Skip to content
Get App

Case Interview Fundamentals: Lesson 17 [Frameworks: The 4-step approach]

A four-step process to build and communicate any consulting case framework: define and structure the problem, raise key issues, form hypotheses, and create tests for them.

Key Takeaways

  • Any conceptual framework can be built and communicated using the same four steps: define/structure, raise key issues, hypothesize, and test.
  • Problem definition is critical, especially for poorly defined cases common in final rounds.
  • Every question asked should serve a hypothesis; hypotheses guide which analyses and tests matter.
  • Problem-specific issues signal genuine thinking rather than memorized framework use.

What the video covers

  • Recaps prior lessons on frameworks as problem-solving maps, issue trees vs. conceptual frameworks, and the 3 Cs framework built from the 5 W's and 2 H's.
  • Introduces a four-step process for building and using any conceptual framework: define and structure the problem, raise key issues, create hypotheses, and test them.
  • Step one stresses that candidates often fail by not understanding the problem or forgetting key categories, and that ill-defined cases require more clarification time.
  • Step two compares raising key issues to a doctor's checklist of relevant questions for each category.
  • Step three explains that each question should have a hypothesis about how its answer could change the final recommendation, even if not voiced.
  • Step four compares testing hypotheses to a doctor ordering lab tests; untestable hypotheses are not true hypotheses.
  • Key issues should be problem-specific, such as tailoring questions to the client's industry rather than generic framework prompts.
  • Good issues include detailed objective questions, non-obvious critical issues, and issues tied to testable hypotheses.
  • Interviewers value a well-reasoned framework built in a few minutes over a mediocre one, and fast framework creation (about 10 seconds per item with practice) is achievable.
  • The next lesson will show how to adapt the 3 Cs to make a candidate's business strategy structures distinct from memorized ones.

Answers

Questions about this video

What are the four steps for building a conceptual framework according to this video?

The four steps are: (1) define and structure the problem, (2) raise key issues for each category, (3) create hypotheses for each issue, and (4) create tests for those hypotheses. The video says this process applies to the 3 Cs or any other framework.

Why does the video compare hypotheses to a doctor's approach?

A doctor asks questions based on a hypothesis about the patient's problem and orders tests to confirm or disprove it. Without hypotheses, they would order unnecessary tests. Similarly, consultants need hypotheses to decide which questions and analyses matter.

What does the video say about ill-defined cases?

When a case is poorly defined, candidates must spend more time clarifying the problem before structuring it. This often happens in final rounds, where interviewers are testing problem-definition skills that partners value highly.

