BCG Casey: Quantitative Question Deep-Dive

In this second article of our Casey Question Type Series, we’ll take a deep dive into the Quantitative question, also known as the QNT question. 

While the name may suggest simple math drills, QNT questions are far more than that. They test whether you can think, analyze, and calculate like a real consultant - navigating complex datasets, identifying what matters, and delivering answers fast and accurately.

This article will serve as your complete introduction to the QNT question type.

Use the links attached to each section in this article to read more about each specific topic (you will need that), as well as the banner on the right to learn more about our product - the BCG Casey All-in-One.

Please read the main article on the BCG Casey if you haven’t done so.

Article overview

Quantitative (QNT) questions are among the most common question types in the BCG Casey chatbot - appearing 2 to 3 times per case. These questions challenge you to do more than just math - they ask you to think like a consultant: making sense of exhibits, choosing the right data, and calculating under serious time pressure.

In this article, we’ll walk you through everything you need to know about QNT questions:

• What QNT questions really test and why they’re more than just number crunching

• How questions are presented and answered

• What types of exhibits and data to expect

• The four common types of calculations used in the test

• Six practical preparation and in-test tips to improve accuracy and speed

Now, let’s dive in and take a closer look at QNT questions.

Overview of quantitative questions

To better understand how quantitative questions work in the BCG Casey test, let’s look at a real example from the Trial Case: Spicy Ramen Truck. You can take the full test at this link. 

After structuring the case problem in Question 1 (you can visit our Structuring Question article to learn more and see an example of this question), you and Brother Hai begin analyzing the data together in Question 2 - what we refer to as a quantitative question. Now, we’ll break down this question type through this question.

QNT questions require candidates to do some calculations to get to the right answers

Quantitative (QNT) questions in the BCG Casey test require candidates to crunch numbers and carry out calculations in order to arrive at the correct answer. 

For example, question 2.1 “Which variant of Brother Hai's ramen saw the largest percentage decline in portions sold in April, compared to Q1 average?”. To choose the correct option, you need to calculate the percentage decline in portions sold of each variant then make comparison. 

But don’t be fooled by the term “quantitative” - these aren’t simple math drills. Instead, they are designed to test whether you can analyze and reason with data in a business context, under serious time pressure.

To solve a QNT question, you’ll start by interpreting both the numerical data in exhibits (tables, charts, diagrams) and the textual information provided in the question prompt. Together, these two sources tell you what needs to be calculated, what formulas to use, and which numbers are relevant. 

In question 2.1, to choose which variant saw the largest percentage decline in portions sold in April compared to the Q1 average, we need data about the number of portions sold of all variants from January to April. 

Once you've understood the full context, you’ll carry out a series of calculations to arrive at the final answer - either by selecting the correct option (in single-select format) or entering it directly (in short-text format). Question 2.1 is an example of single-select format while the latter three questions are of short-text format.

Each QNT question often requires you to complete 2 to 4 calculation steps, with each step involving 3 to 5 smaller math operations (e.g., multiplication, division, percentage change). While the math itself is not advanced, the challenge lies in speed and accuracy under pressure.

QNT questions appear 2-3 times per case

Quantitative questions in the BCG Casey test are one of the most common question types, appearing 2 to 3 times per case, just as frequently as Intuition (INT) questions. 

If you’ve already tried the Trial Case: Spicy Ramen Truck, take a moment to count - how many QNT questions did you spot? 

A single QNT question might contain just one main question, or it can be broken down into 2 to 6 smaller sub-questions, known as mini-questions. These mini-questions can appear in both single-select (SS) format and short-text (ST) format; just take a look at question 2 in the Trial Case for a clear example.

Given the number of QNT questions in each case, candidates are expected to interpret complex data and analyze it both accurately and quickly under intense time pressure. 

Let’s take a simple calculation. Each Casey test includes 6 to 8 questions to be completed in just 25 to 30 minutes, meaning you have only 3 to 5 minutes per question on average. In the most extreme cases, that could mean solving 6 mini-questions in just 5 minutes - a true nightmare if you're not well-prepared.

