BCG Casey: Structuring Question Deep-Dive

As introduced in the article BCG Casey online guidebook, there are five confirmed basic question types in the BCG Casey: Structuring, Intuition, Critical-thinking, Quantitative and Summary questions.

In this first article in our Casey question type series, we will introduce the structuring question, or STR question for short. This article will cover all the basics of this question type, so you can nail this STR question in the real BCG Casey test.

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.

Article overview

Structuring (STR) questions are one of five confirmed question types in the BCG Casey test. Despite appearing only 1-2 times per case, STR questions are among the most important. This type tests your ability to think like a real consultant: breaking down complex problems, building structures, and identifying the right information to solve the case problem.

This article is the first in our deep-dive series on BCG Casey question types, and it serves as your comprehensive overview of the structuring question. 

We’ll cover what it is, how it appears in the test, and how to approach it effectively. You’ll also know four common traps, three preparation guidelines, and key differences between structuring questions in BCG Casey and traditional case interviews.

Now let’s get started with the overview of this question type.

Overview of structuring questions

STR questions require candidates to structure the case problem

These questions require candidates to devise a structure to break down the case problem and then use that structure to gather relevant information.

Usually the question is phrased “what kind of information do you need to gather”. On the surface, it sounds like a data-focused question. But deep down, it requires structuring and drawing a structure. 

Using a structure ensures that information is selected in a reasonable and prioritized manner, rather than based on intuition or guesswork, which are NOT encouraged in consulting.

For example, suppose the question is: “Your client is a coffee chain experiencing a decline in profits. What kind of information would you prioritize for investigation?” 

A strong candidate would NOT choose random data points based on gut feeling, as this risks choosing the wrong options and losing their ticket to the case interview round. 

A strong candidate would first structure the case problem based on the case information available, e.g., using a profitability framework, breaking it down into revenue and costs. From there, they might then go further and divide revenue into number of customers and average spend per customer, and costs into fixed and variable costs.

STR questions appear 1-2 times per case

Structuring questions typically appear 1-2 times in a single case. In some rare cases, there may be up to three. But overall, the number of structuring questions is relatively limited compared to quantitative (QNT) or intuition (INT) questions. While less frequent, structuring questions are still crucial, because BCG evaluates candidates on each question in the Casey test. 

Among the structuring questions in a case, no two are identical. They may differ in question format, content, or both. For example, within the same case, one question might ask you to articulate your thoughts in a short written response, while another might require you to select from a list of options.

... with up to a 90% chance of being the FIRST question

In real BCG Casey tests, structuring questions typically appear at the beginning of the case. When starting the case, you are first provided with the case context, including the client’s situation, their problem, and business objectives. This is immediately followed by a structuring question, asking the candidate to identify the type of information that should be gathered for further analysis of the problem.

The placement of the structuring question right after the case context is intentional. It mirrors the problem-solving process used by management consultants on actual projects. They always start with the structuring of the case problem, which informs them which information to look for. 

STR questions have 2 main sub-types

Structuring questions in the Casey test generally fall into two main sub-types: normal structuring and estimation. 

At first glance, these two may seem quite different - the normal structuring focuses on breaking down and structuring a problem, while the estimation centers on estimating a specific number or quantitative variable.

However, both types share the same underlying requirement: the candidate must break down the given challenge and construct a logical structure to guide their thinking.

Normal structuring questions ask candidates to break down a business problem into manageable components. The prompt given above, “Your client is a coffee chain experiencing a decline in profits. What kind of information would you prioritize for investigation?” is an example of a normal structuring question.

Estimation questions, on the other hand, require candidates to break down a certain number and construct a good structure to estimate. While the objective is numerical, the process still begins with structure. 

For example, “One of the most important impacts of this innovative promotion is to get people talking about GR (our client). Your leader has asked you how you might estimate the dollar value of this increased popularity to GR.”

If you don’t draw a structure, it’s nearly impossible to estimate a number that falls outside your common sense or everyday knowledge, in this case, “dollar value of increased popularity from promotion”. Without a clear breakdown, you’re left guessing, and in a consulting context, that’s a major red flag. 

By breaking down something unfamiliar into smaller components that you are familiar with, you create a logical pathway toward a reasonable estimate.

Since normal structuring makes up the majority of STR questions in the BCG Casey test, the next two sections will be dedicated specifically to this sub-type, focusing on question format and content in greater detail. Following that, we will allocate a separate section to estimation type in structuring questions, to highlight the key differences between the two sub-types within the STR category.

