Issue Tree Example - Step-By-Step Approach with Takeaways

everyone so in this video I'm gonna show you an issue 3 exemple and not only I'm gonna show you how I build the issue tree but I'm also gonna show you my thinking process in each step of the way so you can know how I think as I build the issue tree and at the end of the video I'm gonna share with you the three major takeaways from this specific tree now issue trees are important because they're the main technique they're kind of the blueprint of how makisi BCG bank consultants think to solve business problems and you can apply them in case interviews if you're interviewing for a job at these firms and also you can apply them in real business problems which is why they test these skills in the first place now in this video I'm gonna assume that you're prepared for your cases but you can use the same technique if you're solving a business problem and you're past the the job interview stage already now I chose this specific example for this video because it's an interesting one it's a fairly easy case it's a fairly simple business problem its own airline fuel costs so nothing really complex but about 95 percent of candidates would mess up and not be able to structure and solve this case why is that well because it's a very specific problem and you won't find a framework for it anywhere so you really have to think from first principles and structure from scratch which is the exercise that we're gonna do in this issue tree now Before we jump right into the issue tree example there are two quick things that I want to tell you the first one is that I'm making other videos at least two of them on issue three examples so you can learn from a variety of different problems and different issue trees at least one of those is going to be on public sector and the other one I haven't decided yet but just make sure you subscribe to this channel and look for those videos so that you can watch after this one the second thing I wanna share with is that after I show you the issue tree at the very end of this video I'm gonna share with you two great resources so you can learn to make issue trees better one of them is a comprehensive guide on how to build great issue trees and the other one of them is a step-by-step again comprehensive guide on how to be missing the five ways to be missing alright so I hope you stick to the end so you can get those two resources and watch me creating an issue tree so you can learn this technique better so let's jump right in so here's an example of a case that it could get in the case interval or a real business problem that it could get when working on an airline suppose that an airline's fuel costs have recently risen and the CEO is concerned and he wants you to find out why has the fuel costs risen for that airline so the first step in building the issue tree is to define the problem well and here I want to do two things first of them is to confirm the metric the numerical variable that the CEO is concerned is it total expenditure we'll fill in the dollar amount is it the percentage of revenue spent with feel how is he measuring fuel costs let's suppose he's concerned that the total amount has risen now the second thing I want to do to confirm the problem is to understand why what exactly is he concerned about so I'm gonna ask have revenues risen accordingly and the reason I ask this is maybe the costs simply have gone up because they're doing they've expanded they've grown their airline's suppose that yes revenues have risen slightly but not proportionally so fuel costs are have risen in a higher rate than revenues which explains the CEO concern right so this is the first step just getting a clearer picture of what the problem is and what we're solving for now the next step to build initial tree is of course to build the first layer and the problem we'll have is a metric is dollar spent on fuel per year or per month and because we have a metric a new memo on numerical variable we want to be numerical if we can in our structure because it's easier to understand where the problem comes from just using data if we did a qualitative structure and started talking about the type of aircraft and the type of feel and the supplier and so on it's hard to quantify where the problem comes through so within the five ways to be missing paradigm which is the way we teach you to be missing you can find more about that at crafting cases comm slash Missy or in a YouTube video called the five ways to be Missy within the paradigm within the framework we're gonna use an algebra structure which is a math equation right so to build a math equation to explain an airline's fuel costs I'm gonna start with the driver of you costs and one driver that I think of one numerical driver that I think of here is the total distance flown so if distance flown by the airline summing all the aircraft's and all the trips if that distance foam has risen we can expect that fuel costs have risen right so this is a numerical driver of fuel cost distance wrong now given that we have this driver how can we complete the formula and make an equation well if we have distance and we want the dollar cost of view we need to multiply distance times the dollar costs per distance so I'm going to multiply total distance flow in kilometers or miles and multiply that with the dollar costs per kilometer or per mile distance will consult with distance and we're going to end up with the dollar cost so this is the simple way to build an equation to break down any numeric problem you find one driver and then you multiply that by a metric that will result in the metric that you're targeting so now we have the first layer of our issue tree now I want to build a second layer that's the next step I want to keep a math structure if I can because if I can do a math structure as far as it makes sense I can quantify in even greater detail where exactly the problem comes from so I'm gonna just repeat the process in the second layer if the total distance flown was higher well one driver for that is that we're flying more trips right if we used to do a thousand trips per year or whatever and now we're doing 10,000 trips per year of course the distance form has risen and of course the fuel costs have risen now if we have the number of trips flown and we want to get to total distance what's left is the average distance per trip so if we multiply a number of trips times the average distance per trip we get total distance flown and thus if either of those have risen then total distance one has risen so same process to build the second layer and now we're gonna do that in the bottom bucket dollar costs of fuel per kilometer or per mile flown right so for fuel costs we're either using more fuel per kilometer so more liters or more gallons of fuel per kilometer