Building hypotheses is very difficult. Most candidates in a McKinsey, BCG et al interview would not know when to build the hypothesis, what comprises the hypothesis, how to test if it is MECE etc. This simple technique is one way to build hypotheses and used on real consulting engagements. It was developed to help candidates prioritize their analyzes and ensure the hypotheses are MECE. When practicing this technique note that the development of the decision tree must be done quickly and cleanly.

Training by ex-McK, BCG et al. Partners

Receive free exclusive episodes to advanced strategy and case interview training programs, plus a chapter from Bill Matassoni's Memoir. This is the ONLY way to sample FC Insider material.

Where else can you learn from ex-partners?

Sign up to receive exclusive FC Insider episodes. Start now:

Privacy Policy


8 responses to Build Hypotheses With Decision Trees

  1. Dear Michael,

    Thank you for the clarification; that was helpful.


  2. Hi Femi,

    In the ideal world you could tell the interviewer your plan is to quickly get to the hypotheses and then quickly build out the decision, talking aloud, as you work your way to the hypotheses.

    That will work best. Silence is okay and is allowed, but talking as you write and think is always better.


  3. Dear Michael,

    Thank you for the podcast. I must have listened to it for the umpteenth time but I still have a question and my question builds on Singh’s.

    My guess is that one’s worksheet will have the 2-level decision tree terminating in hypotheses but that during presentation one should only talk about the hypotheses part – the decision tree merely serving as a structure on which the hypotheses rest. An alternative will be to build the decision tree on one’s rough sheet and then transfer/write the corresponding hypotheses on one’s worksheet i.e. as the problem-solving plan and then present the plan. I ask because you emphasize the need to be silent on the decision trees and I wonder how that translates to the world of practice. Will the former irritate an interviewer? Will the latter appear complicated?

    Thank you for your time and attention.


  4. Hi Singh. This means that you are using the decision tree as a tool for your benefit, behind the scenes to help you understand which parts of the case are important and prioritize these areas. Once you have them prioritized, build an hypothesis for the prioritized area. McKinsey only want to see the hypotheses and could care less about the detailed structure. So, if you choose to discuss the structure, do so lightly knowing that they want the hypotheses. Hope that makes sense. Michael

  5. Michael, you mention here that at McKinsey “don’t show them your decision tree, just state / discuss the hypotheses”.
    Could you please elaborate on that ?

  6. Thanks Michael.

  7. AS, it is called a “Trend drill hypotheses” and there are many such videos in Felix’s section. Simply go through the list of cases and see where we have done them. Michael

  8. Michael,
    You referred to “the fish technique” with reference to building hypotheses in one of Felix’s coaching videos. Is there a podcast/video that explains the technique ?

Leave a reply

You must be logged in to post a comment.