Full Transcript — Download SRT & Markdown

00:01
Speaker A
In the last video, I've shown you the fundamentals of frameworks. I walked you through how a framework is nothing more than a map to solve a problem. I've also shown you the difference between issue trees and conceptual frameworks and when to use each one. And I've shown you the first and only conceptual framework I will ever teach you, the 3 Cs. More than that, I've shown you the anatomy of the 3 Cs, why it is the way it is, and I've shown you how to create relevant questions for each category of the 3 Cs or any other framework for that matter without having to memorize anything, using the 5 W's and 2 H's.
00:16
Speaker A
to use each one. And I've shown you the first and only conceptual framework I will ever teach you, the 3 C's. More than that, I've shown you the anatomy of the 3 C's, why it is the way it is, and
00:31
Speaker A
The point was not to have you memorize the 3 Cs or have you use it in every single location. The point was to show you a tool before I can teach you how to build one specific for your cases. But before we do that, we need to know how to use these tools, how to use these frameworks. In this video, I'm going to show you a simple, easy-to-use, and effective four-step process to build and communicate any framework, regardless of it being the 3 Cs or something else that you come up with, as I will show you how to in the next lessons in this module. Anyway, let's not waste any more time and jump right in.
00:44
Speaker A
The point was not to have you memorize the 3 C's or have you use it in every single location. The point was to show you a tool before I can teach you how to build one specific for your cases. But
00:57
Speaker A
So, the easiest, most effective way to build and use a conceptual framework is in four steps. This is true when applying the 3 Cs to a business-related competitive strategy problem or when using any other framework created by you or not. The four steps are: Number one, define and structure the problem. Number two, raise key issues for each category. Number three, have a hypothesis for each key issue. And number four, create tests for these hypotheses.
01:14
Speaker A
it being the 3 C's or something else that you come up with as I will show you how to in the next lessons in this module. Anyway, let's not waste any more time and jump right in.
01:27
Speaker A
Defining and structuring is important because it will enable you to come up with a good structure for your case. Most candidates fail already in this step: not defining and not understanding the problem well, or forgetting key categories. For now, we only have the 3 Cs as categories, but I'll show you how to come up with more soon enough.
01:43
Speaker A
or not. The four steps are: Number one, define and structure the problem. Number two, raise key issues for each category.
01:53
Speaker A
Coming up with key issues means basically asking relevant and insightful questions on each category that you come up with in step number one. This is basically a checklist of questions that you must ask to understand the problem and solve your client's problems. Think of when you go to a doctor with a problem. They'll ask you a bunch of questions, right? To figure out what disease you have or what problem you have. So, this is just like when you go to a doctor. You have questions around different areas for your client, or in your head, and in this step, it's where you talk about those specific questions to the problem.
02:03
Speaker A
Defining and structuring is important because it will enable you to come up with a good structure for your case.
02:10
Speaker A
Now, the third step is to create hypotheses for each issue that you raise in step number two. Think about it. When you go to a doctor and they ask you questions, they usually have a hypothesis of what problem you have in their mind, otherwise they wouldn't be asking that question. So, each question they ask has a purpose, and the purpose is to test that hypothesis. Same for consulting. You only ask questions whose answers could change the final recommendation. Which means that you need a hypothesis for each question you ask, of how that could change the recommendation, even if you don't proactively communicate all of your hypotheses to the interviewer. You need to have them at least in the back of your head.
02:26
Speaker A
Coming up with key issues means basically asking relevant and insightful questions on each category that you come up with in step number one. This is basically a checklist of questions that you must ask to understand the problem and solve your client's problems. Think
02:43
Speaker A
And finally, in step number four, you're going to create tests for your hypotheses. So, to keep up with the doctor analogy, your doctor is going to confirm or disprove his hypothesis with data from whatever you tell them, but also from clinical exams, imaging tests, lab tests, or even genetic tests. Now, the importance of having a hypothesis here is that if your doctor has no hypothesis, they're just going to order more and more tests and exams. If they have hypotheses, they know which ones to order. So, it's similar here for consulting. You need the hypotheses, and then you test those hypotheses. You don't have to communicate all of the tests of your hypotheses to your interviewer, and you don't even have to think of them beforehand, but you'd better be able to test each and every one of them if prompted. Each and every hypothesis. An untestable hypothesis is not a true hypothesis.
02:55
Speaker A
uh different areas to your client or in your head, and in this step, it's where you talk about those specific questions to the problem.
03:05
Speaker A
Now, let's go back and look at each step in more depth. I'll start with step number one, defining and structuring. Just like in brainstorming, you need to define the problem before you structure it. Why? Well, because you can't structure a problem that you don't understand. So, how do you define a problem? And is there a situation where you don't have to bother with it?
03:18
Speaker A
their mind, otherwise they wouldn't be asking that question. So, each question they ask has a purpose, and the purpose is to test that hypothesis. Same for consulting. You only ask questions that the answers to it could change the final
03:32
Speaker A
First of all, you need to have a clear understanding of any problem before you start structuring it. But problem definition becomes absolutely critical when the problem is poorly defined. Contrast these two questions, for example. I got this one from McKinsey's website practice cases, and it's pretty typical of a McKinsey first-round case. I'm not going to read through the whole thing, because the point is that they give you a ton of information. The problem is pretty well defined and there's little room for mistakes here.
03:48
Speaker A
of your head. And finally, in step number four, you're going to create tests for your hypothesis. So, to keep up with the doctor analogy, your doctor is going to confirm or disprove his his hypothesis with data from uh whatever you tell
04:06
Speaker A
