How do you solve any high school physics problem?
Most people would look for the appropriate formula to solve that specific problem. A few complicated problems would require a few formulas instead of only one, but it doesn’t go much further than that. In real-world physics, you would need to design experiments, interact with other researchers and do a bunch of other stuff. But in the simulated environment of a classroom, knowing the right formulas and how to adapt them to each problem-set is enough.
How do you solve any case interview?
One answer is to look for the appropriate conceptual framework: the combination of relevant concepts for that specific problem. Real world consulting involves doing other things, such as cleaning databases, leading teams and changing clients’ minds. But in the simulated environment of a case interview room, knowing the right frameworks and how to apply them goes a long way.
Case interviews are much more complex than high school physics. There’s a lot more moving parts and often no correct answer. Knowing the right framework to that problem is not enough, but a huge help. As you’re going to see, this does not mean you should start memorizing every framework you see in front of you. If you’ve got that itch, hold on and give me a chance to change your mind.
Conceptual frameworks are the hardest type of structure to achieve a MECE result, but using them is essential to many cases. If the problem cannot be broken down as a formula or as a process, conceptual structures are pretty much the only way to go.
And even some cases that could be structured using algebra or as a process would be a better fit for a conceptual framework.
Conceptual frameworks are structures based on categories of concepts. They usually come straight from theory but can be adapted if you understand it well enough. Examples of these frameworks are the 3Cs, the 4Ps and Porter’s 5 Forces.
But there are others, lesser known, as well. When I was at McKinsey, a simple framework that consultants commonly used for simple organizational problems was the “People, Process, Systems” framework. Any simple organizational problem can be pinned down into a problem with People, a problem with (or lack of) Processes or a problem with (or lack of) Systems. Another example of a lesser known, but widely used framework is the Trust Equation, which says to build and maintain trust you need four factors: Credibility, Reliability, Intimacy and (a lack of) Self-Orientation.
Your final goal should be to be able to build a conceptual framework from scratch. This is the holy grail of case interview structuring, and it can be done with enough understanding and practice. I’ll show you how on another occasion. But most people need to memorize one of two before learning to build their own. And that is fine for now. Think of these as training wheels. They’re useful, but you want to get rid of them eventually!
Although highly useful, there are three risks of using conceptual frameworks:
I’ve never seen a candidate who wasn’t proficient with the use of conceptual frameworks get an offer, so you gotta learn this if you want a high chance of getting the job. Let’s go into each of these risks to see how they manifest themselves and what can you do to minimize them.
- Not knowing a framework good enough for your specific situation
I remember when I was preparing for my interviews the first time. I hated that there were so many frameworks. Case in Point alone had 10 or 12 of them. They were all crammed in a single page or two. Weird trees with arbitrary words written on them. No context given, no explanation of why they were set up that way. I knew it must have some logic, but I couldn’t fully grasp what it was. And how could I possibly memorize so much stuff?
The worst part: I tried one or two of those in a couple of cases and it didn’t really work. I was stuck.
Then I found Victor Cheng’s method: two simple frameworks, once conceptual and one analytical. They were fairly well explained: the questions they asked were ones with a high probability of being relevant. I started only using those structures, and they worked most of the time! The exceptions were some weird cases that weren’t really common, I thought.
But when I started interviewing, you know, in real firms with real interviewers. I realized most cases were, in fact, “weird”.
And when that happened, those frameworks stopped working for me. I was rejected on the first round from all firms, except for McKinsey. I got lucky there. I tried my best to learn to create my own structures from scratch before the final round, but I didn’t have enough time and was finally rejected. It took me another year and months of preparation to learn how to structure so well I smoothly got a double offer.
Most candidates think they have two options: learning to use one broadly useful framework (usually Victor Cheng’s or one they create themselves), or memorizing many. Unfortunately, neither option is good enough to get an offer from McKinsey, Bain or BCG.
But there are two more options: to learn a few useful frameworks and adapting the hell out of them or to learn to create your own frameworks from scratch.
