How to master the quantitative part of the MBB case interview

The ‘Quantitative section’ of the MBB case interviews is the part of the case many interviewees stress most about, and (in my years of interviewer experience) the part most great interviewees fail under stress. But the fact is, these sections typically have easy math, calculations you could do easily when not under pressure. 

The key challenge is that candidates typically don’t prepare for this section correctly. 

In this article, coach Thor will help you master the quantitative part of the case interview, by covering: 

1. How the ‘Quantitative Case Section’ is typically structured
2. How do you best prepare for and tackle this section under pressure
3. How you gain ‘extra points’ with your interviewer  

The role of the ‘Quantitative Section’ in Your MBB Interview 

As you are hopefully aware of already, MBB interviews are typically structured into three distinct parts:  

1. Introductions & background/resume
2. PEI (Personal Experience Interview)
3. Case 

If this is new to you, keep in mind that parts 2 and 3 (PEI and Case) carry the same weight. Many candidates fall into the trap of preparing almost exclusively for the case and neglecting the PEI. Coach Florian has a great McKinsey-focused article on that topic.

Now, let’s shift the focus back to this article. We are looking at the Quantitative Section of the case.

How important is the Quantitative Section in an MBB interview?

The case accounts for roughly 50 percent of your overall interview score. Within the case, you will almost always find three equally important parts:

1. The Analytical Section, where you give a structured answer to a broad case question
2. The Creative / Data Section, where you generate hypotheses and insights from limited or messy information
3. The Quantitative Section, where you perform calculations in a structured way and draw conclusions from the results

These three components each carry the same weight within the case evaluation. This means that even if you perform well in two of them, completely missing one section can still hurt your chances of moving forward.

From my experience as a McKinsey interviewer with more than 100 interviews, I was surprised to see how often candidates struggled specifically with the Quantitative Section, including many engineers.

MBB Quant-question example and what's difficult about it 

The truth is, the ‘math-difficulty’ of the Quant-questions is not high. Over 90 % of typical interviewees could easily tackle the math if there is no stress/pressure involved (with the small exception being candidates with minimal math exposure in their educational backgrounds).  

Lets explore a typical (complex) Quantitative Case Question: 

Retailer X is a large clothing retailer, with presence in Germany and Poland. Due to declining market share, they’ve decided to launch a new marketing campaign, but they want to understand its effectiveness well before full launch. Thus, they focus their marketing on specific regions in Germany (Test group), whilst the rest of the German market is unimpacted (Control group). Their main focus is to increase revenues from existing customers. They’ve estimated that ~20,000 customers in Germany will fall into the Test group), whilst they have ~80,000 in the control regions.  

Question: What is the profit impact of the new marketing campaign. 

Additional data includes: 

• Cost of Campaign: 100,000 EUR
• COGS (Cost of goods sold) percentage: 50% 

Table 1: 6-month purchase change by channel during marketing campaign 

Table 2: Average orders by customer group and channel past 6 months pre-marketing : 

The answer to this question is as follows: 

1. Revenue from Marketing affected customers during marketing campaign:
   • Spend per customer = “Avg. Spend per purchase” * “Number of purchases”
   • Marketing affected customer Online spend = 30*(100%-20%) * 6*(100%+66%) = 24*10 = 240 EUR
   • Marketing affected customer In-Store spend = 40*(100%-25%) * “5*(100%+20%) = 30*6 = 180 EUR
2. Revenue from Control group customers during marketing campaign:
   • Control group customer Online spend
     30*(100%) * 6*(100%-10%) = 30*5,4 = 162 EUR
   • Control group customer In-Store spend
     40*(100%) * 5*(100%-20%) = 40*4 = 160 EUR
3. Marketing affected customers increased spend: 240 + 180 – 162 – 160 = 98 EUR
4. Total revenue increase in marketing affected customers
   • “Increased Spend” * “Number of marketing affected customers”: 
     98 EUR * 20,000 customer = 196,000 EUR
5. Profit of increased spend
   • “Total Revenue Increase” * “(1-COGS)”
     196,000 EUR * (100%-50%) = 98,000 EUR

Total profit impact of Marketing Campaign = “Profit of increased spend” - “Cost of marketing campaign” = 98,000 EUR – 100,000 EUR = -2,000 EUR 