… and are often used to “test” specific branches of a structure 

A quantitative question is often placed in the context of testing a specific branch within the structure drawn by the structuring question at the beginning of the case. Hence, candidates can typically expect QNT questions to appear soon after the start of the case.

In the Spicy Ramen Truck case, after structuring the case problem in question 1, we arrive at a structure like the one illustrated below. You can visit our Structuring Question Deep-Dive article to understand how this structure is built.

And the first QNT question follows immediately, aiming to test the “Number of portions sold” branch. 

This setup is designed to mimic the problem-solving process of management consultants on actual projects. In real consulting work, consultants gather and analyze a large amount of data to extract insights and draw meaningful conclusions - either to confirm or reject a hypothesis. And of course, as part of that process, they perform a great deal of calculations.

That’s why QNT questions are included in the Casey test and why they appear more frequently than many other question types. They are meant to assess your ability to think and solve math problems like a real consultant.

That said, in the Casey test, you’re not expected to extract insights. Your task is simply to perform the calculations and arrive at the correct number.

QNT questions are math problems within business context

QNT questions in the BCG Casey go far beyond simple number-crunching tasks. Instead of merely solving mathematical equations, candidates are expected to apply math within a real business context. That’s because the Casey test itself is built around a business case.

Take question 2.1 as an example. You’re not just asked to subtract or divide one number from another. Rather, you’re required to identify which ramen variant experienced the largest percentage decline in portions sold in April, compared to the Q1 average. This is a simple case of solving a math problem in a business context - you must interpret the data, calculate the relevant numbers, and determine which variant is contributing most to the revenue drop. 

Candidates often see these questions appear interwoven with Critical-thinking (CRI) or Intuition (INT) questions, making them more complex and layered.

To perform well, candidates need more than just strong calculation skills; they must also demonstrate sharp graph-reading abilities and solid business intuition - specifically, a foundational understanding of how different business concepts work in practice.

Most of the time, QNT questions are nearly always accompanied by exhibits 

In QNT questions, exhibits are charts, tables and diagrams.

This setup reflects the day-to-day work of consultants, who rely heavily on data visualizations to extract insights, test hypotheses, and support recommendations.

And in the BCG Casey, these visual aids are not just for reference; they often contain the key numbers and formula hints needed to perform the required calculations. 

Being able to quickly interpret and analyze these exhibits is therefore essential, not only to solve the question correctly, but also to demonstrate the data literacy expected in real consulting projects.

Question flow

Main questions come first, 

Each QNT question typically begins with a block of information, often in the form of text and data, that sets up the question problem. This introduction provides the context, figures, and any necessary background for solving the upcoming question. 

If there is only one question, it will appear immediately alongside the data. However, if the QNT task includes multiple mini-questions, candidates must first review the information and then click “Continue” to reveal each question one at a time.

These mini-questions are presented in a fixed sequence and cannot be skipped or rearranged. You must answer one before the next becomes visible. 

Importantly, the questions are logically connected, meaning that the answer to one question may be required to solve the next. As a result, a mistake early on can create a ripple effect, impacting the accuracy of all subsequent answers. Therefore, precision and careful reasoning are essential throughout the entire question chain.

… followed by one or some follow-up question(s) (optional)

In some cases, QNT questions may include follow-up questions or additional mini-questions that appear after you've submitted your initial answer. These only occur in the short-text format, where you're asked to input a number into an answer box. Follow-ups are triggered based on your response and serve to probe deeper into your reasoning or calculations.

After answering the main question, one of two scenarios may occur:

• You may be asked to briefly explain the reasoning behind your answer. For example: “Please explain briefly the reasoning behind your answer in Question ABC.”
  

• Or, the chatbot may ask you to break down the specific calculations or intermediate steps involved in reaching your answer.

