Structuring & qualitative/quantiative part of case

Hi, here some tips and tricks as well as explanations of what the interviewer expects:
 

 1. General Approach to Structuring a Problem

Instead of relying on rigid structures, I encourage you to learn to have flexible thinking. Chances are there will be cases that are unusual and cannot be solved with any learned framework. For example, the one you shared I have never used.

However, there is a general convention that is expected of you, which is to go as broad and as deep as possible into all elements that can be relevant to the problem (qualitatively). 

Here a few steps to help you:

1. Brainstorm Broadly:

-       Dump Ideas: Write down all possible factors related to the problem you can think of (e.g., prices)

-       Group into Themes: categorize the factors into logical groups that are more higher level (e.g., prices and volumes would relate to a higher category of revenues)

2. Increase Breadth and Depth:

-       Expand topics: Think about all stakeholders involved in the demand and supply side of the situation (e.g., if the problem were about a football match you would think about both teams, the referee, and the audience.. ) and environmental factors (e.g., the football field).

-       Dive deeper: ask questions for each of the elements you have identified and for all stakeholders (what do they want, how do they want it, why, where, when…)

-       Make sure you are deriving a tree structure, such that higher themes (e,g, profits) sit on top and branches diverge to deeper levels (e.g., the color of the football field)

Your goal is to come up with as many elements as possible, organize them in a hierarchical way and only then move on to double checking which elements are the most relevant to the problem at hand and hypothesize why.

3. Hypothesize:

-       For each topic in your tree simply ask “how does it impact the problem at hand?”, and the answers will be your initial hypothesis (e.g., revenues are down in football matches because someone changed the color of the field to orange).

-       Rank those that are most like.

At this point you should have a tree with great breadth and depth, and plenty of hypothesis for all elements.

4. Think About Data:

-       Identify Data Needs: for each of your hypothesis you will have very specific data needs, think of those and be as specific as possible (e.g., share of tickets sold via each existing channel within the UK market)

 5. Guide the interviewer:

-       Present your logic tree from the high level buckets to the deeper ones, then the hypothesis from most likely to least, and then the data you would need to test each. Then ask the interviewer for their view.
 

2. Some techniques to help you practice:

1. Mind Mapping: Regularly create mind maps for different business problems to enhance your ability to think broadly and categorize effectively.

2. Flashcards: Use flashcards with various business scenarios to practice grouping ideas and forming themes quickly.

3. Timed Drills: Set a timer and practice structuring problems in 5-10 minutes to improve speed and clarity.

4. Brainteasers: Try doing the same with brainteasers and topics that are abstract and not about pure financials, you will get very strange cases in real interviews that have nothing to do with math and may be purely qualitative.

3. Qualitative vs. Quantitative Approach in Cases

Refer to steps in #1.

What the interviewer wants to see is that if you were alone with a client and the client had a question you would find your way to the answer. To do that, the interviewer wants you to go very broad, very deep, and have plenty of hypothesis that are very concrete. The data requests you will have for clients will also be extremely precise.

So, guiding and discussing with the interviewer at a qualitative level is really the most important part of the case. Avoid asking generic questions such as “do you have any data?” 

Good luck!

Hi there, 

Let me start with a broader explanation, in case this is helpful for anyone as a way of thinking. 

To me, questions such as “What is the general approach to quickly structure a problem?” sound exactly like “How to become fit?”, meaning there is not any simple / single answers. 

But, just like in the analogy, structuring is like building a muscle: the more you practice it, the more clear it will be become and eventually easier to do. Some people might have a more natural tendency to structure (or learned through their school, jobs), while for others it requires more active learning. 

With that long intro, let me try to more specifically answer your questions: 
1. No, I don't think there is a standard first level of buckets that would work well for multiple problems. I would love to be proven wrong, but if that would be the case, no one would even need consultants anymore. 

2. Yes! You can do drills for example where you just try to structure the problem at hand. Have an approach in mind (my favourite: what should I absolutely know in order to prove / disprove a hypothesis regarding the stated problem or situation), practice it and pay good attention at the feedback you receive or the stated solution (if you solve from a case book). 
Be inquisitive and critical by asking: what are the differences in approach? What are the differences in results? Why were those considered good answers? How could I think of them in the future? What was my approach missing? 