Now, you may be thinking that this seems quite difficult. You’re not wrong – I made up a question on the harder end of the spectrum for sure. But, the difficulty of the question is NOT in the Math – the most difficult calculation above is 30*5,4 (which can be simplified to 30*5 + 4*3). The difficulty really stems from the following:  

1. The amount of information provided
2. Number of variables & steps to keep track of
3. Knowing what variables to use how / what to calculate 

Tips for how to nail the Quantitative Section of MBB interviews 

A lot of the conventional wisdom (and coaches) out there recommends candidates to spend a considerable amount of time practicing math. Many even highlight the importance of mastering being fast at simple calculations (with tons of tools out there to help you practice this. I would argue that for 80-90% of candidates, practicing ‘quick’ math is NOT an effective use of your time (because your skill level is likely already sufficient). 

Below are the steps that helped candidates I’ve coached in the past become rockstars at the Quant-Section of MBB interviews: 

1. Create a standard structure you always utilize when presented with a Quant-question
   • Using this structure, you will identify the relevant information provided, capture the variables and build your approach  
2. Outline the calculation steps required to reach the final solution BEFORE you calculate anything
3. Remember to finalize the solution by reacting to the outcome in relation to the broader case
   • E.g., in the example quant-question above – even though the marketing campaign was net-negative, a key insight from the information provided was that the control group was buying less frequently over time. Why this is, will need to be analyzed.
4. Practice real case quant-questions with a sparring partner (e.g., coach or friend), utilizing the steps above 

Practice real case quant-questions with a sparring partner (e.g., coach or friend), utilizing the steps above 

Before I outline tips for how to effectively address each step above, let me reiterate why these are the steps that best prepare you for the Quant-Section. In my +100 interviews as interviewer for McKinsey, the two most common reasons I’ve seen candidates flop on in this section are:  

1. Fail to capture a crucial piece of information needed to arrive at a solution (e.g., don’t capture all required variables or the question itself / end goal function)
2. Start calculating numbers too quickly, resulting in losing track of the steps required for the full calculation (e.g., due to stressing over speed/accuracy of calculations performed) 

Each of these common pitfalls can be easily diminished by a standardized and structured approach, and practice of real cases. Let’s dig deeper. 

Create a standard structure for the Quant-Section 

Every Quant Question at an MBB interview will have certain common features: 

1. A goal function / final answer (what you’re calculating) - typically presented as a question
2. E.g., “What is the profit impact of the new marketing campaign.”
3. Variables (numbers) needed to arrive at a final answer
4. Calculation steps required to arrive at that final answer
5. A final step where you put your result into the broader context of the case

Whether in person or virtual, you should/will have a pen and paper available to you. When the interviewer presents you with the Quant Question, make sure you start filling out your standard structure/template. Please find an example of my own structure below: 

 Figure 1: Template example of a standard structure for the Quant-section of the MBB Case interview.

As you can see from the image, an effective standard structure for your Quant Answer will have 4 categories: A) Goal function; B) Variables provided; C) Calculation steps; D) Final Insights.

Note: typically I would advice you to put the calculation steps on a separate paper, with space between steps for actual calculations. Here it is presented on one page for conciseness.  

Why is this standard structure helpful? 

Having this standard structure ready to use in your mind (or even on paper) when entering the Quant-section of the case interview will reduce the cognitive load of organizing the provided information (thus leaving more brainpower for figuring out calculation steps and actual math). This also ensures that you won’t forget what you’ve actually been asked to solve for – the goal function (a surprisingly common failure in this section), nor forget to contemplate the potential insights of your answer.  

Most importantly though, this forces you to focus on the most important aspect of the Quant-assessment – figuring out the calculation steps needed to arrive at the final answer. Finally, it’s worth noting that if you have a robust structure for your approach and get into challenges mid-way through, it will be a lot easier for your interviewer to assist you back on track. 

Outline the calculation steps 

This is the hardest part of the Quantitative section, and the real test the interviewer is giving you: “Can you take a sizeable amount of information and structure it to arrive at a calculated conclusion?”. As mentioned above, one of the most common pitfalls in the Quant-section of an MBB case is to jump straight into calculations, before first outlining the needed steps to arrive at an answer.  

So, how do you get good at this? There are two key aspects to this: 

1. Practice, practice, practice!
2. Take your time, and don’t start calculating until you have a clear path to the answer 

Lets use the Quant-question above to examplify how you would arrive at the calculation steps in practice. 