In the second scenario, these follow-up sequences usually contain 2 - 4 questions, each guiding you through a single step of the required math. Your job is to perform the specific calculation requested, using the same data and assumptions from the main question.

In the past, candidates used follow-up questions as a clue about whether they answered correctly. A request for explanation typically signaled a correct response, while a prompt asking you to revisit specific steps might indicate a mistake. 

However, this is no longer a reliable strategy. Now, follow-up questions can appear regardless of whether your initial answer was right or wrong. You might be asked to explain your thinking or show each calculation step-by-step, even if your final result was accurate or inaccurate.

Question format

QNT questions begin with a question text displayed as a message,

Note: In this section, we focus on the question text of a QNT main question - the part that delivers most of the data and context needed to solve the problem. 

In the Casey test, quantitative question text is presented in the form of a message-style prompt, designed to simulate a conversation. 

And a standard QNT question text typically consists of four components, each serving a specific purpose.

1. (Always present) Context and data block

Every QNT question begins with a block of information that sets the scene and provides essential data. This section is always present and typically includes a description of the business situation, often written in several lines of dense text. The length can vary, but it usually requires strong reading comprehension skills to digest the context quickly and accurately.

In addition, this section almost always contains multiple exhibits (charts, tables, diagrams), sometimes embedded within the text, but often followed by a clearly separated section dedicated to data presentation. 

Efficiently interpreting these visuals is critical, as they contain the numbers needed for calculations later on. Mastery of both reading comprehension and data interpretation is key to navigating this portion of the question successfully.

2. (Optional) Statement on number of mini-questions

If the QNT problem includes two or more mini-questions, you may see a line stating how many calculation questions will follow. This helps set expectations about the length of the question. However, if there is only one question, this line will be omitted entirely.

3. (Always present) Main calculation prompt

This is the core of the question, clearly asking you to compute a specific number based on the context and data provided. Even if follow-up questions appear later, this line identifies the primary goal of the task.

4. (Always present in short-text format) Answer format instruction

At the end of the question text, you will always find a note specifying how your answer should be formatted. This includes rounding rules (e.g., to two decimal places, or to the nearest thousand or million), and unit sign omitted. Following these instructions precisely is critical - formatting errors may lead to a wrong answer, even if your calculation is correct.

... followed by one of two possible answer formats

In the BCG Casey, QNT questions appear in two distinct answer formats: Short Text (ST) and Single Select (SS). While the interface for each format is slightly different, the underlying logic remains the same.

Short Text (ST)

Short Text (ST) is the most common answer format in QNT questions, accounting for roughly 70–80% of all cases. 

In this format, candidates are required to calculate a specific result and input only one number into the answer box. 

The interface closely resembles that of long-text (LT) questions, with just one key distinction: short-text (ST) only accepts numerical input. You cannot type in words, symbols, or unit signs - only numbers are allowed.

Each question or mini-question provides a single answer box, and the chatbot will clearly specify how your answer should be formatted. Instructions typically include rounding rules, such as rounding to the nearest thousand or to a set number of decimal places. If your answer doesn’t follow the required format, the system will reject it and prompt you to submit a corrected version.

It’s important to note that pressing Enter will immediately submit your answer. There is no confirmation step and no option to revise your response afterward - once you hit Enter, your answer is final.

Single Select (SS)

Single Select (SS) is the less common format in QNT questions, appearing in roughly 20–30% of cases. 

In this format, candidates are asked to calculate a result and then choose only one correct option from a list of predefined answers. The interface and layout are similar to the multiple-select (MS) format, but with one critical difference: you can select only one option, and there is only one right answer.

Answer choices can take various forms, not just numbers, but also statements, objects, or business scenarios. All available options will be displayed directly within the answer box. You can scroll sideways to see all options when there's not enough space.

What makes this format especially tricky is that the moment you click on an option, your answer is immediately submitted. There is no confirmation step and no way to go back or change your response. 