Now, take a look at this case question, which was given in a final round at BCG. The CEO of a bank has just visited one of the bank's branches and found it to be too crowded. And that's it. The interviewer didn't say anything else. There's not even a case question in this case question. In the second case, you obviously need to clarify and define the case much more than in the first one. The first one is much better defined, so you don't have to spend as much time doing it. So, this is the first lesson. Some cases are better defined than others, and you need to spend more time defining the case when it comes to you in a poorly defined manner.
04:23
Speaker A
Now, if they have hypothesis, they know which ones to ask. So, it's similar here for consulting. You need the hypothesis and then you test those hypothesis. You don't have to communicate all of the tests of your hypothesis to your
04:37
Speaker A
A quick hint here. When the case comes ill-defined, that usually means that the interviewer expects to see your problem definition skills. That usually happens in final rounds, because partners really value this skill, as it's one of the key things that they do for their clients: define their problems or help them define their problems.
04:52
Speaker A
Now, let's go back and look at each step into more depth. I'll start with step number one, defining and structuring. Just like in brainstorming, you need to define the problem before you structure it. Why?
05:07
Speaker A
Now, the second lesson in problem definition is that you define the case by using clarifying questions. And the third lesson, which goes together with the second one, is that you should aim to have three to five clarifying questions to define the problem. Try to have no more than five, unless the case is so absurdly amorphous that you really need more than that. Let's dig into these last two lessons.
05:18
Speaker A
First of all, you need to have a clear understanding of any problem before you start structuring it. But problem definition becomes absolutely critical when the problem is poorly defined.
05:31
Speaker A
Clarifying questions are broad questions aimed at defining the problem more specifically. You need to understand the specifics of the problem so you can create a framework specific to the case you get. That's true even if you're just going to use the plain 3 Cs, because you need the details for steps two, three, and four of building and communicating the framework. You can use any question to clarify, but there are three types that are particularly useful.
05:42
Speaker A
I'm not going to read through the whole thing because the point is that they give you a ton of information. The problem is pretty well defined and there's little room for mistakes here.
05:52
Speaker A
Number one, questions that explain the objective of the case in detail. This would be useful, for example, in the BCG case I just showed you. In that case, the CEO seemed concerned with crowded bank branches. So, you could ask your interviewer, "Okay, I understand his concern with that. Is there anything specifically he's concerned with? Is he concerned with customer retention, for example, or with something else?" So, here you're trying to dig deeper and be more specific about the objectives of the case.
05:58
Speaker A
The CEO of a bank has just visited one of the bank's branches and found it to be too crowded. And that's it. The interviewer didn't say anything else.
06:09
Speaker A
Number two, questions related to understanding the business model of the client or their industry. You can't structure a case without understanding how that business works. At a minimum, you need to understand how that business generates revenues, so everything related to its products, marketing, and how they sell, and also how that business operates, which is what will determine the business cost structure. So, ev
06:21
Speaker A
The first one is much better defined, so you don't have to spend as much time doing it. So, this is the first lesson.
06:28
Speaker A
Some cases are better defined than others, and you need to spend more time defining the case when it comes to you in a poor, poorly defined matter. A quick hint here. When the case comes ill-defined, that usually means that the
06:43
Speaker A
interviewer expects to see your problem definition skills. That usually happens in final rounds because partners really value this skill as it's one of the key things that they do to their clients to define their problems or help them
06:57
Speaker A
define their problems. Now, the second lesson in problem definition is that you define the case by using clarifying questions. And the third lesson that goes together with the second one is that you should aim to have three to five clarifying questions
07:12
Speaker A
to define the problem. Try to have no more than five unless the case is so absurdly amorphous that you really need more than that. Let's dig into these two last lessons.
07:25
Speaker A
Clarifying questions are broad questions aimed at defining the problem more specifically. You need to understand the specifics of the problem so you can create a framework specifically to the case you get. And that's true even if you're just
07:40
Speaker A
going to use the plain 3 C's because you need the details for steps two, three, and four of building and communicating the framework. You can use any question to clarify, but there are three types that are particularly useful.
07:55
Speaker A
Number one, those questions that explain the objective of the case in detail. This would be useful, for example, to the BCG case that I just showed you. In that case, the CEO seemed concerned with crowded brain bank branches. So, you
08:10
Speaker A
could ask your interviewer, "Okay, I understand his concern with that. Is there anything specifically he's concerned with? Is he concerned with customer retention, for example, or with something else?" So, here you're trying to dig deeper and more being more
08:26
Speaker A
specific into the objectives of the case. Number two, questions related to understanding the business model of the client or their industry. You can't structure a case without understanding how that business works. At a minimum, you need to understand how that business
08:44
Speaker A
generates revenues, so everything related to its products, marketing, and how they sell, and also how that business operates, which is what will determine the business cost structure.
08:56
Speaker A
So, everything related to how they manufacture and and and transport the products, for example.
09:03
Speaker A
Number three, questions that ask for specifics of generic statements. And this includes asking for critical numbers when your interviewer gave them and asking about definitions of words that you might not know. So, if your interviewer said that you need to figure
09:20
Speaker A
out how to grow a client's sales, you want to ask by how much and in which time frame, and if it's in dollar amount or percentage, things like that. If the interviewer says that customer churn has risen, you need to understand exactly
09:38
Speaker A
how is customer churn defined in that client and by how much it has risen.
09:44
Speaker A