Question format

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

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

And there are four components commonly found in STR question prompts. 

1. (Optional) Lead-in sentences with question-specific information 

At the start of the question text, there may be 1-5 sentences providing additional case context directly related to the question. They introduce new facts, updates that are critical to answering the question correctly. Hence, candidates must draw from two sources of information:

• The case context presented at the beginning of the case

• The add-in information embedded in the STR question itself

2. (Always present) One imperative sentence stating the task

This component is the core of the question. Every STR question includes a clear directive telling the candidate what they must do - typically to identify what information is needed. 

3. (Always present) One instruction sentence defining the answer format

Every STR question also specifies how the candidate should respond. Common instructions include 

• Select all that apply

• Select the fewest possible items

• Select three / four / … 

4. (Optional) A note providing technical clarification

Some STR questions include a brief note offering additional clarification relevant to the question. These notes often explain specific terminology or provide formulas needed to answer the question. 

In the example given, the note helps ensure that candidates draw a structure around the correct definition of “expected profit”, which may vary across different industries and firms.

... followed by one of two possible answer formats.

Structuring questions in the Casey test appear in two main answer formats: multi-select format and long-text format. 

Candidates will never encounter the single-select or short-text format for STR questions, because inherently a structure cannot be described by one option or a number. 

Multi-select format presents a list of options and asks candidates to select those that are most relevant for analyzing the case problem. STR questions in this format typically appear early in the case - often immediately after the case context is introduced. 

Long-text format asks candidates to type out a short written response, explaining what information they would seek and why. STR questions in long-text format can appear at any place in the case, but often in the middle or end of the case. 

Each format is paired with a distinct interface in the Casey test, which will be introduced below.

Multi-select format (MS)

The MS format interface is displayed as shown in the images below.

In MS format, candidates are asked to select multiple correct answers from a list of 4 to 10 options. 

These questions always give candidates a clue as to how many options to select. Sometimes they give an exact number, like “select the three best”. But sometimes it is just a hint, like “select all that apply” or “select the fewest possible items”. These are more difficult because no exact number is given. Candidates must rely on their judgment to determine which options are selected. Typically, around 30%–60% of the options are correct.

After carefully reading the question text, candidates click the “See options” button to reveal the list of options. In questions with up to 10 options, these choices may take up the entire screen. The imperative sentence remains pinned at the top of the screen for reference. If you'd like to see the whole question text again, candidates can click the “Collapse” button in the top-left corner of the screen.

Candidates simply click on the option once to choose it. If you change your mind, click it again to reject. You can freely click and unclick as many times as you'd like while deciding. If you want to reset your selection entirely, click the “Clear selection” button.

Once you’ve made your final selections, press the “Done” button to submit your answer. Important: after clicking “Done,” your answer is locked in - you cannot go back or make any changes, so be sure before submitting.

Long-text format (LT)

The MS format interface is displayed as shown in the images below.

In the long-text format of structuring questions, candidates are required to express their ideas in short paragraphs. Unlike the multiple-choice format, long-text responses test your ability to apply structured thinking and communicate clearly in your own words.

The answer to this format is typically limited to three to four sentences or lines. This word limit is not just a recommendation; it is an instruction explicitly stated in the question prompt. BCG expects a short and concise response from candidates.

The interface for long-text format is relatively straightforward. Candidates will type their answer in a message box resembling those used in common messaging apps, such as WhatsApp or Messenger. 

Note that pressing “Enter” does not submit the response - it simply moves the cursor to a new line, just like in Microsoft Word or Google Docs. To submit an answer, candidates must click the send icon located on the right-hand side of the message box.

The platform does not support special text formatting such as bold, italics, or bullet points. If candidates need to emphasize something, use CAPITALIZATION or hyphens “-”. 

For example:

“To analyze the client’s decline in revenue, I would first break it down into TWO main components - VOLUME and PRICE.”

While the message box technically allows up to 1,500 characters, candidates will rarely, if ever, need that much. BCG has clearly stated in the instruction sentence that strong answers should be brief, focused, and limited to 3–4 sentences or equivalent. Writing more than necessary could signal a lack of clarity or structure.

To get a strong answer, you must draw on both the immediate context of the question and any relevant insights from earlier parts of the case. 

Question content

In this section, we focus on a key component of structuring questions in the BCG Casey test: the imperative sentence - the part of the question that directly tells candidates what to do. 