Question content

QNT questions are worded to require candidates to read data and perform calculations

In the short-text format, QNT questions are usually easy to recognize. The question prompt often includes a direct instruction such as: “Based on the data provided above, calculate…”

This is a strong signal that you’re dealing with a QNT question. It tells you exactly what needs to be done - perform a calculation using the given data, then input only one number into the answer box. This format is straightforward and transparent in its expectations.

However, QNT questions in single-select format are less obvious. You might not see the word “calculate” at all. Instead, what signals a QNT question here is the large amount of data provided in the question text. The chatbot may simply ask you to select the correct option, without explicitly saying you need to do any math.

But to answer correctly, you will need to process the data, do the math behind the scenes, and then pick the option that matches your result. So while the wording may feel qualitative, the question still tests your quantitative skills, just in a more indirect way.

It’s important to note that QNT questions do not require much accounting or finance knowledge, as the calculations involved are generally straightforward. Even in cases where the question relates to accounting or finance, candidates are usually provided with the necessary formulas.

That said, to be on the safe side, non-business candidates should review basic financial terms such as: gross profit, net profit, profit margin, net present value (NPV), break-even point, and return on investment (ROI), as these commonly appear in business-related math problems.

Exhibits

The exhibits used for QNT questions are often relatively simple, compared to those used by the PST or Potential Test. 

Each QNT question usually includes 1 to 3 exhibits, including charts, tables, and diagrams. 

The most common types of exhibits include tables and charts, with charts typically falling into familiar types such as line charts, pie charts, and bar charts. These are visual types that most candidates will already be comfortable interpreting.

The data presented in the exhibits is intentionally kept clean and minimal, typically around 4 to 25 rounded data points (arranged in tables of about 2 - 5 rows and 2 - 5 columns) plus any necessary legend and footnotes. 

Calculations

In QNT questions, the calculations themselves are usually quite easy. 

Most of the time, the test uses clean, rounded numbers, not real-world messy data, but data specifically designed for testing purposes. This makes mental math and calculator use faster and more manageable.

In addition, you are allowed to use a calculator, just like in McKinsey’s Solve. BCG does not impose any restrictions on the resources you have at your disposal while taking the Casey test from the comfort of your own home. Plus, the test often allows a small margin of error, with answers typically rounded to the first decimal point (e.g., 12.3456 becomes 12.3). 

Despite the simplicity of the arithmetic, the real difficulty lies in finding the right numbers and the right formula. Up to 80% of the data provided is irrelevant or unrelated to the actual calculation. In most questions, you’re not given a formula, you have to figure out which formula applies and which data to plug in.

This is what makes QNT questions more than just number-crunching exercises. They test your ability to structure your thinking, filter out unnecessary information, and build a logical calculation path.

To navigate QNT questions effectively, it's not enough to spot the right data, you also need to know what kind of calculation to apply. Below are the four common calculation types you’ll encounter in the Casey test.

1. Basic operations (+, - , *, /)

There are four basic mathematical operations used in QNT questions: addition (+), subtraction (−), multiplication (*), and division (/).

This is the most frequently appearing type, found in every QNT question.

However, note that stand-alone basic operation questions are rare. Instead, these operations are usually combined with other types of math, such as percentages, probabilities, or weighted averages, to form more complex calculations.

2. Percentages

This type appears almost as frequently as basic operations and is a core component of many QNT questions.

There are two sub-types within this category:

• Simple percentages: These involve calculating a percentage, ratio, or fraction between two items.

• Compound percentages: These require performing multiple percentage steps in sequence. For example, calculating a percentage of a percentage.

3. Probabilities

This type involves the calculation of probabilities of possible outcomes.

Candidates may be asked to work through simplified business scenarios where probability logic must be applied to interpret data and estimate chances or risks.

4. Weighted averages 

In this type, candidates are required to calculate averages where each value has a different “weight.”

There are two sub-types:

• The sum of all weights equals 1: This is common when weights are expressed as percentages or proportions.