So, even if you don't know what churn means, [clears throat] you should ask. But even if you know what churn means, you still need to know how that specific client defines churn because different industries define the number different differently. Also, in
09:59
Speaker A
this category are questions related to units of measurement. So, if you need to get profits to last year's level, for example, you need to understand if it's the profits in the dollar amount or as a percentage of revenues or something
10:14
Speaker A
else. Now, another guideline for good clarifying questions. Density matters. What I mean by density here is that each question that you ask must add a lot of value in your understanding of the situation. You don't want to ask 15
10:32
Speaker A
different questions. Otherwise, 5 or 10 minutes of your interview will have gone by and your interviewer will will assume that you're not going to create a structure to the problem. So, you want every question to count. And this is
10:45
Speaker A
easier said than done. Practice will help you with that as well as focusing on the three types of questions that I just mentioned.
10:54
Speaker A
Three to five well-formulated questions seem to be the sweet spot in most case interviews. You can't usually do less than three because most interviewers leave some issues ambiguous or ill-defined to see if you're going to be able to define the case well.
11:12
Speaker A
In the vast majority of cases, you're going to be able to fully understand the situation with three to five questions.
11:20
Speaker A
One more thing. Explicitly say that you're going to clarify some things that weren't fully clear before you pause and structure the case. This lets the interviewer know that you're aware that you should structure your thinking in a case interview and that leaves them a
11:36
Speaker A
bit less anxious that, you know, is this guy going to stop and structure or is he going to just going to keep asking me questions?
11:44
Speaker A
And my final and most important point. The role of clarifying questions is not to solve the problem.
11:53
Speaker A
Any questions that will help you solve the problem should be within your structure. The role of clarifying questions is to be the minimum set of questions that will enable you to define and understand the problem so that you
12:08
Speaker A
can structure it. Now, after you're done with clarifying and defining the case, it's time to structure it.
12:18
Speaker A
In practice, the first thing you're going to do is to ask your interviewer for a couple minutes so you can structure the case.
12:24
Speaker A
In case you're not exactly sure how the dynamics of asking for time and being silent while structuring the case work, you can watch videos of case interview examples. To help you out, Julio has curated the best videos of case
12:39
Speaker A
interview examples from the web and reviewed each one of them. Just Google for Crafting Cases, Case Interview Examples, or something like that and you'll find it.
12:49
Speaker A
In these couple of minutes in which you're silently structuring, you will want to write down on paper steps one and two, your categories and your issues. And you'll make a mental note of steps three and four so that you don't
13:03
Speaker A
waste time writing them down but are able to say your hypothesis and tests to the interviewer. Anyway, to structure the case, you can have two paths. Number one, creating an issue tree in case the problem is simpler, shorter term, and
13:20
Speaker A
more prone to be solved by problem isolation. Number two, creating a conceptual framework in case the problem is more complex, multifaceted, long-term, and prone to be solved by understanding what's going on from multiple angles. If you create an issue
13:37
Speaker A
tree, it should always be built from scratch. If you choose to create a conceptual framework, you again have two paths. You can adapt from a proven framework such as the three Cs as I will teach you in the next video, or you can
13:52
Speaker A
create one from scratch as I will teach you later in this module. With a well-defined problem and some practice building frameworks, you should be able to come up with the categories in about 15 seconds. The rest of the time you get
14:07
Speaker A
to create your framework is mostly dedicated to the next few steps. So, let's move on to step number two, raising key issues. As I mentioned before, the way to raise good key issues is to ask insightful questions using the
14:24
Speaker A
five Ws and two Hs. But there's a bit more to it. These questions should be asked in a problem-specific way.
14:33
Speaker A
I'm going to give examples of generic questions you could ask in a case and then show you the same questions specifically for an airline strategy case. But before I do that, I must warn you. Almost no candidate who learns to
14:47
Speaker A
memorize frameworks raises key questions in a problem-specific way. 95% of them will just ask generic questions that could fit any case. It's funny. A couple days ago I was watching one of the videos at McKinsey's recruiting page to
15:05
Speaker A
see what they're currently talking about their own recruiting process. One of the consultants in the videos said that what differentiates great candidates from the others is that when they talk about a client situation, they're talking about that specific client situation with
15:22
Speaker A
things relevant to that industry. So, why don't more candidates do that? Well, one reason is that they've memorized frameworks and pre-made framework questions are generic almost by definition. Another reason is that it's tough to learn to make your questions
15:39
Speaker A
problem-specific. But here's the good news. Once you learn to do that, you instantly become two to three times better than everyone else.
15:49
Speaker A
This is one of those things that with just a little effort will bring you a ton of results. And the reason is that consultants never ask generic questions to their clients. They always ask problem-specific questions. By doing the
16:04
Speaker A
same, you will sound more like a real consultant and less like a clueless candidate. So, let's run through some examples so I can show you that in practice. Suppose that I'm thinking of key questions for the customer part of
16:18
Speaker A
the analysis for an airline strategy case. Instead of asking, "Who are my customers and how much do they buy?" I could ask, "What are the key segments within business and leisure travel and how large is each in terms of number of
16:34
Speaker A
seats sold and revenues?" Instead of asking, "What products do they buy and what do they want?" I could ask, "What routes and add-on services are in large and growing demand and what are the the customers' main complaints?" Instead of
16:52
Speaker A
asking, "Where and when do they buy?" I could ask, "What main channels do they use to buy tickets and how price sensitive are they regarding to when they buy?" Notice how the questions on the left could work in any industry, while the
17:08
Speaker A
questions in the right will only work in an airline case. This improves your interviewer's perception of your skills due to a bunch of different reasons. For one, it just makes the problem-solving better. Specific questions have specific answers and they're better than generic
17:26
Speaker A
