward: that going back is either forbidden or somehow penalised. Nothing on screen contradicts it, so the assumption survives the whole section.
The cost is invisible, which is why it persists. You never see the question you would have got right if you had flagged the hard one and returned — you only see a score. Knowing the exam is module-adaptive, not question-adaptive, is what makes the first-pass/second-pass structure available to you at all.
Each section gives you 30 minutes. The item counts differ:
| Section | Items | Minutes | Average per item |
|---|---|---|---|
| Integrated Reasoning | 12 | 30 | 2:30 |
| Verbal | 14 | 30 | ~2:08 |
| Quantitative | 14 | 30 | ~2:08 |
Those averages are the least useful numbers in this chapter, and you should know why before you build a plan around them.
Nobody spends the average. A Sentence Correction item you recognise takes forty seconds. A Multi-Source Reasoning item with three sub-questions takes four minutes and is supposed to. If you try to spend 2:08 on everything, you will rush the easy ones — where points are cheapest — to fund the hard ones, where they are most expensive.
Pick two moments in each section and know where you should be:
| Section | At 10 minutes | At 20 minutes |
|---|---|---|
| IR | 4 items done | 8 items done |
| Verbal | 5 items done | 10 items done |
| Quant | 5 items done | 10 items done |
Two checkpoints, three numbers to remember. That is all the arithmetic you should be doing about time during an exam, because every second spent computing your pace is a second not spent on a question.
Integrated Reasoning has the most uneven items of the three sections. A Multi-Source Reasoning set with three sub-questions and a Graphics Interpretation item with two dropdowns are not the same size of task, and treating them as such is how candidates run out of time in IR specifically. Plan for the MSR sets to eat four minutes each, and take that budget from the single-answer items.
This is the highest-value section in this chapter, and the hardest one to actually do.
Here is the arithmetic that should govern the decision, and it is brutal in its simplicity. Every item is worth the same. A question you fight for four minutes and get right is worth exactly what a question you answer in forty seconds is worth. The four-minute question cost you the time of two other questions.
So the question is never "can I solve this?" It is "can I solve this in the time it would take me to do two other things?"
At ninety seconds, decide. Not finish — decide. Ask yourself one question: do I have a clear next step, or am I hoping something will occur to me?
If you have a next step — you know what to compute, you know which two options you are choosing between — continue. You are working.
If you are re-reading the question hoping for inspiration, you are not working. You are hoping. Eliminate what you can, guess, flag it, and move.
It is not leaving the item blank. It is:
- Eliminate any option you are confident is wrong — this takes ten seconds and roughly doubles your odds
- Choose among what is left
- Flag the item
- Move on without looking back
That last part matters. An item you have abandoned and keep thinking about is still costing you time, just on a different question.
Every other abandonment rule people try is a rule about the clock — "never spend more than two minutes" — and clock rules fail because they cannot tell a productive third minute from a wasted one. Sometimes minute three is where the item cracks.
"Working or hoping" is a rule about your state, not the clock, and that is what makes it decidable in the moment. Most candidates cannot tell the difference at first, only because nobody has ever asked them to notice it. Once you start noticing, the ninety-second decision becomes almost automatic — which is exactly what Drill 1 is for.
There is a specific failure mode that costs more points than any single topic in this course, and it has nothing to do with the exam. It is about you.
You hit a question you should be able to do. You cannot do it. And instead of leaving, you stay — because leaving would mean admitting you could not do it.
This is the anchor, and it is worse in experienced professionals than in students. You have spent fifteen years being the person who solves the problem. You are not used to walking away from one, and the exam is built to exploit exactly that.
- The exam contains items you cannot solve in the time available. That is not a defect in you; it is how the instrument works. A test where you could solve everything would not distinguish you from anyone else.
- Nobody sees your process. There is no partial credit for effort, no record of the beautiful approach you almost completed. The only thing that exists is the letter you selected.
- The anchor does not cost you one point. The four minutes you spend proving you can do a hard question are funded by two easy questions you will now rush or never reach. It costs you two or three.