A framework is like a toolbox. A great craftsman has not one generic toolbox, but at least a few, each with a specific function. Craftsmen may have a toolbox to work with wood, another one to work with metals, a third to work with softer materials and a generic one to have widely useful tools so they always have the right one for the job. Call a person like this for a specific job and they will bring the right toolbox, not a random one. Not only that, they will also bring in additional tools from other boxes that will make the job even easier.
Learning a few frameworks is like having different toolboxes to choose from when a job is given. Conceptual frameworks are toolboxes for decision-making. The 3Cs framework, for instance, has a handful of tools to understand the customer demand, and a few tools to understand your company and your competition. Structuring the case well is showing your interviewer you can set up a toolbox that is specific for your job. That you know what to do with each tool.
You can’t have one toolbox that is perfect for every problem, but you can assemble a toolbox that is perfect for each job if you have a few different boxes to choose your tools from and you have familiarity with them.
Knowing how to use well a few different frameworks is exactly like that. It is to know a few different frameworks and to have familiarity with them so you can adapt each to the specific job you’re given. You can solve almost any case if you know the right ones AND you know them intimately enough to adjust well. Sadly, memorizing the 3Cs, 4Ps and Porter’s 5 forces won’t do it. This skill runs a bit deeper than that.
But even being a master adaptor of structures will leave you with a (small) blind spot: some cases are just too novel to be solved with any structure you might know.
My last case when I was interviewing with Bain involved a large company being acquired by a competitor. The owners of this company were already decided to sell and wanted me to come up with a way to bring more of the merger’s value creation to their side vs. the acquirer’s side. In other words, they wanted to sell for more money rather than less money. Naive candidates will try to make the perfect sales pitch, as I’ve seen several coaching candidates do. But being the second time I was applying and given my level of preparation, I was far from naive. I knew I needed a framework on how to get more money when selling a company to a competitor. I also knew no framework on this and had a strong hunch that my interviewer did not expect me to have one, but did expect me to create one. I was lucky my skills were sharp, so that’s what I did. I created a framework for a problem I had never pondered about before, on the spot.
Creating your own frameworks from scratch is HARD. But consultants do it everyday, so partners expect you to do so as well. Consultants are craftsmen, and many world-class craftsmen make their own tools. Programmers build software to help with their own software building. Some surgeons build their own surgical tools. Musicians create new tools to make music as well. With the right techniques and enough practice, you too can learn this skill.
And I highly recommend you to.
I have been interviewed using memorized frameworks on my first try and knowing how to create structures from scratch on my second try. I can tell you it is much more pleasant to be the second guy. No problem is scary, and the conversations with the interviewer just flow naturally. You’re super confident; all your answers are received with a broad smile. You fear no situation. It’s case interview heaven.
- Not being able to adapt the chosen framework to the specifics of your situation
You’ve chosen a framework that fits, but have you adapted it well enough?
The second risk of using a conceptual framework is to be unable to adapt it to the specifics of your situation.
Every case is unique. And most interviewers choose to use cases that aren’t best solved by using the common structures 95% of candidates use. I’m sorry to bring the bad news, but they know which structures you guys use in every single case. As you know from practicing, everyone else is using the same old structures.
Even if you found a suitable framework, or created one yourself, be sure to adapt it to your case.
To do that effectively, you need to know which specifics does your specific case require. You can’t assemble a toolbox from a set of tools or build a custom toolbox without knowing what the specifics of the problem you’re solving are and what that implies to the tools you need to be using.
Using a poorly adapted framework will get your interviewer asking you questions. You better know how to answer them.
Great candidates, as great consultants, use a different structure for each different problem. Every problem has specifics, take them into account.
- Not seeing how your framework connects with other potential structures
If you happen to create and adapt a good framework for your case, you’re in good shape. But there’s one last mistake you can make: not being able to see how your framework connects to other potential structures you could use to solve that case.