questions. Another reason, it shows that you can interact with real clients. No one likes the consultant who comes in and doesn't seem to be thinking about the problem.
17:37
Speaker A
Also, it's impossible to assume that you're not thinking about that specific problem when you're asking very specific questions. Just by doing this, you will immediately see your interviewer's face in relief as they realize that you're not a framework robot.
17:55
Speaker A
Notice that you don't need to be full of knowledge about the airline industry and not even know specific terms related to it. All you need is to put yourself in the shoes of someone who buys plane tickets. Just being aware that when you
18:09
Speaker A
board a plane, there are tourists and business travelers, or reminding yourself that each route is a different product, and that airlines would then try to sell you extra luggage and priority boarding and Wi-Fi and so on.
18:24
Speaker A
Or reminding yourself about the last time you tried to buy a plane ticket and the prices kept going up on your favorite website.
18:32
Speaker A
Just having these basic things in mind will help you a lot on being specific.
18:39
Speaker A
Now, being problem-specific is really important, so let me give you another example. Usually, when you're doing competitor analysis, you want to know about your competitor's key resources because they give you a clue about what what your competitors can do and are
18:55
Speaker A
likely to do. The generic way to do this is to ask, "What are my competitors' key resources?" But suppose you're in a case in the mining industry. You might ask, "What are my competitors' mines and how are they different from others?"
19:11
Speaker A
Or if you're in a case in the retailing industry, you might ask something different such as, "What are my competitors' brands, store locations, and suppliers' contracts?" And if [clears throat] you're in a case on the pharma industry, you guessed it, the
19:26
Speaker A
questions are going to be different again. Something like, "What are my competitors' brands, patents, and R&D pipeline?" Now, a lot of candidates feel that this is riskier. They'll say, "What if I forget a type of resource that is
19:40
Speaker A
important in that industry? Am I not better off just by asking the generic question and having the interviewer tell me what resources are important?" And the answer is no. You can and probably should mitigate some of the risk by your
19:55
Speaker A
working on how you phrase the question. Something like, "I want to know about my competitors' key resources. For example, what mining sites do they own and how are they different from each other?" This keeps the conversation open for
20:09
Speaker A
them to suggest other types of resources that might be relevant. However, just being generic and not giving any specifics is the riskiest strategy because it will almost guarantee that you're not going to get the offer. If you're lucky, your interviewer will give
20:26
Speaker A
you a second chance and ask you, "What types of resources do you think would be relevant to take a look at?" But why count on luck?
20:34
Speaker A
Now, one thing that you might have noticed is that the variable that I used in these examples to make the issues problem-specific is the industry that the client is in.
20:43
Speaker A
This is probably the most important variable, but you don't have to limit yourself to it. Let's dig deeper into the mining example.
20:53
Speaker A
Some types of minerals are mined in thousands of mines across the globe, while others have just a few possible mines. We call these industries concentrated. And if the mineral that you're dealing with is in a concentrated industry, you might want to ask if your
21:10
Speaker A
competitors have the ability to control prices as a reaction of your strategy. There's just a few competitors, just a few mines, so they might uh put increase or decrease the the supply so to change prices. Another example, say
21:27
Speaker A
that the mineral you're dealing with has plastic substitutes. So, aluminum is a good example here. Many of its applications are on food packaging, for example, and that can be switched to plastics by the industrial companies that buy it. Now, in that case, you
21:44
Speaker A
might want to add a question regarding your substitute competitors' production facilities, including their capacities and the cost of production. Final example here. Suppose that we're dealing with a mineral that has few buyers.
21:59
Speaker A
Maybe it's a mineral that is only bought by, I don't know, chip manufacturers. And there's just a few of those around the globe. In this case, a key resource would be the relationship your competitors may have with these
22:12
Speaker A
companies and their buyers, and you'll want to ask about that as well. Now, you should always think about how you can make your issues case-specific, including dimensions that aren't the industry itself. You can think of types of products sold as I did in this
22:29
Speaker A
example, but you can also think of the country the company is in. So, a company in China is different than a company in Europe. You can think of the specific goals of the client or the context of the case. So, for example, if a case
22:43
Speaker A
happens in a recession, this is different uh than if it happens in an economic boom. One last thing regarding the issues. A good list of key issues for each category of your framework should have five characteristics in order of importance. First, the question
23:01
Speaker A
should be problem-specific. This is what I've shown you so far. Second, you should always have the most obvious issues to solve the problem.
23:10
Speaker A
They might be obvious, but they're obvious for a reason. Third, you should have a few non-obvious issues as well, so interviewers can see that you can think outside the box.
23:20
Speaker A
Fourth, the list should be as exhaustive as possible. Because we're dealing with qualitative concepts here, it's impossible to be completely exhaustive. This is not a math equation, but still, having a list of questions that seem to be comprehensive is much better than one
23:38
Speaker A
that doesn't. Finally, number five is an extension of number three. You should have all the non-obvious issues that are critically important to solve this problem. Every case has a few of those, the incredibly important issues that no candidate mentions because they're so
23:55
Speaker A
specific to that particular case. I put this as the fifth in this list, not because this isn't important. These issues are actually the ones that make or break real consulting projects, and especially in final rounds, they're really, really important. I put this as
24:11
Speaker A
a fifth because less than 1% of candidates can pull that off consistently. It requires amazing business sense. Now, because this is so difficult, I highly recommend that you focus on the other four factors first, and after you're good there, you can
24:27
Speaker A
focus on this fifth. The best way to improve in all of these five factors is to practice structuring extensively with access to well-developed answers. This is one of the reasons why you find comprehensive drills in this chapter, and it is also one of the reasons why