or per mile flown or the dollar per litre has risen so the price so this is the classic price times quantity framework but I don't want you to think in terms of framework I want you to think in terms of drivers if we're spending more fuel per kilometer we're using more fuel or we're paying more for the same quantity of fuel so so far we have an issue tree that's mathematical all the way up to the second layer and this is awesome because the structures meaningful we're actually analyzing things that make sense to analyze and that would explain a rise in fuel costs but also it's open - fireball right we could ask data to the interviewer if we're in the case interval or we could get real-world data if we're solving a real problem and we could see how the numbers changed and where the problem is in a real case interview I would probably try to get data here in a real-world problem I would certainly get data here because then I could focus the problem find the problematic driver and then just dig into that but in this exercise I want to teach you how to build issue trees I want to build the third layer here and to do that I'm gonna pick and choose some buckets the first cup of buckets number of trips flown an average distance per trip they're almost self-evident right we don't really need to explain them further if total number of trips long has risen that's the reason why fuel costs have risen so I don't need to dig deeper on that I could try to explain why total number of trips has risen maybe we included new routes maybe it's a regulation but I don't need to same for average distance per trip I could explain that and if that was the problematic buckets I would explain that but just saying to someone well distance per trip has risen that's why our fuel costs have risen that's probably enough right so let's focus on the bottom buckets let's start with litters of few per kilometer foam or gallons of fuel per mile phone if you're American now I can't structure that mathematically me personally because I don't know Aeronautics physics or Aeronautics engineering if we got an engineer here he'd probably be able to come up with a formula so but because I'm not an engineer I don't know the formula let's try a different approach I see two different scenarios where how much fuel per kilometer would rise the first is worse flying the same aircrafts and the efficiency of you has dropped anyway the second is we've changed our aircraft May now if the second one is true then of course that's the reason why litters of your per kilometer foam has risen and we just need to confirm that by confirming two pieces of data the first is to confirm that the new planes are actually less efficient than the old ones we had and the second is to measure if this lesser efficiency fully explains the problem if not then we need to look for other sources of a surge in fuel costs now if we're flying the same aircraft that doesn't really explain anything right so we need to dig another level deeper and to do that it's quite hard how can we explain litres of fuel per kilometer Foam has risen even though we're flying the same aircraft now I'm going to speak in terms of the the language of the five ways to be missing so if you haven't seen that might be a bit confusing but I think you will understand so one option we have is to try to find an equation for liters of fuel per kilometer flow but as I said I can't do that because I'm not a aeronautics engineer or anything like that the second option we have is to do a process structure so maybe we're spending more fuel on takeoff than before or maybe during altitude flight or maybe on landing but my guess is that this is not very insightful because most of the causes I can think of actually would cause a surge in fuel costs in all three of them so the process structure doesn't really work here now another option we have is to come up with a conceptual framework which is the hardest missi technique to use but I'm gonna use it here and if you want to learn how to do that check out our free course you can find that in our website it's a free course encase interviews and among other things it teaches you how to build these frameworks now here's my framework maybe the aircrafts are flying with more weight they're heavier thus they're spending more fuel maybe the routes have changed not the average distance per trip but maybe we're doing shorter trips so the takeoff and landing actually spending more fuel per kilometer because in a shorter trip you have to take off you have to land both consume a lot of fuel and have less time in altitude flight so the there's less kilometres to dilute take off or we're fly at lower altitude for example or maybe it's the maintenance we're doing less maintenance our maintenance has gotten worse or maybe it's the flight behavior our pilots are flying faster order I don't know they're flying a certain pattern that spends more fuel so how did I come up with this framework again check out our free course in our website the frameworks chapter specifically but the point is the way I did it is I did an analogy with driving so if my car spending more fuel its heavier or it's doing a different route or the maintenance is worse or I'm driving it in a way that spends more fuel but there is another way even if you can't do an analogy you can't really rely on analogous to build frameworks and it is the landscape technique you find that in our free course and the structure you would have gotten is either the aircraft is different or the route is different or the pilot is different or what's within the aircraft is different which is the weight part of this framer right now let's go down into litters the price per liter has risen again I'm gonna use that Dakota me the opposite words technique to separate different things as I did in liters of fuel per kilometer so we're it's either more expensive because we've switched the type of view or supplier or it's more expensive even though we haven't switched anything and this is my issue tree it's not that complicated but there's a bunch of different techniques that you should use and one thing that I do is I think of one part at now let's do some takeaways here takeaway number one this is not a general rule but one thing you can do is to go quantitative and then go qualitative now you don't always do that if you have a qualitative problem of course you can't break it down in a coincidence of way but if you have a quantitative problem what I recommend you do in most cases is to start breaking it down in a quantitative way because you'll be able to quantify and know with numerical data where the problem comes from and then when doing a quantitative analysis a algebra structure or a formula is not insightful anymore you start using qualitative structures to dig deeper both are important if you