Usually, when you start a case you’re given some information on the nature of the problem, but little (if at all) on the nature of the data you will have available to solve it. The goal of the initial case question is to see if you can come up with one reasonable structure / plan to solve the problem. Any structure that is able to solve that problem is fine, and for any given problem there are many possible structures.
But as you start solving the problem, your framework may not be fit to the way the data is organized. The most common instance this happens is when your framework is more conceptual and heavily relies on qualitative data and your interviewer’s data is more numerical. Or vice-versa.
There is nothing bad about starting with the “wrong” structure as long as it could solve the problem. Your job when structuring is to develop an effective plan to solve the problem, not to guess what data your interviewer has or what’s in his mind.
In fact, this happened to me at a Bain final round. The interviewer had told me I was in charge of opening a call center for an automotive company and had to figure out how many people to hire. He asked me how would I do that. I came up with an elaborate conceptual framework involving demand fluctuations, productivity, desired service levels and a few other factors, but all he was looking for was a basic, in-depth estimation. I quickly switched from my original framework to the estimation structure and secured my offer. I might’ve gotten extra points in that partner’s mind for being able to change from one structure to the other seamlessly.
But here’s the thing: most candidates cannot do what I did. Even final round candidates.
If they build a conceptual structure, they can’t see how it would connect to an algebraic issue tree. If there’s an underlying process, they can’t see it unless they’ve chosen to use a Process structure. Having the mental flexibility to switch structures at will is critical when working with other people, both in case interviews and in real projects.
I remember a math teacher I had in high school. He was the guy who really taught me math. He would explain the same concept in many different ways. Many different proofs to the Pythagorean Theorem, the same geometrical problem solved through the lenses of geometry, trigonometry and calculus.
This is the same level of thinking flexibility you should aim to achieve with your structures. Say you have a market entry problem: a company wants to enter a certain market. You should be able to solve this problem using a numbers approach (how profitable can we get), using a purely conceptual approach (assuming no numbers are available), or using a mixed approach (using numbers where they exist and concepts where they don’t). These approaches are different, have slightly different premises but should reach the same result. The decision to enter or not should be the same given different valid approaches.
You could solve a problem involving sides of a triangle by merely memorizing the Pythagorean Theorem, but you need to really understand triangles to how different proofs connect with each other.
The same goes from quickly switching from one structure to another: you need to understand why each part of the structure is there and how it connects with another. A good start is to realize how the 3Cs connect with the Profitability Framework.
Conceptual frameworks are hard and risky to use. So, why bother?
Because mastering these structures will give you superpowers.
You’ll be able to quickly devise a way to solve just about any problem. You’ll discuss highly complex issues with razor-sharp focus. While others are confused with a complex problem, lost in a sea of interconnections and “what-ifs”, you’re three steps ahead, prioritizing solutions based on key issues you’ve distilled from your break-down of the problem.
And I don’t mean this just for case interviews, but for real world problems as well. Remember, a case interview is a simulation of the real job.
If you’ve ever seen a partner from a top firm speak and was astounded by how clear and quick their thinking is, be aware one way they do it is through their incredible structuring ability. You can learn that as well. I was terrible at structuring and I did it.
And there’s another, more urgent reason: if you want to get an offer from a top management consulting firm, if you want to go seamlessly through the process and never fear another case again, there’s no better skill to learn than to create your own conceptual frameworks from scratch.
. . .
If you learn the three core structuring techniques, you can structure any case.
Breaking down equations, breaking down processes and organizing concepts with conceptual structures are truly all you need to build MECE structures. They are the bread and butter.
But man does not live by bread alone.
In the next two parts you’ll learn about Segmentations and Opposite words, two techniques to add flavor to your interview. They are great complements to the three staples. Segmentations are like spices: they add nuance. Opposite words, in contrast, are like olive oil, they’ll make just about anything taste better (until you use too much of it).
These two techniques are as quick to learn as reading the article on them, so why not do it?
In the next part I’ll show you how to use segmentations to refine your structures, and why you shouldn’t rely on them as much as many candidates do.