1. You know that the end goal is to understand the ‘profit impact’ of the marketing campaign, but that the marketing impact is presented as spend (revenue)
   • Thus, note for last step – remember to subtract the cost of marketing to arrive at profits
2. You also know that you are provided with revenues (spend & frequency) for a Test and Control group on a per customer-basis – so clearly you’ll need to subtract the Control from Test
3. Thus, a logical first step would be to identify A) The revenues of the Test group (marketed customers); B) The revenues of the Control group; C) Find the difference
4. Then, since the information is provided on a per customer-basis, and you know you want to find the ‘total impact’, you’ll want to: D) Find total Revenue impact across the Test customers
5. Finally, you need to make sure to convert the revenues to profits by subtracting the cost of the goods sold and then the cost of the marketing itself 

Perhaps this looks easy. Perhaps it seems very tricky. Your reaction to it may give you a decent idea of your current practice level. If this seems hard, make sure to practice creating the calculation steps in more cases. Regardless, remember that most importantly, focusing your attention on devising these steps BEFORE you calculate anything will always lead to a better result. 

Get those extra interview points 

If you have nailed the steps above in your interview, and arrived seamlessly at the correct answer, you’ve likely already passed this part of the case interview. However, you can still increase your chances for an offer. This is where the ‘final insights’ you can draw from the answer comes in.  

In my experience, most candidates (+75%) that manage to get through the calculation itself are so relieved that it’s over that they fail to provide any reaction to the outcome. If you really want to impress the interviewer and maximize your Quant-score, you’ll want to show you can relate the outcome to the broader case. 

How do you do this? 

This is honestly quite simple in most cases. You just need to get out of calculation-mode and back into creative-mode. Lets reuse my example question from above: 

The final answer was that the marketing profit impact was –2,000 EUR. Here are a few final insight examples one could point out (in ascending difficulty): 

1. The client should consider tweaking their marketing campaign in some way, to ensure it has a positive return. Unchanged, this campaign loses money. (Obvious insight, minimum contribution)
2. The Control group customers were buying considerably less often during the 6-month testing period. This is worrying, and we should analyze what may have caused it – e.g., is it simply seasonality, or are we losing market share? (Medium-hard insight)
3. Although the Test group clearly increased total spend during the 6 month marketing campaign, they did so through a large increase in frequency, whilst avg. spend per purchase decreased significantly. We should analyze whether we could alter the marketing towards higher cost items to minimize the decrease in avg. spend (or even increase it), whilst maintaining the frequency gains (Medium-hard insight)
4. The marketing campaign focused solely on Germany, ignoring the Polish market. The client should be careful to also test future marketing efforts on the Polish market, as the same marketing messages may have different results in that market – the same might even apply in different German regions (Hard insight) 

With a full case, you may even be able to tie the results back to something meaningful that you learned in previous questions of the case. But, let me be clear, you are not expected to point out all (or even many) of the potential insights of your calculation answer. Providing even one insight beyond the obvious (this campaign loses money) will impress your interviewer and earn you those extra points. 

What if the math itself IS my biggest weakness? 

My opinion that the math of the Quant-Section is not the difficult part of it, is likely obvious by now. That said, I did admit at the start of this article that I believe this applies to over 90% of candidates – not everyone. So, you might be thinking, if the math itself is my biggest weakness, what do I do? First, make sure you are indeed separating ‘math’ and calculations from figuring out the calculation steps, and that your challenge truly is the calculation. If that is the case, my main advice is: 

Figure out which types of calculations you struggle with, and focus on improving those through practice.  

It is very unlikely that you struggle with all types of calculations equally. Maybe large numbers throw you off because it’s hard to keep track of the zeros. Maybe fractions or percentages are more challenging for you. Or perhaps your difficulty lies in estimating results when an exact calculation would take too long, such as dividing large numbers that don’t produce a clean result. Whatever your specific weaknesses are, start by identifying them. And once you have, practice, practice, practice!

1. There is a neat ‘mental math tool’ on Preplounge you can utilize: INSERT LINK
2. Don’t get too hung up on speed – aim to go through the 60 seconds without any errors
3. Use Youtube – there are many great videos that deep-dive on specific types of calculations (e.g., division) and how to make them mentally easier to tackle
4. Practice on the math in Cases – you can find numerous cases online and in the Case library of Preplounge