Guide to Improving Speed in Written Tests such as the PST
As with case interviews, the best way to get good at written tests is through practice. The second best way is to practice. And the third best is… practice. The good news is that there is ample training material available on the internet, as well as several official practice tests.
Despite some similarities across the consulting firms (e.g., a lot of them use GMAT style math questions) there is no single ‘standard’ model test that you have to prepare for. In this article we will give you some general advice on how to make sure that you are able to solve the typical questions fairly quickly. We will also go over the most prevalent question types to give you an idea of good strategies to tackle them. You will also learn how to avoid common pitfalls.
Time is usually the single most crucial factor in solving PST-style tests
Overall, the questions given in this type of tests are actually not that hard. What makes them hard is the time restriction consulting firms place you under when taking the test.
Apart from simply practicing the questions, there are some tips you should know that will help you succeed in PST-like tests.
Efficiency: don’t waste time by not being as quick as you could
This one seems to be the most straight forward and also the hardest to implement: practice to get quicker at otherwise time-consuming tasks. Typically, the two best candidates for simply getting quicker are case math and reading.
This is the most obvious skill to improve in terms of speed. The trick here is the right kind of preparation:
- Practice accurate calculations first. There is no point in quickly calculating wrong results.
- Practice getting faster once you are certain that you end up with the right result most of the time.
It is also always a good idea to have a look at the simple multiplication tables and calculating with percentages calculations. PrepLounge allows you to practice these skills extensively.
Consultancies will throw tons of text to read at you during the test. To get through it quicker there are two things you can do:
- Improve your reading speed: there are various speed reading exercises and even smart phone apps. If you have enough time this is a valuable skill to learn (that will help you your entire life).
- Prime yourself for the information you are looking for. Priming is one of the great features of our human brains. In essence, priming means that your brain is better at discovering information it already knows it is looking for. You can best utilize this by approaching questions that will require digging for information this way:
- Read the question
- Read the answers
- Then read the additional information: by reading the question & answers first you prime your brain to look for the information to solve it and skim through the useless parts quicker
Effectiveness: do it right the first time
While very good candidates are able to finish the written tests on time, hardly anyone has time to go over the test a second time to check everything. Since we all are prone to errors – especially in high pressure situations – it is paramount that you give yourself as little chance to slip as possible without affecting your speed too much. This is actually something you will be taught at the beginning of your consulting career as well. Sometimes this is dubbed as limiting your ‘f*ck up potential’ or FUP.
Transfer answers to the answer sheet right away
Some consulting firms require you to do their tests paper-based and to record your answers on a special answering sheet. If this is the case: transfer every answer right after answering it. Do not batch-process answers in 5 or 10 question blocks or transfer all the answers in the end. While batch processing gives you a very small time advantage it also bears a very high risk of error or even worse: forgetting one of the answers.
Forgetting one of the answers can lead to your answer key shifting by one question which can lead to several wrong questions in a row. We do realize that there is practice material that recommends you this batch processing approach but at PrepLounge we feel very strongly that the best approach is transferring 1:1. Chances are very high that you will run out of time towards the end of the test. Save yourself the stress and FUP of having to transfer 26 questions in 15 seconds, thereby also making sure that you hit the right column and line for each of them.
Mark both the question and eliminated answers graphically
In most cases you will receive much more information than you both need and can handle while the key information hides somewhere in the text. Marking numbers and information you need in the text helps you quickly rediscover the answers. Time is too precious to spend it by quickly and accurately calculating with the wrong row of a data table. The same goes for answer possibilities you eliminate: cross them out. That way you do not have to re-think in case you want to revisit the question.
If you start solving a question: always finish by marking an answer as well
Don’t leave questions open for later review. Chances are that you will have other questions that your time will be better spent with. What do you do if you discover that a question proves more difficult than you thought in the beginning? The answer is: you do not persevere. A general rule of thumb is that if you cannot figure out how to solve the question within the first 30 seconds, there will probably be other questions you should be solving first. (You do not need to be able to actually solve it within those 30’’ but you should be able to know how you can solve it).
If you encounter a question that proves difficult, resist the temptation to crack it. Trust your gut feeling on what the answer might be, mark the question and move on. If there are penalties for wrong answers you might consider if it’s worth it to transfer the answer to the answer sheet. Refer to the section ‘Know the rules’ below for further guidance!
Eliminate some answers quickly
Almost all PST-like tests have multiple choice answers. That gives you one advantage: you do not necessarily have to find the solution; you (only) have to find the best answer that is given.
- After you have set up how to calculate the question, you can actually guess 1-3 answers that probably will not be right by doing a very rough estimation (be it that they are too far off, the decimal is not right etc.). See the general math question section for an example.