In your working life, persistence on a hard problem is almost always correct. The problem does not expire at 30 minutes, nobody is counting how many others you solved that hour, and being the person who does not give up is a large part of how you got here.
The exam inverts every one of those conditions: fixed clock, equal weights, no observer. The instinct that serves you at work is precisely the one that costs you points here — which is why this is a failure mode of the experienced, and why it has to be handled as a decision made in advance rather than a judgement made in the moment.
You do not have a calculator in Quant. You do have one in IR. Both facts matter less than knowing when precision is required at all.
Calculate when the margin is tight. If the options are 103, 107, 110 and 112, or a statement asks whether something exceeds a threshold and your estimate lands within a few percent of it, estimation will betray you.
The exam writes these deliberately. An item where the true value is 14.83 and the threshold is 15 exists for one reason: to catch the candidate who rounded to “about 15” and then had to guess which side of the line they were on.
Nothing about that item looks dangerous while you are doing it. The danger is entirely in the gap between your estimate's precision and the margin the question was built around — and you can see that gap before you compute, by looking at the options first.
Look at the answer choices before you compute. They tell you how precise you need to be. Candidates who compute first and look at the options afterwards routinely do exact arithmetic where a glance would have done — and occasionally estimate where they needed to be exact.
Where estimation is most valuable: percentage change, ratios, and anything involving large numbers where the trailing digits cannot affect which option is correct.
Where it is most dangerous: near a boundary (“more than 25 percent”, “at least 350”), and in Data Sufficiency Yes/No items, where the entire question is often whether a value falls on one side of a line.
You will guess. Plan for it rather than treating it as failure.
The arithmetic is straightforward. Five options, blind guess: one in five. Two options eliminated: one in three. That is a meaningful difference, and eliminating two options usually takes fifteen seconds — which makes it one of the highest-return activities available to you on the exam.
- Multiple choice, Verbal. Options that are true but do not answer the question asked. Options whose scope is wider than the argument's — anything saying “always”, “never”, “all” when the argument was about a specific case.
- Sentence Correction. Options with an obvious grammatical error you can name. You do not need to know which of the remaining three is best to eliminate the two that are broken.
- Quant, Problem Solving. Options with the wrong sign or the wrong order of magnitude. A quick sanity check on size often kills two.
- Data Sufficiency. If you are certain one statement is sufficient, you have already eliminated three of the five. If you are certain one is not sufficient, you have eliminated two.
Two-Part Analysis, Table Analysis and most Multi-Source Reasoning items require every part correct for any credit. Guessing on one part of a three-part item does not give you a third of a point — it usually gives you nothing.
Which means: on multi-part items, spend your effort on getting all parts right, or accept that the item is likely lost and spend the time elsewhere. Half-solving a three-part item is the worst possible use of two minutes.
Something happens to judgement in the last five minutes of a timed section, and it is worth deciding what you will do before you are in it.
A blank is a certain zero. A guess is a one-in-five that costs three seconds. If you have flagged items and the clock is running out, the very first thing you do with the last two minutes is put a letter on everything unanswered. Then go back to thinking. Candidates lose points here every single administration, and it is the most avoidable loss on the entire exam.
With four minutes left and one flagged item, the pull to revisit an answer you were unsure about is strong. Resist it unless you have a specific reason — a concrete error you can name, not a feeling.
- Scan for blanks. Fill every one. Non-negotiable, do it first.
- Return to flagged items in order of how close you were to an answer.
- Change an existing answer only if you can name what was wrong with it.
- At one minute, stop working and verify once more that nothing is blank.
Your first read had the full context of the question fresh: the stem, the constraint, why you eliminated what you eliminated. Your last-minute reconsideration has fragments of that plus adrenaline. It feels like a second opinion; it is actually a worse-informed first opinion. That asymmetry is why "found a concrete mistake" is the only threshold worth acting on.
Thirty-two hours total. That is a real constraint, not a pessimistic one, and it is enough — if none of it is wasted.
