Segmentations: The 4th Way To Be MECE In Case Interviews

[@CraftingCases]

If you've found the 5 Ways to be MECE helpful to improve your MECEness, you're gonna find our free course 3X more helpful. That's because we show you how to answer any case interview question in a step-by-step method through our signature system "The 6 Building Blocks". Join now at www.craftingcases.com/freecourse

[@javimaci4615]

Hi Bruno, This is Javier a Kellogg's EMBA alumni. I watched Victor's video and yours. I have to say that I am impressed. You guys take Victor's explanation to another level. I will probably join the paid services out of respect of the work you put in. Well done and thank you!

[@CraftingCases]

Hey Javier, super glad to hear that we’ve been of help!

[@eugenlow585]

The mix effect has really got me thinking deeper on a more insightful level! Thanks

[@Goldeggyolk]

Hi Bruno, really appreciate your content out there. Today, I just discovered another use of segmentations in brainstorming, especially when one could have infinite possibilities through conventional means. A brainstorming question included me to specify what products and services can a telco cross-sell, and I was dumbfounded, as there were literally so many opportunities, but I didn’t know how to structure it correctly. I was constantly thinking about this specific case, and after a number of hours, I determined that segmentation regarding how telco can cross sell - felt like a surprisingly great pick. I proceeded by segmenting the channels by customer service, physical storefronts and online channels and plan renewals. It was so evident that the kind of products that could be sold through those differed significantly (for online - customised products such as streaming services and discount vouchers as part of bundles through tracking customer data, smart devices through physical stores and cross/up sell suggestions through customer complaints). It was so amazing to see that your techniques fit so perfectly in any case, it was just difficult to recognise in the first go. I was convinced that some harder cases required something more sophisticated, and didn’t think about this technique until much later. Again, thank you so much for your content here. I am discovering new applications of them every single day! :)

[@CraftingCases]

Seems like you're getting the gist of structuring and problem solving Akanksha, congrats :)

[@monideep82]

Bruno this is great video. May I suggest you to publish a book with these insights. Your way of thinking is superlative !!!!!

[@edisonngu7838]

Thank you so much for existing. Great tips and handome presenter - you rock!

[@dinocollins720]

Thank you Bruno! This video was extremely insightful!

[@eug3r]

your videos are such great value, keep it up!

[@melaniehelinger5635]

Dear Bruno, thank you for your amazing videos. You are really helping me getting through my preparation! I have one question regaring segmentation of a problem. Before choosing a segmentation, is it okay to ask the interviewer if this is the right path or if a segmentation is irrelevant for this case (e.g. because the market share dropped in all customer segments/regions/product categories)?

[@rondovk]

Victor Cheng often says in his materials that you can ask for a 'correct' segmentation from the interviewer by asking. Is this not true? Seems that would be too easy, so I am inclined to agree with your points in the video.

[@CraftingCases]

What has happened in the past 5+ years is that interviewers started asking what do YOU think would be the most useful segmentation patterns if you ask them that. Because of that, I think being proactive is a better approach. After you offer a few segmentation patterns, they’ll typically tell you which one to pursue or hand you a chart that makes it obvious. This makes a lot of sense since in real consulting you’ll have to choose how you’ll segment your data and no one wants a junior consultant who needs to constantly ask this kind of stuff and has no clue. I think back when Victor made his materials, no one segmented (all they had was case in point), so merely asking for a segmentation was better than everyone else. Nowadays candidates are more sophisticated and everyone’s asking for segmentations - interviewers will then ask you to go a step further and suggest the most useful patterns (as I explain in the video, there ARE patterns that are more useful than others for each situation). This is all just speculation, of course. I wasn’t around back when Victor started teaching, so I don’t really know if people really ignored segmenting back then. What I can tell is that I have never seen nor heard of an interviewer that will react positively with his prompt of “what segmentation should I use here?”

[@LuisHenrique-me7gm]

@@CraftingCases I had the same doubt! Thanks! You mentioned that in real consulting you have to choose how you'll segment your data, but, do consultants/junior consultants usually pick only one? Or do they try a few more? I find it weird because some data are so relevant (such as revenue of the company, e.g.) that should have "all" info in a Data Base (gender, age, region, dist channel, ....). Then, the work you would have to segment this data wouldn't be so hard using Excel or Power BI (the ones I know and use, for sure, there are other options). Companies dont usually have this DB well structured and filled?

[@orr4jk]

Hi Bruno, Amazing I really appreciate this work. In some of the books I read on case interviews they suggest asking out right if there is information on segmentation. What do you think of this approach. Edit: I read the response you made below, so I'd like to rephrase the question. Do you think it'd be smart to say "based on what I know so far here's a few ways I'd like to try segmenting, do we have any data for these segments" and if she or he says no a follow up "is there any data on segmentation I might have missed?"

[@CraftingCases]

I think that's a fine approach!

[@Nerminalaly]

keep on thanks alot ✌🏻

[@KaFeSee]

Thank you for this really insightful video on how to use segmentations! I have a question regarding your tips. You mentioned that I should use more than one criteria to segment a category like Customers to Age, Income, Region, ... I absolutely see the advantage in terms of better hypotheses generation. On the other hand, following the theory of MECEness, to segment a category into more than one part is not clearly mutually exclusive isn't' it? Because customers belong to a segment each. What are your thoughts on that? Thank you in advance!

[@13579saez]

15:35