STR question is usually worded to prompt the candidate to gather the information.

In most STR questions, the imperative sentence prompts the candidate to identify, select, or explain what information is needed to address the problem. 

On the surface, these instructions may appear to be simple data-gathering prompts, but in reality, they demand structured thinking and often require candidates to build a structure before selecting or explaining any data.

Here are some common examples of imperative sentences in the multi-select (MS) format:

• “Select the fewest possible items from the list below that are necessary to determine…”

• “Which of the following data would be most helpful?”

• “Select the top four data points required for this analysis from the list below.”

• “Choose the four most relevant pieces of data needed to estimate…”

• “Which three of the following topics would you prioritize for investigation based on the context you have been provided?”

In the long text (LT) format, the imperative sentences take on a more open-ended tone, but the logic remains the same:

• “To structure the decision-making process effectively, what are the three primary questions that need to be addressed to help…?”

• “In two to three sentences, identify the key data you would request to verify that…”

Let’s take an example of the Trial Case: Spicy Ramen Truck.

The case centers around declining revenue at Brother Hai's ramen truck. 

In Question 1, the structuring question asks the candidate to identify information that would help investigate the cause of this revenue drop.

At first glance, it looks like a typical data request. But a strong candidate would recognize that they need a structure to answer it. The structure helps ensure the selected data points are comprehensive, targeted, and actionable - far more effective than relying on guesswork.

The answer choices provided are branches or sub-branches of a pre-defined structure.

In this section, we focus on multi-select (MS) format questions, where answer options are provided by the system, unlike in long-text format where candidates must craft their own response.

Each MS-format STR question comes with 4 to 10 answer choices, and typically only 30–60% of these are correct.

These options are not random - they represent branches, sub-branches, or even sub-sub-branches of a pre-defined structure. 

Hence, the candidate can use the answer choices themselves to find out the underlying structure, and then, based on that to select the correct options.

Let’s take the case of Brother Hai's ramen truck as an example.

Looking at the options, we can see clues hinting at a structure splitting Hai's revenue across different variants (i.e., products). Within each variant, you can further break revenue into: Price per portion and Number of portions sold (or Number of customers * Average number of portions per customer)

In an ideal scenario, the correct options would be MECE - mutually exclusive and collectively exhaustive - covering all branches of the structure without overlapping. 

However, in the real test, that’s not always the case.

Candidates are expected to find 'the best options' that cover as much as possible of the structure. You can recognize this expectation by looking at how the structuring question is worded:

• “Top four data points”

• “Four most relevant pieces of data”

• “Three topics you would prioritize for investigation”

• “Select the fewest possible items…”

This is because for some questions, the options available do not cover every branch of the structure. And this is not uncommon in the BCG Casey.

The objective in STR questions is to choose the FEWEST number of options that cover AS MUCH of the issue tree as possible, not to cover every part of the issue tree.

Hence, based on the structure, we can identify that the correct answers are “The prices of different versions of Hai's spicy ramen” and “Number of people buying different variations of his spicy ramen”.

Equally important as identifying the correct answers is eliminating the incorrect ones. Understanding how wrong options are crafted can help you avoid traps. 

In general, incorrect options fall into one of the following four categories:

1. Too broad 

“Too broad” means the option covers not only part or all of the structure, but also strays out of it and covers irrelevant information.

For example, if the problem concerns the profit of one product line, but Option A covers revenue and cost across all three product lines, then Option A is too broad.

2. Too granular

“Too granular” means the option covers a too-small part of the structure that is already better covered by another option, provided that the higher-level option does not stray out of the structure.

For example, if the problem concerns the profits of a company, Option A covers revenue of all of its 3 product lines, while Option B, C, D each covers only one product line, then Option A is the correct answer.

3. Irrelevant

“Irrelevant” means the option does not relate to the problem in any way, shape, or form – i.e., it is entirely out of the structure.

For example, if the given problem is about revenue and the option is about cost.

4. Wrong data

“Wrong data” means the option “looks” correct, but the unit or the timeframe is wrong. 

For example, if the problem is set in 2025 and the data is from 2023.

Taking the case of Brother Hai's ramen truck, we have:

• Possible changes in customer behavior (such as health food trends): This option is too granular because this should be examined after we found which variant (i.e., product) causes the decline in revenue.

• Whether there's a new day-time pasta shop opening up to compete nearby: This option is irrelevant because a day-time pasta shop should not affect the business of a night-time ramen truck

• The ingredients that go into the sausages that he buys: This option is too granular because this can only be examined if we found the root cause to be the sausages.