• The sum of all weights does not equal 1: In this case, an extra step is needed to normalize or adjust the weights before calculating the final average.

How to prepare for quantitative questions?

Learn to filter data quickly

In the BCG Casey test, candidates can easily feel overwhelmed by the large volume of datasets, charts, and tables included in each question, especially under time pressure. These visuals are filled with information, but not all of it is relevant to the actual calculation.

The key to solving this challenge lies in quickly identifying the truly relevant data. This skill can be built through two actions:

(1) Understand what problem you’re trying to solve before looking at the data, so you can focus on what truly matters.

For example: If the question asks for Year 2 profit growth, there's no need to analyze unrelated metrics like headcount or pricing breakdowns.

(2) Master your reading speed using skimming and scanning techniques. These are essential for navigating long question texts efficiently

Skimming means quickly looking over the text to get the main point. While scanning means searching for specific information.

For example: If you're asked about market share in Asia, scanning lets you skip other regions and jump straight to the relevant exhibit section.

Familiarize yourself with 4 calculation types

You should be comfortable with the four main types of calculations used in the Casey test: basic operations, percentages, weighted averages, and probabilities.

The best way to get there is through progressive practice:

First, practice each type individually in a “normal context”, where your only goal is to solve the math. These exercises will help you understand the formulas and logic behind each concept.

Then, transition to solving these types in a business context, similar to how they appear in the Casey test. We recommend practicing with our BCG Casey Simulation, which mirrors the real test format and forces you to apply math within a business scenario.

If you’re not a native English speaker, make sure you’re also comfortable with math-related terminology in English. Practice with phrases like gross margin, weighted average, or year-over-year growth, so you don’t waste time decoding vocabulary during the test.

Practice mental math skill

One of the biggest challenges in QNT questions is performing calculations under time pressure.

On average, you’ll have 3 - 5 minutes per question, so mental agility is crucial. You should be able to do math in your head without using a calculator.

To build this skill, we recommend you practice regularly with Numerical Reasoning Tests (in the Aptitude Test Package) and Comprehensive Math Drills from MConsultingPrep. In daily life, avoid relying on a calculator for simple math. Instead, try doing it in your head, such as estimate percentages, compare ratios, or break down growth rates mentally.

Over time, these habits will make you faster, more confident, and more efficient when facing Casey’s time-constrained environment.

3 useful tips during the test

Even with solid preparation, how you manage your process during the test makes a big difference. Below are 3 practical tips to help you stay accurate, efficient, and in control while working through QNT questions in the Casey chatbot.

Don’t panic when seeing follow-up questions

If you encounter follow-up questions, don’t jump to conclusions or panic as they’re not always a sign that you made a mistake. Instead, stay calm and take this as a prompt to double-check your calculations.

To respond correctly to follow-up questions, begin by reviewing whether your overall approach to the question was sound.

If your approach seem correct, go through each step again and recheck the specific calculations within your process.

Follow-up questions often point you to the exact part that needs reviewing, so use them to your advantage.

Keep calculations clear & concise

Keeping your calculations organized and clearly structured is key, not just to avoid making mistakes, but also to quickly identify the source if something goes wrong.

To do this well, define your approach upfront: What steps will you take to solve the question? What does each step involve?

Write down each step logically on paper, show each calculation clearly, and underline or circle your final answer so it stands out.

When steps are neatly laid out, it’s easier to stay on track and much easier to troubleshoot if a follow-up question appears.

Note down key points after each question

Once you’ve completed a question, take a moment to jot down key data points or useful insights before moving on. This serves two purposes:

First, it helps you organize your thought process and ensure that your answer flows logically.

Second, it acts as a reference point for later questions, especially because many QNT questions are interconnected and build on earlier ones.

If you answer incorrectly and a follow-up question appears, these notes can also help you quickly retrace your steps and locate where you went wrong.

Taking notes isn’t just about memory, it’s about setting yourself up for consistency and accuracy throughout the case.