- Especially if the multiple choice answers are numbers, it can also be quicker to reverse engineer the result rather than calculate it. Example: If the question reads like ‘By which percentage did revenues grow from Year 1 to Year X’ with the answers being something like 2.9%/4.2%/5.4%/5.7%, some answers can be eliminated quickly by estimating how much the respective percentages would be of the base revenue in Year 1.
Work smart, not hard: reverse engineer wherever possible
A story every first-year consultant will be able to relate to at some point goes like this:
After your successful start in a major consulting firm you approach your case team leader and suggest him that you could do a super interesting and clever piece of analysis. He looks at you with a slightly confused expression and asks you ‘Does it change the answer?’
Time is scarce on consulting projects, therefore consultants can only do so-called answer changing analyses that might change the hypothesis they are currently working on. You can also use this principle to your advantage. Take the example of a question in a McKinsey PST test that asked you to order five ratios (medium/hard divisions) from smallest to largest (if done accurately: medium/hard difficulty calculations). The MC answers went like this:
- 1, 2, 4, 3, 5
- 1, 2, 3, 4, 5
- 1, 3, 2, 4, 5
- 1, 3, 2, 5, 4
As you can see each answer starts with ratio 1. Calculating ratio 1 therefore is a complete waste of your time – it is the least answer changing analysis. On the other hand three out of the four answers end with a 5 – knowing that 5 would indeed be smaller than 4 would be answer changing: it would allow you to eliminate 3 out of 4 answers (therefore finding the right answer). If ratio 5 would indeed be larger than ratio 4, you could go on by calculating if ratio 3 or ratio 4 are actually larger (one additional calculation). Only if this failed you would have to calculate a 4th ratio (ratio 2).
Reverse engineering is especially powerful when coupled with doing rough estimations first (in this example, it would have shown you very quickly that 3 is indeed smaller than 2).
Finding these ‘smart’ ways of solving the questions usually saves you a lot of time in PST-style tests; after all, this is what consultancies want to test for. In the example above, D was indeed the right answer. Another ‘clever’ (but slightly less obvious) approach for the example above would have been to start with ratios 3 and 4, to test if Answer A is the correct one.
Since by now you probably all have been honing your math skills to the edge of the humanly possible you already figured out that even if the consultancy did not want to test for your ability to work smartly (which they did by conceiving the question with answer D being the right one), starting with trying to exclude answers A or D would have been beneficial anyways. Why? Because starting by testing for one of those answers would require you to calculate an average of 3.25 ratios to solve the question. This is compared to 3.5 ratios when starting by trying to decide between A/B and C/D. Calculating all ratios from 2 to 5 and simply looking up what the right order is would require you, on average, to do 4 calculations. (We trust that at this point no one is wondering why in this case you would never have had to calculate 5 ratios.)
Know the rules
Usually a lot of candidates end up in a ‘Borderline-Fail’ bucket in written consulting tests. That means that, for those candidates, 1 or 2 additional right questions might make the difference. In order to score these crucial points, make sure you know the rules of the test you’re taking and spend some time to think about what they mean for your strategy. Of course, you should do this before the test day. If you do not get information prior to the test, make that little extra effort to call or email the recruiting department and ask them if there is any additional information available.
For example, know if there are penalties for wrong answers. If there are no penalties for wrong answers, you obviously want to mark an answer for every single question even if that means marking ‘c’ 5 times in a row at the end of the test.
If you have no clue, never mark answer A in a multiple choice test. Answer A is statistically less likely to be correct than the other answers if a human designed the order of the answers (sometimes right answers are randomized automatically). Marking an answer while not being sure might pay off even if there is a penalty. Take for example BCG’s potential test where a correct answer scores you 3 points and an incorrect one scores you -1 point. In such a test, if you have eliminated 1 of 4 possible choices, you should guess. Your statistical score will then be 66% * -1 + 33% * 3 = 0.33, which is positive.
At each new section: get a quick overview first
Some consultancies nowadays offer computer-based tests where questions are arranged in sections around hypothetical clients (e.g., the BCG potential test). The information you get can be on several pages throughout the section. Feedback we got from test takers is that it pays off to get a first impression of the information on each page of the sections before starting to answer questions. Do not read it in detail yet; just skim through it quickly to get an idea of what it is about.
Do not answer questions in the order they occur
Not all questions in a written test are of the same difficulty level. Instead of going through the test solving questions in order of appearance, it has proven to be more efficient to capture the ‘low hanging fruits’ first. Typically, all questions get you the same amount of points when the test finally is graded. Therefore it makes more sense to make sure that you first get the easy ones right and worry about more difficult ones later.