24:44
Speaker A
we've created a course dedicated to structuring that has a huge practicing components through tons of drills.
24:53
Speaker A
And hey, one quick tip here. If you're having trouble coming up with good frameworks and you get all the categories right, having a checklist of these five points on your side as you practice will help you figure out where
25:06
Speaker A
to improve. Okay, cool. With the categories and the key issues laid out, we have the bulk of our framework ready to go. But even though we know what are the key areas to look at in order to solve the problem, and even though we
25:21
Speaker A
know what are the key questions within each area, we still don't have a hypothesis-driven approach.
25:27
Speaker A
Which leads us to our third step, creating hypotheses for each key issue. Now, a lot of people get confused about why you need hypotheses and why consulting firms have a hypothesis-driven approach in the first place. In fact, a lot of people don't
25:43
Speaker A
even understand what makes a hypothesis a hypothesis. Let's take care of those concerns. There are two key reasons why you need a hypothesis that underlies each question that you ask when you're raising key issues.
25:58
Speaker A
Reason number one is that because you want to be effective and efficient in your work, each question must be important enough to potentially change the final recommendation. You need to explain or to be able to explain to your
26:12
Speaker A
interviewer why you're asking each question. The thing with people drawn to management consulting is that many of them are curious people. They want to know everything about the situation, the industry, and the client. I'm like that, you probably are, too. However, you
26:31
Speaker A
don't have the time to learn everything in your typical consulting project, which means that you should focus on what's going to move the needle. You need a mechanism to sort the critically important from the merely interesting.
26:45
Speaker A
And that's one role of the hypothesis. If you have an underlying hypothesis for each issue, that makes it more likely that each issue is critical, not just interesting.
26:57
Speaker A
Now, you don't need a formal hypothesis to do that. You could just explain why each issue is critical, and that's one way to go. If you mention why you need to answer each of the questions that you you're
27:11
Speaker A
raising the key issues step, that's already good enough for most standards. But there's a second reason why you need a hypothesis behind each issue.
27:22
Speaker A
You need to answer the question in a very tangible, specific way. You're not just answering your questions in a generic way, you actually need to answer them in a data-driven, fact-based, specific way. And to show your interviewer that you can do that, you
27:41
Speaker A
need a testable hypothesis. Now, this is all very theoretical so far, so let's bring that to practice.
27:49
Speaker A
I'll use an example of the airline strategy case that we discussed a few minutes ago.
27:54
Speaker A
Remember that one of the problem-specific issues that we had under the customer analysis was, "What are the key segments within business and leisure travel, and how large is each in terms of number of seats and revenues?" Well, when you mention that issue to
28:13
Speaker A
your interviewer, they don't know if you're asking that because you have a purpose in answering that question or because you read in some case interview book that you're supposed to talk about customer segments in strategy cases. Now, there are
28:28
Speaker A
several potential hypotheses that would justify this question. And you need to show at least one of them to your interviewer.
28:37
Speaker A
So, I'll show you a simple one that works. If you want to learn more about how to formulate hypotheses and how to test them, we've got you covered in the hypothesis testing module. We've actually got a whole module for
28:51
Speaker A
generating and testing hypotheses because doing that is a skill that appears in multiple parts of the case, not just in the framework. So, it's a building block in itself. Anyway, here's a simple hypothesis. After stating your issue, you could say to the interviewer,
29:09
Speaker A
"And my hypothesis is that there must be at least one highly profitable segment in the market that is not served well by anyone, us or our competitors, and that we could target that segment from now on. An example of that would be
29:25
Speaker A
last-minute business travel between large business centers." Now, this is a good hypothesis not because I've used the word hypothesis in it, but because it makes our issue super relevant, because it is testable, and because if it's true, it moves the case
29:45
Speaker A
forward. If we found such a segment that is highly profitable and underserved, we'd have way more focus on how to create this airline's strategy. Now, the most important of these three things is that it's testable. If this hypothesis
30:02
Speaker A
were not testable, it could never make the issue relevant nor drive the case forward in practice, which brings us to the fourth step, creating tests for your hypothesis.
30:16
Speaker A
But before we do that, let's get something straight here. You're probably worried about how to create and present a framework with all these steps in enough time. I'll address that directly right after walking you through creating tests for your hypothesis. But still,
30:33
Speaker A
it's important for you to be aware that you only have to do the hypothesis testing step as prompted by your interviewer. In a candidate-led case, you might prompt yourself and choose to test a specific hypothesis right after you present the framework so you can
30:51
Speaker A
drive the case forward, but still, you're doing it as prompted and only one hypothesis at a time.
30:59
Speaker A
Let me show you how to test the hypothesis that I just raised. I showed the interviewer that I can do it by showing them a hypothesis testing plan. It is basically a list of questions I want to answer with the data
31:14
Speaker A
that I need in order to answer it. I'd first say that I'd build a list of segments in different ways, such as the reason for them to be traveling or the routes that they take or how many flights these segments uh flight per per
31:29
Speaker A
year or the demographics of the person. So, I'd segment the population in different ways and try to find a segment that is large and high paying. This will ensure that they're highly profitable. So, I'd look at their size by both the number of tickets sold
31:47
Speaker A
to this segment and the revenues they generate, and I'd double-check their price sensitivity by looking at how much they pay per ticket versus the average ticket price on the same flights.
32:01
Speaker A
So, an example of a segment like this would be last-minute flights by executives in business travel routes.
32:08
Speaker A
Another example could be high-net-worth individuals who travel every weekend to their leisure homes, for example.
32:16
Speaker A
Then I'd survey the customers from the most promising highly profitable segments that we find to see what are their main frustrations with the flights they take. And I'd also look for what they desire the most if they could change anything about their