skip the quantitative then you're not going to be able to quantify and if you've skipped the qualitative in this issue tree for example you wouldn't have mentioned important things like aircraft mix weight maintenance flight behavior all those things that are important in the real world and you won't consider if you just do quantitative so use both if you can not all problems will allow for that and some constitutive problems will be better structured using a qualitative structure we're going depth into that in our course which again you can find in our website now takeaway number two stop each branch where it reasonably explains the problem so I've mentioned this throughout the video but I'm gonna repeat because this is really important a lot of people ask me Bruno when should I stop my issue tree and if you notice this issue tree has some parts of it stopped at layer two some parts at layer 3 and some parts at layer 4 now of course in a real interview in a real problem-solving situation you can get data and then just dig deeper where you must but if you don't have data and you wanna plan a study or if you're looking into the future into how to reduce fuel costs and then of course the past the data from the past is not very relevant then you should stop each branch word reasonable explains the problem number of trips flown has risen this reasonably explains if this were the case you would reasonably explain why fuel costs have risen we're flying more trips now in the same layer you have litters of fuel per kilometer foam has risen if I say that it doesn't really explain anything if someone asks me well why have fuel costs have risen and I answer while litters of fuel per kilometer has risen or even if I use more worldly language such as the fuel efficiency has dropped I'm still like why have fuel efficiency dropped so in this case we need to move all the way on to the fourth layer on - well we're flying heavier airplane we're doing less maintenance and these reasonable explains the problem so let's take away number two take away number three is you can still go deeper where you need so even though we've stopped building the issue tree where it reasonably explains the problem you can still go deeper right so let's suppose that in this example you did this issue tree and the interviewer said well the problem is that we're flying heavier the airplanes have more weight why might that happen now now we're focusing on the weight factor and because we're focusing on the weight factor we can go deeper on that specific bucket so I could segment different items on the airplane that could be making it heavy we're flying with more people we're flying with more fuel because we're not calculating exactly how much fuel we need so we're carrying more fuel or we're flying with heavier equipment right another example if the interviewer said well the price per litre has risen even though we haven't changed the type of fuel nor the supplier why might that happen I can't say well either their costs has gone up because oil prices have gone up or their margin has gone up because we haven't negotiated well so I can dig it another layer a third example they might ask why average distance per trip has risen say that this is what has caused a surge in fuel costs now if they say that they may ask why why could that have happened and I can say well it could be either because the average distance per trip has risen in the same origin destination pairs which means we're flying a longer route for some reason we're deviating from some country from which we have no access to their airline space or we've chosen a different route I don't know maybe some Pacific Ocean storms or whatever or the average distance per trip has risen because we're flying different routes or we're flying the same routes but we're flying more of the longer routes and less of the shorter routes so you can still go deeper where you need so if you find data that points to some specific part of the issue tree you can go deeper on that but you don't have to go deeper when your tree can explain the problem now at the beginning of the video I told you that I was going to share with you two great resources to go into more depth and learn more about issue trees now the first one of these resources is called the definitive guide to issue trees and I've spent the whole past month write it it's really really really in-depth it has the three best techniques that I know of to create great issue trees it shows you the six main principles that you have to adhere to in your issue trees so that they're really good at solving the problem that they're designed to solve it also shows the tons of examples real mistakes from real candidates with my commentary on why they've made those mistakes it shows you four ways to practice the shoe tree so that you can actually get good at it and there's a bonus chapter at the end showing you five ways you can apply issue trees on the job after you get a job as a consultant or as a an executive in a big company or in your own company as well so to get that resource you can just google the definitive guide to issue trees plus crafts and cases or you can click the link on the description I'm gonna put a link there so you can get it it's really extensive really comprehensive and well I did my best to make the best resource to create issues now the second resource that you might enjoy teaches you how to be missing in each step of the way on building your issue tree and in other situations as well I know the Missy principle something that is really hard to learn to apply and as you've seen in this video it's essential to build really good issue trees you have to be missing in every step along the way in every layer of your tree this resource is called the five ways to be missing there's a video on YouTube on it you can find the video right here or if you prefer reading you can just google the five ways to be messy or crafts and clay cases craft some cases missing and you're gonna find it now finally there's a third way to learn to build issue trees better which is to to watch the other videos on each of trees examples that I've made on YouTube you can watch another issue tree example right here and as you do that subscribe to the channel below so that you can get the new videos we put out on case interviews and on solving business problems better I hope you've enjoyed this video and I'll see you in the next one

Channel: CraftingCases

Views: 44,222

Rating: 4.9514351 out of 5

Keywords: issue tree, case interview, mece principle, consulting interview, problem solving, issue tree analysis, issue tree examples, issue tree case interview, issue tree mckinsey example, case interview frameworks, case study, issue tree mckinsey

Id: VAsTEvAJQ8g

Length: 23min 55sec (1435 seconds)

Published: Sat Nov 24 2018