Estimation in structuring questions

Question format

Estimation-type STR questions in BCG Casey share the same four-part structure as normal structuring questions. 

Each question text typically includes: 

• Lead-in sentences with question-specific information

• One imperative sentence stating the task

• One instruction sentence defining the answer format (either MS orr LT format)

• A note providing technical clarification

Compared to normal structuring, the note part is almost always present in estimation questions. 90% of such questions include a definition or formula related to the item being estimated. These notes appear to guide the candidates’ thinking and help them identify the right inputs.

Let’s take an example of the following case.

The case asks candidates to estimate the change in expected profit if GR runs a promotional campaign. 

The note provided defines expected profit as “the probability-weighted average of all possible profit outcomes”. It also gives an illustrative example of how expected profit is calculated using the given numbers. 

Answer formats for estimation-type questions remain consistent with other normal structuring questions.

The candidate will either:

• Select options from a list of 4-10 options that are necessary to make an estimate (MS format)

• Write a short answer explaining what information they’d need to make an estimate (LT format)

Question content

Estimation structuring questions are often phrased around the idea of “what information is needed to estimate”. 

For example:

• MS format: Which of the following items are necessary to determine …? 

• LT format: Your leader has asked you how you might estimate the dollar value of ... Please share your thoughts on this in 3-4 lines.

Let’s take the case above as an example. Candidates are required to gather information to estimate the change in expected profit if GR runs the promotional campaign.

At first glance, these seem like quantitative questions, but in reality, they are still structuring questions. This is because before you can estimate anything, you need the right structure.

The right structure lets you take something unknown (in this case, the change in expected profit) and break it down into smaller parts connected by math operators (+, -, *, /).

Each part should either be directly given, inferable, or a number that can be reasonably estimated. So you can arrive at the final answer. 

This process - breaking down an unknown into multiple components you know - is the essence of structuring. That’s why estimation type falls under structuring questions in BCG Casey.

Note that candidates are not required to do the actual math or estimate any specific numbers. Their task stops at identifying which pieces of information are needed, NO need to estimate them. 

That’s why the quality of your structure matters more than your ability to crunch the numbers.

How to prepare for structuring questions?

Understand the consulting problem-solving fundamentals

Problem-solving is at the heart of management consulting. It refers to the ability to break down complex, ambiguous problems into structured, manageable components. 

This is the exact skill that structuring questions in BCG Casey are designed to assess. If your problem-solving skills are strong, you’ll find it much easier to build a solid structure, and from that, choose the right information to answer the question.

More importantly, if you’re aiming to become a consultant, problem-solving skill is non-negotiable. Every part of the recruitment process, from CV screening to BCG Casey to the final case interview, will test this ability. 

To develop this ability, try to apply it to everyday problems and cases you read in business publications. Practicing with the BCG Casey Simulation from MConsultingPrep is also an effective way to build real-time problem-solving skill.

Grasp thoroughly the MECE principle - mutually exclusive, collectively exhaustive

MECE (pronounced “mee-see”) is short for Mutually Exclusive, Collectively Exhaustive. It is a principle used in consulting to break down items into small pieces. 

“Mutually exclusive” means no overlap between each piece, while “collectively exhaustive” means all the pieces combined form the original item without any gap.

Why does it matter in structuring questions? Because your structure must be MECE to choose the correct options. When your structure is MECE, its branches must not have any overlap while covering ever aspects of the case problem. So it helps you avoid incorrect answer choices that are either “too broad” (covering too much, including irrelevant data) or “too granular” (covering too little of the structure, when broader options exist).

A useful tip: Apply common case frameworks to create a MECE structure more quickly, when applicable. 

Frameworks are pre-existing and intuitive templates used to draw structures.

Many frameworks, like the Profitability Framework, 4Ps, etc., are already MECE by design. So using those in structuring questions and case interviews is a good place to start, at least you don’t have to worry about the MECE thing.

However, be cautious: blindly applying inappropriate frameworks can be counter-productive. For a more detailed guide and practice exercises, visit this article on MECE by MConsultingPrep.

Master common case interview frameworks

Frameworks are templates used to break down and solve business problems in case interviews. They can also be used in structuring questions in BCG Casey to break down and structure the case problem.

