Hypothesis

Short answer: it depends, but most of time you may not need it.

In an interviewer-led case study, you don't have the chance to hypothesize and generate a tree for it to problem solve with. You can be asked to hypothesize, for example, “what might be the possible levers to boos sales?” just feel free to give your hypotheses.

In an interviewee-led case study, you can decide yourself if you want to use a hypothesis tree approach. Issue-tree approach is easier for most people. If you want to use hypothesis-tree approach, just state explicitly in the beginning “my hypothesis is xxx, and a few areas to validate such hypothesis is A, B, C…”. But:

• You may need to help some interviewer understand why you have such hypo (some interviewers don't like the approach)
• When walking the interviewer through the tree, do describe how it my change your hypo

I personally find hypo-tree approach easier for business problems: 1) seen a lot, 2) requires go / no-go decision, or other yes / no situation

Hi there, 

The hypothesis' main goal is to help you structure the case. The way I think of it is: “what do I need to look into in order to prove hypothesis?”. Those areas “you need to look into” are esentially your case framework. 

For same cases, the hypothesis will be more easy to state out. I personally am a fan of saying it out loud, as early in the case as it makes sense. But most people don't do / recommend that anymore. For some other (typicailly more unusual cases), stating a hypothesis would feel … very unnatural and then you should not force it. 

To recap: 

1. Hypothesis comes early in the case to help you structure it and then drive the case

2. Throughout the case, you can revise your hypothesis if you find info that changes it

3. Conclusion should be your updated hypothesis with a summary of the supporting arguments. 

Using the exact word “hypothesis” during the case is not the important part, but having a “hypothesis driven approach” is the key to solving a case. 

Hope this somehow helps :) otherwise, feel free to follow-up with additional questions. 

Best, 
Ariadna 

Forget the hypothesis. 

It doesn't matter what your hypothesis is, or that you state an hypothesis.

What really matters is whether your structure allows to test an hypothesis (whatever it is) and it is clear on how it will drive to a decision.

Let me say this in a more practical way. If the question is “should we enter this market”, it doesn't matter if the hypothesis is that you should or that you should not. What it matters is that you are able to state what would be the criteria that would lead to a certain recommendation.