32:33
Speaker A
air travel experience. So, what I'm looking for here is for that uh highly profitable segment to have lots of frustrations, which would indicate that they're poorly served currently. And finally, as a last step, I'd see our company's competencies to
32:52
Speaker A
see which of those segments that are poorly served and highly profitable that we can serve better without reinventing ourselves. So, this is an example of a hypothesis testing.
33:05
Speaker A
Now, one quick thing here. Some would say that all of this is obvious. It's obvious that you want to find a highly profitable segment with lots of frustrations that it's easy for your company to serve. And that's true, but
33:18
Speaker A
it's not the point. The point is that I can uh show the interviewer that I can find a way to analytically prove that hypothesis and quantify its impact instead of just knowing what is obvious to everyone.
33:34
Speaker A
Right? So, proving in a very systematic way is important because any gaps in reasoning show, and if there aren't any gaps in reasoning, it's easy to convince your clients to take action. It's It's much better to have that than to just have
33:51
Speaker A
intuition about what's obvious. So, that finishes our four-step approach to frameworks. But there's still one concern left to be answered. How the heck are you going to do all these four steps in two or three minutes and still have the time to
34:10
Speaker A
communicate all of it? Well, let's break that down. I'll start with the best practice on creating the framework in under 3 minutes and then show you how to communicate it so you can maximize your interviewer's perception of your work.
34:26
Speaker A
So, let's start with creating first. You got your piece of paper in the interview. You've defined the case clearly, and you've clarified any questions that you had, and you've asked your interviewer for a couple minutes to structure your case.
34:42
Speaker A
Now, the first thing you'll do is step number one, creating the categories. You'll think which categories you want to consider. For now, you only know how to create the three Cs and other pre-made frameworks that you might have
34:55
Speaker A
learned about before taking this course, uh but in the next video, I'll show you how to adapt the three Cs, and later you'll know how to create structures from scratch. So, you'll have more flexibility in creating the categories.
35:10
Speaker A
And you'll write that down on paper. Thinking which categories you want to use and putting them on paper should take you about 15 to 30 seconds, no more than that.
35:21
Speaker A
Now, the next thing you'll do is to think about key issues for each category and to write them down. You don't have to write them down as whole phrases as you see in the drills that I do, but uh
35:34
Speaker A
but you just have to write down short snippets of text that will help you remember what are the key questions so that when you're presenting them to your interviewer, you can speak in a whole phrase format.
35:49
Speaker A
So, write short snippets so you don't waste time, but uh be able to speak uh about these key issues in a whole phrase format. You should be able to come up with and write down each key issue in 5
36:02
Speaker A
to 10 seconds each once you have the practice to do so. So, this is quick, but it's doable.
36:10
Speaker A
It also means that you should be able to write down uh 12 key issues in a minute or two.
36:17
Speaker A
While there's no right or wrong number of issues in your framework, you'll see through the examples in this module that most of my frameworks have between 10 and 15 issues to consider, which is why I used the number 12 as a reference.
36:32
Speaker A
Now, as you write down these key issues, you should also think about the hypothesis that underlie them. However, you don't need to spend time writing down these hypotheses. All you need to do is to have a hypothesis for each
36:49
Speaker A
issue you write down so you can remember why you're asking each question, what is the logic behind it.
36:58
Speaker A
For that reason, it shouldn't take you any extra time on top of uh creating the the issues to do that.
37:05
Speaker A
Your mind will come up with the hypothesis at the same time that you think through the issues, right? They're inherently uh related to each other.
37:16
Speaker A
Finally, you don't need to spend any time even to think about the test for each hypothesis in the few minutes that you have in silence to think through the problem.
37:27
Speaker A
Although this is a critical step for the framework to work, it is a step that you only do after you present the framework for the few hypotheses that your interviewer will point you out to test.
37:39
Speaker A
So, this whole process of creating the framework should only take you a couple minutes, no more than 30 seconds for the categories and about 2 minutes for uh writing the issues down and thinking of the hypothesis in your head.
37:55
Speaker A
Will you be able to do this in a couple minutes as you're starting out? No, you'll probably need 5 or 10 or 15 minutes to create a decent framework before you get more experience. But once you master it, it will be easy to do so,
38:10
Speaker A
which is why you should practice structuring separately from whole cases so you get more and more practice on it.
38:18
Speaker A
One question that comes up a lot, isn't 3 minutes too much to spend in silence creating the framework? I heard from someone that I shouldn't take more than 1 or 2 minutes. You'll hear that from a lot of people. Now, here's the deal.
38:32
Speaker A
Which do you think is better from an interviewer's perspective? A candidate who has spent 3 minutes and created an amazing framework or one that has spent 1 or 2 minutes and created a mediocre one?
38:47
Speaker A
It's really, really hard to come up with a great framework in 1 or even 2 minutes. Even I can't do it.
38:55
Speaker A
And most people can't do it in 3 minutes, either. So, they didn't have the practice, but as you go through this course and if you want to take a step further and and join our structuring course, you'll be able to do an amazing
39:10
Speaker A
framework in 3 minutes. And that extra minute, if spent to make your framework better than 99% of frameworks out there, is the best well-spent minute you can have in any interview. The results you'll get are going to be much better,
39:26
Speaker A
and this will set the tone for the rest of the interview. The interviewer is going to be really impressed with you.
39:32
Speaker A
So, they'll help you out more uh and and they'll just have a better impression overall.
39:38
Speaker A
Now, eight or nine out of 10 interviewers will be glad that you took a little bit extra time and actually did a good job.
39:46
Speaker A
In fact, most of them won't even notice you took a minute longer. They'll be answering email in their smartphones or they'll be just hanging out. I have never seen or heard of an interviewer using a stopwatch or anything like that.
40:00
Speaker A