3. The first part of the case should be focused on understanding the problem and laying out the plan on how you would solve it > this rather means in which areas you would look to investigate rather than what data you need.  

In real life, consultants do not go at the client and just ask for all the data they can think of. They have hypothesis of what matters when solving the clients' problem, make a plan on how to approach it and only then ask for the data - if this sounds similar to how one would approach a case, trust me, it's not a coincidence!

To conclude, it sounds like you have very good and normal questions for someone starting their interview prep journey! I would recommend you maybe have a quick chat with some current or former consultant, so you can get a first-hand impression how structuring works for them, in real life and what ultimately drives the case approaches. Whatever you do, please don't learn structuring frameworks by heart :) 

Hope this helps, 

Ariadna 

Issue 1: General approach to quickly structure a problem

I would advise against seeking that harmony.

Instead, you should better understand what a framework is supposed to do. When you have clarity on that - you will find a sudden ease in creating a bespoke framework, on-the-fly.

A framework is supposed to provide you with an answer to the client's problem/question.

Are you entering a new market? Then answering each bullet in your framework should give you Yes/Nos.

Are you diagnosing a profitability decline? Then answering each bullet in your framework should help you validate/invalidate possible causes of the decline.

A “structure” is merely the art of bucketing your ideas & hypotheses to make it easy to communicate, easy to understand, and thorough in essence. Happy to help you out with the right way to put this approach into practice.

=================

Issue 2: Practicing alone

Practicing with a peer in a timed fashion will push your mind into thinking urgently and will build interview momentum. But the real practice is after the session when you spend an additional 1-2 hours per case analyzing it in many ways.

In that post-practice analysis you try creating alternate realities of the case - alternate frameworks, alternate storylines, alternate numbers, alternate outcomes. This will show you how the same case prompt could have gone in 100 different directions. It will give you the ability to think in multiple ways quickly. And it will give you the confidence to feel right and solve a case well.

=================

Issue 3: Qualitative vs Quantitative

Most fresh casers think of a case as a math exercise to a business question. But solving business problems is a combination of art and science.

If you have faced pushback on jumping into math too quickly, then it means you are pushing too hard on the math aspects without marinating yourself into the business aspects of the problem. Often, without exploring the Qual aspects - you may end up doing excess and unnecessary math.

Some candidates often also have the other problem - they speak far too much philosophy on a case and go into numbers too late into the case.

There is no rule on how many minutes you need to spend on Qual vs Quant. If you are receiving the feedback - just adjust the balance until you reach a sweet spot that gives you a good hearty feel of the business problem - which then gets validated by your quant following on.

Hi there,

1. There is no general approach to structure a problem in the sense that you need to learn X buckets that you can then reuse whenever it fits. There is, however, a general approach related to 
   1. thinking about a problem the right way (problem deconstruction via first principles)
   2. communicating it in the right way to the interviewer
   3. If you want to know more, reach out, I have a document where I lay this out in more detail
2. Yes, you can use drills for that → see point C. from above
3. You need to present your MECE analyis/framework first, then dive deeper into the data and ask for it automatically (shows case leadership). Quant/qual is not black or white → most problems have quant and qualitative aspects to consider at the same time (e.g., if you figure out to reduce labor cost, there will be an impact on the affected employees that you need to consider = interdependency).

Cheers,

Florian

Hi there,

I would be happy to share my thoughts on your questions:

• First of all, I would advise you to consider reaching out to an experienced coach like myself to improve on the skill of structuring case studies given your rather fundamental understanding of the initial structure. I have designed the Case Structuring Program specifically for candidates who struggle with structuring a case study like a consultant would do.
• Moreover, I would advise you to just keep practicing, because with time you will surely figure out when to ask for data (and especially how).

If you would like a more detailed discussion on how to best prepare for your upcoming interviews, please don't hesitate to contact me directly.

Best,

Hagen