Here is what wastes it: doing practice questions without reviewing them. It feels productive, it fills the time, and it teaches you almost nothing. If you have thirty-two hours, roughly half of them should be spent on questions you have already answered.
- Week 1 — Diagnose. Take a full practice exam under real conditions: timed, no interruptions, no phone, all three sections. Then spend as long reviewing it as you spent taking it. You are not looking for your score. You are looking for what kind of wrong you were — the five categories in Self-diagnosis, below.
- Weeks 2–3 — Formats. Learn the formats that do not exist outside this exam. Data Sufficiency first — it is roughly half of Quant and nobody arrives knowing it. Then the IR formats. This is the highest-return period of the entire eight weeks, because it converts questions you cannot approach into questions you can.
- Weeks 4–5 — Content. Now go after the specific topics your diagnostic flagged. Not everything: the two or three that cost you most. Trying to review all of arithmetic in two weeks means learning none of it.
- Weeks 6–7 — Timed practice. Sections under the clock, then full review. This is where the strategy in this chapter becomes habit rather than theory. The ninety-second decision has to be automatic by exam day, and automatic takes repetition.
- Week 8 — Taper. One full exam early in the week, light review after, nothing new. Do not learn new content in the last week; it displaces things you already know and it costs you sleep, which matters more.
Two sessions of two hours beats four of one. The exam is 90 minutes of sustained concentration, and if you have never worked for 90 minutes straight in preparation, exam day is the first time — which is a bad day to find out what that feels like.
Every error this chapter has described, in one place. When strategy costs you a point rather than knowledge, it is almost always one of these eight. Open each to see the shape it takes and the defence against it.
This chapter is not learned by answering questions. It is learned by changing how you behave while answering them. These five drills build the specific habits, and each one is done away from this page — with a timer, a practice set, and something to write on.
Set a repeating timer at ninety seconds. Work a set of 10 mixed items. Every time it rings, stop and write one word: working or hoping.
Do not change what you are doing — just label it. After ten items, look at your labels. Most people discover that “hoping” started well before they left the item.
The point of the drill is noticing, not fixing.
Take a full section timed. At the 10-minute and 20-minute marks, write down how many items you have completed. Compare against the checkpoint table in X.2. Do this three times across different sections.
You are looking for a pattern: are you consistently behind in one section, or does it vary? A consistent gap is a pacing problem; a variable one is usually an abandonment problem.
Work a set of 10 items with a hard rule: you may not spend more than two minutes on any item, no exceptions. When time is up, eliminate, guess, move. Score it.
Most candidates score the same or better than they do unconstrained — which is the lesson. The time you thought you needed was not buying points.
On 10 Quant items, read the answer choices before doing any work, and write down whether you will estimate or calculate. Then solve.
Check how often your prediction was right. This builds the X.5 judgement into a reflex.
For one week, track two numbers: minutes spent answering new questions, and minutes spent reviewing questions you have already answered.
If the second is less than the first, your preparation is inefficient regardless of how many questions you have done.
After your next full practice exam, categorise every wrong answer into exactly one of these. The distribution tells you what to work on far more precisely than the score does.
| Category | What it means | What to do |
|---|---|---|
| Ran out of time | Never reached it, or rushed it | X.2 and X.3. Content work will not help |
| Misread the question | Answered a question that was not asked | Slow down on the stem specifically — re-read it after choosing |
| Did not know how | No approach came to mind | Content. This is the one that needs a chapter, not a habit |
| Knew it, executed badly | Right approach, arithmetic or sign error | Practice under time. Not more theory |
| Second-guessed | Changed from a correct answer | X.7. This is the most fixable category on the list |
If more than a third of your wrong answers fall outside “did not know how”, then strategy is your cheapest available improvement — and this chapter is worth more to you than the next four content chapters combined.
Every item is worth the same. The only question that matters, on every single one, is whether it is worth what it is costing you.
X applies to all three sections rather than sitting inside one. The drills above are worth repeating throughout your preparation — the ninety-second decision only becomes automatic with repetition.