Now, from time to time, you will find a more hurried interviewer that will cut you off and ask you to start presenting the framework to them uh at minute 1 and 1/2 or 2. Now, you won't lose any points if this happens.
40:16
Speaker A
All you have to do is to say that, you know, you were taking a bit of extra time to think through some issues specific to that client situation and that, sure, you can start showing your work to them
40:28
Speaker A
uh and then think of more issues as you present it to them. Now, once you create your framework and announce to the interviewer that you're ready, you need to communicate it to them.
40:41
Speaker A
Here's how you do it. First, you walk them through each of the categories that you've written down on paper and explain why they're important.
40:51
Speaker A
This should take you half a minute and it's important because it gives your interviewer a high-level view of your structure. You walk them through each category before you even start talking about the issues.
41:03
Speaker A
Then, you go category by category explaining the key issues. Now, for only a few of them, you should explain what's the underlying hypothesis.
41:15
Speaker A
The other hypothesis will be just in your head in case they ask. Just to make this extra clear, you will walk them through all of the issues you've raised and for a few of the most important, you will also mention what's
41:29
Speaker A
your hypothesis behind that issue. Why is that? Well, you want to go through at least a few hypotheses so that you can show your interviewer that you're hypothesis-driven and very purposeful about the questions that you ask. You want to show them that
41:44
Speaker A
there's some thinking behind each issue. And in most interviews, you don't want to go through all of the hypotheses because that would take too much time.
41:54
Speaker A
You basically want to show your interviewer that you can do it and that you have hypotheses for every issue in your head and if they want to know your hypothesis for all the issues, they're going to ask follow-up questions.
42:06
Speaker A
Now, going through all of these issues and a few hypotheses should take you from two to four minutes depending on how much detail you put into everything.
42:16
Speaker A
Now, you must be thinking, what about the tests for these hypotheses? I don't need to think about them when creating the framework nor to communicate it to the interviewer. So, what's the point of that step? Well, what you need to make
42:30
Speaker A
sure is your hypotheses are testable, but you don't have to come up with the specific test that you're going to use until you actually have to test the hypothesis.
42:41
Speaker A
And that will come after you present your framework. In an interviewer-led format, the interviewer will ask you, how would you test a few specific hypotheses in your framework? They will do this after you present the framework as follow-up
42:58
Speaker A
questions. And in a candidate-led format, they will ask you, well, what would you like to do next as soon as you finish presenting your framework? And guess what you'll do next? That's easy.
43:09
Speaker A
You pick one or two critical hypotheses and test them to reveal critical information to help your client solve their problems.
43:19
Speaker A
Testing the prioritized hypothesis is always the next step after presenting the framework unless your interviewer just asks you to do something else.
43:30
Speaker A
One important thing though, I highly recommend you to practice this hypothesis testing step a lot when practicing structures or when doing mock cases. And the reason is that most candidates never ask for tests as follow-up questions as often as
43:47
Speaker A
interviewers do. So, when you're practicing with other candidates during mock interviews, they'll never ask you to test hypotheses.
43:56
Speaker A
And this means that if you don't make a deliberate effort to practice, you'll get no practice before your interviews and this is going to be tough on you.
44:06
Speaker A
You're going to have a hard time knowing how to test hypotheses. So, to wrap up this video and the four-step approach to frameworks.
44:16
Speaker A
First of all, make sure that the problem is well-defined before you try to structure it. You can't structure a problem that you don't understand.
44:24
Speaker A
Once you have a well-defined problem, you'll know what type of framework to use as well as how to come up with the right categories and the most relevant issues and hypotheses within each category. Of course, you're going to
44:38
Speaker A
learn how to create these categories uh in the next video on how to adapt the three Cs, but this step is critical.
44:45
Speaker A
Second step. Make sure your key issues for each category are problem-specific. Of the five characteristics of a good list of key issues, this one's the most important. Remember to add specificities about your client's industry, country, or the specific context that they're
45:04
Speaker A
going through to make your framework sound more specific and less rehearsed and memorized. Third thing. Use hypotheses for key issues to show why they're critical to your investigation. Hypotheses not only help you show the interviewer specifically what's on your mind
45:24
Speaker A
regarding that issue, but also will help making the broader question of your issue specific enough so it can be tested. And finally, know how to create tests for the hypotheses to show your interviewer that you will know not only
45:40
Speaker A
which questions to ask in a real project, but also how to answer them. A consultant that can generate hypotheses, but that cannot find ways to test them is not doing their job right. And regarding communication, when creating the framework, focus on the first three
45:58
Speaker A
steps. And when presenting it, focus on the first two, but also sprinkle in some hypotheses to the most important key issues so your interviewer knows you're being hypothesis-driven in how you create your questions.
46:14
Speaker A
Now you know how to use frameworks. You also know from the previous video how to manipulate the three Cs so you can actually create a framework. But how can you create more specific frameworks to your problems?
46:28
Speaker A
How can you guarantee that you're touching on all categories relevant to the problem you have at hand?
46:35
Speaker A
Well, the next video will show you how to adapt the three Cs. Learning to do that will not only show you how to make your business strategy structures different and better from those of all other candidates who have memorized
46:51
Speaker A
frameworks, but it's also an essential step in the in learning how to create frameworks from scratch so you can solve non-business problems or very different business problems as well. So, uh this is the end of this video. Jump into the
47:05
Speaker A
next one and I'll teach you how to adapt the three Cs to create better frameworks.
Topics:case interviewconsulting frameworks3 Cshypothesis-driven thinkingproblem structuringissue treesMcKinsey caseBCG caseconsulting interview prepbusiness strategy

Get More with the SozAI App

Transcribe recordings, audio files, and YouTube videos — with AI summaries and speaker detection. 30 minutes free.

Or transcribe another YouTube video here →