Some common frameworks are: 

• Profitability Framework

• Business Situation Framework

• External vs Internal

• Qualitative vs Quantitative

• Cost vs Benefits

• McKinsey M&A Framework

• 4P and 7P Frameworks

• Porter Five Forces Model

• 2×2 Matrix (e.g.: BCG Growth-Share Matrix)

• SWOT Analysis and more

These frameworks can be off-the-shelf or highly customized for specific cases, but NOT all the cases. 

Business cases in Casey are similar to real business problems, where frameworks need a lot of customizations to be useful. 

So while frameworks are helpful, not every case will align perfectly with one. And applying frameworks without careful consideration can lead to ineffective outcomes.

To ensure your structure is both relevant and insightful, it's safest to take the initiative to build it yourself directly from the case context. Doing so not only helps you address the problem more accurately but also demonstrates your strong ownership and a sharp problem-solving mindset - two qualities that every consulting firm looks for.

If you want to dive deeper into frameworks, MConsultingPrep has an excellent guide available here: Case Interview Frameworks.

Structuring questions in BCG Casey vs. Case Interviews

Question format

The testing environment of case interviews are more interactive than BCG Caey. In case interview, the candidate and interviewer engage in a live, spoken discussion about the case. The candidate is free to ask follow-up questions, clarify case details - just like in a real client meeting. 

In contrast, in the BCG Casey candidates interact with a chatbot through a chat-style interface, similar to messaging apps. Because the chatbot is pre-scripted, all the information about the case are pre-determined and shown at once. The candidate cannot ask follow-up or clarification questions. 

This fundamental limitation leads to some differences in how structuring questions are presented and answered between case interviews and BCG Casey.

In a case interview, the candidate answers structuring questions based on the case information they have previously gathered from the interviewer. 

The case information is not provided all at once but unfolds gradually through an ongoing exchange between the interviewer and the candidate. Candidates can ask clarification questions, request more context, or confirm understanding at any point. How much information is uncovered depends on the candidate’s capability to ask the “right” questions, making the case more flexible and adaptive to each individual’s approach.

While in Casey, the candidate answers structuring questions based on provide-all-at-once information from a scripted chatbot. 

Since the chatbot is fully scripted, the conversation is fixed. All the information and questions are provided upfront - no further clarifications can be requested. The candidate can only answer the questions given by the chatbot.

In case interviews, since questions are delivered verbally, the question text is usually just one concise sentence that gets straight to the point. For example, “What information do we need to…?”

But in Casey, the full context and question are displayed on-screen, which allows for a longer question text. As mentioned earlier, these often include four components: lead-in sentences, an imperative sentence, an instruction sentence, and optional notes.

In terms of answer format, candidates in case interviews respond verbally and may also sketch their structure in a notebook to explain it to the interviewer. 

In Casey, however, STR questions appear in one of two formats: Multi select (MS) - where the candidate selects items from a list, or Long text (LT) - where they write out a short answer.

Question content

Despite the differences in question format, structuring questions in case interview and Casey both test the candidate's ability to structure the case problem.

In both settings, the questions often appear under the guise of data-gathering, phrased as “What are the key factors / criteria to consider...?” or “Which information would help us...?” While they may sound like simple data questions, answering them properly requires the candidate to build a MECE structure. This structure then becomes the basis for identifying which pieces of data are truly relevant.

Structuring questions in case interview also have two sub-types: normal structuring and estimation. 

Normal structuring: These are typically asked at the beginning of a case, after both the candidate and interviewer are on the same page about the case. The candidate is expected to break down the problem and identify information needed to analyze it.

Estimation: These questions tend to appear in the middle or later stages of a case. They require the candidate to estimate a quantitative figure, such as profit, revenue, or cost. 

This is where the difference begins.

In BCG Casey, the candidate doesn’t need to construct a structure entirely on their own. They are supported with available answer options. These options often reflect branches or sub-branches of a hidden structure already designed for the case. In other words, the case structure is pre-defined. And the candidates’ task is to recognize it and select the right options.

In case interviews, however, candidates are given minimal hints. They must build their own structure independently, drawing only from the information they actively uncover through conversation with the interviewer. As such, multiple valid structures may emerge depending on the candidate’s line of inquiry and assumptions. And it’s up to the candidate to define the structure and justify theirs.

Another important difference lies in the estimation questions. 

In BCG Casey, estimation within structuring questions usually stops at the structuring phase. Candidates are expected to identify the inputs needed to estimate a number, not to perform the actual calculation.

While in case interview, estimation questions require the candidate to go further, not only constructing the structure but also estimating each input and calculating the final number.