Question merged
This question is read-only because it has been merged with How to solve vague cases that don't fit into any frameworks?.

How to answer questions that have no frameworks associated to them?

Case
Recent activity on Nov 17, 2018
3 Answers
2.3 k Views
Michael
Certified
asked on May 25, 2018
Looking for solid partners. Preparing for MBB interviews

(edited)

Overview of answers

Upvotes
  • Upvotes
  • Date ascending
  • Date descending
Best answer
Sidi
Expert
updated an answer on May 25, 2018
McKinsey Senior EM & BCG Consultant | Interviewer at McK & BCG for 7 years | Coached 350+ candidates secure MBB offers

Hi Michael!

I think the basic premise of your question needs to be discussed. What I mean is the following:

I am absolutely no friend of frameworks to begin with! I believe most (or probably all) casebooks are teaching a fundamentally flawed way how to think about business / strategy / organizational problems! A framework as such is worth nothing if it is not embedded into the specific context of the situation! This means, each element that you want to scrutinize ("building blocks" of the framework so to speak) needs to clearly relate back to the question that you want to address!

This is why you ALWAYS start from the specific question that you want to answer! From there, you define the criterion or criteria that need to be met in order to anwer this core question in onw way or another.

In 95% of cases, value creation will be the central element. This is nothing else than profit generation over a specific time frame. You then draw a driver tree for profitability in order to isolate the numerical drivers for your solution. And then, only after you have drawn out the driver tree, you can map out the relevant qualitative "framework elements" to the sub branches. This approach is much much clearer then any framework you will find in any case book. And, contrary to frameworks, which are hanging in the air and do not logically relate back to the specific question, his is a bullet proof approach when done rigorously.

The caveat is: this requires time and coaching to internalize. But ultimately, this is how consultants think about problems - how can we optimize for value creation?

Cheers, Sidi

P.S.: Have a look at the structure I outlined here: https://www.preplounge.com/en/consulting-forum/how-would-you-structure-this-1737#a3813

(edited)

Upvotes
KP
Skilled
replied on May 25, 2018
Preparing for final round McKinsey interviews (late Oct). Done 70+ cases so far.

Thanks for that insight on structuring, Sidi! Is it a good idea to define the core question as soon as you're done clarifying the case question or after you've had a chance to understand a bit more about the business. In other words, should the order of response be: 1, 2, 3 as below or 1, 3, 2?

1. Para-phrase/ clarify case details

2. Define and confirm core question

3. Ask questions to understand how the business works, etc.

Upvotes
2
Sidi on May 28, 2018

Hi KP! In most cases, [1,2,3] will work well. If you, however, struggle to understand the context, are not clear how the business makes money, etc., then [1,3,2] is the way to go. Cheers, Sidi

Anonymous replied on Nov 17, 2018

Stay curious and focus a lot on identifying the problem at the beginning. In consulting, there isn't really a framework for most of the things we do. It is all iteration and continuing to ask questions. Sure, it takes place over a week or so and an interview you have to do it in ten minutes, but the key is just to keep asking more questions until you have enough info to propose your own structure.

Once you have your structure, see what the interviewer's reaction is to it and then iterate more or dive deeper into the next level of the structure you created.

Upvotes
0
Sidi gave the best answer

Sidi

McKinsey Senior EM & BCG Consultant | Interviewer at McK & BCG for 7 years | Coached 350+ candidates secure MBB offers
429
Meetings
5,830
Q&A Upvotes
78
Awards
5.0
134 Reviews
How likely are you to recommend us to a friend or fellow student?
0
1
2
3
4
5
6
7
8
9
10
0 = Not likely
10 = Very likely