Page 1 of 1
Consulting Case Interview Preparation
Posted: January 31st, 2013, 12:10 pm
Pick up a copy of Business Doctors. A fun and crazy story set in the world of management consulting and crime.
If you are targeting Management consulting jobs, you need to be a pro at cracking Case Interviews.
To the untrained eye (and brain), these might seem like extremely tough puzzles.
With practice you'll get better, your analytical skills will become sharper, and you'll get closer to your goal of becoming a high-flying management consultant.
If you know very little about the field read these introductory posts first:
Management Consulting 101
1. What is management consulting?
2. Types of management consulting jobs
3. Top 10 strategy consulting firms
4. Top 10 technology, HR, financial consulting firms
Also, check out this list of the best consulting books.
In this thread we'll help you with your Consulting Case interview prepartion by posting some sample case studies.
Amit (MBADropout) who has been managing the free profile evaluation for managementing consulting jobs will be the master of ceremonies here as well.
Explain to us how you will tackle these questions and if you are up for the challenge, give us your final solution as well. If you get stuck along the way, we'll provide tips and advice to help you.
Re: Consulting Case Interview Preparation
Posted: February 6th, 2013, 7:51 am
One of the primary concerns that I hear during recruitment drives is that interviewees haven't had enough exposure to solving cases. The primary intent of this thread is to offer a forum to prospective candidates to give them such exposure.
A simulated environment such as this thread isn't going to prepare prospective candidates / interviewees for all aspects of a case interview - the stress, body language, tone, and such other elements of a real interview are extremely difficult to recreate, even in face to face simulations. However, it should serve as a forum for interviewees to test (and hone) their problem-solving skills, and I will try my best to play the part of a live recruiter, and give (critical) feedback.
To get started, here are a couple of cases from one of the other threads. These are from recent interviews I have conducted:
1. "I own a company that manufactures premium fountain pens, each costing Rs.5,000, to be sold only in India. How would I determine who to target? My marketing budget being Rs.50 lakhs, what mix of marketing channels would I employ? What would be the total sales in 2013 (this question was asked back in 2012)?"
2. The lady that was being interviewed was shivering with fear (cold?). Therefore I followed a more conversational approach.
Amit: "Today is a very cold day (this is from one of the top campuses in North India, and it was a cold December evening). Is it usually this cold in the campus?"
Lady: "At this time it is. We have a generator running for this building - there are frequent power cuts. No lights on the streets - sometimes we have no power for hours."
Amit: "On the topic of power cuts, let's say I am a manufactuer of solar panels. The city (this was a large city in North India) wants to install solar panels atop every street light. How many solar panels would I need to manufacture to be able to cater to the requirements of the entire city?"
I will keep adding more cases to this thread. In the meanwhile, please feel free to take a crack at solving these.
All the best.
Re: Consulting Case Interview Preparation
Posted: February 6th, 2013, 11:29 am
On Sameer's advice, I am re-posting my answer from a previous thread. I hope that we can continue the conversation that we have been having.
1) This is an exciting question. Before I proceed, could you clarify your objective for entering this market? Is it to maximize sales revenue, or profits, or is it something else? Based on the objective, I will have to structure my analysis and offer the right solution.
2) I am assuming this to be a market sizing question. Before solving this, I would like to make some assumptions. They are -
The population of India is about 1.2 billion.
The total length of roads in India is around 4.4 million km. I am considering the target city to only have paved roads. So, I would like to assume that about 50% of all roads in India are paved, which works out to 2.2 million km approximately. Please note that these figures have been rounded off to make my calculation easier.
Using the total population and the length of paved roads, we get an approximate ratio of paved road per person as =
2.2 million / 1.2 billion = 0.0018 km/person.
That means there is about 1.8 km of paved road per 1000 people in India. Again, this is based on the assumption that paved roads are uniformly distributed throughout the country.
Next, I will calculate the total length of paved roads in our target city. I live in a large city (in South India), and I am assuming the average population of large cities in India to be around 8 million. We can now get the total length of paved roads in this city to be =
8 million * 0.0018 km/person = 14,400 km
This is the total length of paved roads in the target city. Now I will calculate the number of street lights in these roads. From this, I can arrive at the total number of solar panels that the city will need (since every street light has a solar panel). Today, while walking to work I noticed that my street had a light for every 30m, on both sides. Assuming this to be constant throughout the target city, I can get the number of street lights to be =
14,400 km * 2 / 30m = 960,000 street lights
Let us round this off to a million. As every light would have a solar panel, the total number of solar panels that would be required = 1 million. To calculate how many solar panels would be required, I am assuming that none of the city’s street lights have any solar panels. Also, I am assuming that you are the only manufacturer and the only supplier to the target city.
If the above assumptions hold true, then based on my initial analysis, you would have to make approximately 1 million solar panels to be able to cater to the requirements of the entire city.
In response to your post, I must say that I am thankful to God for having given me the time and the access to preparation materials. Although I have not studied at a target B-school, I am looking to put in as many hours of practice as I can before an opportunity comes knocking at my door. Looking back now, the two companies I missed were for my good. Had I been short-listed then, I would not have converted them because of my lack of adequate preparation. I now realise that I cannot afford to let such an opportunity slip by just because I did not give my best.
On behalf of all the aspirants, I sincerely hope you would continue to give us your valuable feedback, so that we can improve our performance. As for me, I am simply excited to solve more such cases with you!!
Re: Consulting Case Interview Preparation
Posted: February 6th, 2013, 12:02 pm
Thanks for re-posting your answers here, Rohith. It'll help segregate the general consulting profile evaluation requests from the hard-core technical case interview practice content.
For the benefit of others reading this, I'm copy Amit's feedback on your response.
Before I get started with my feedback to your solutions, please note that there is no single solution - every case can be solved in many different ways, and there could be several "best" solutions.
Now on your solutions to the two cases:
1. Never complicate the question more than it already is. If your question back to me is whether this is a revenue vs. profitability maximization problem, my answer as an interviewer would be, "...let's assume it is both. How will you solve the problem differently to achieve either objective?"
2. Since you have actually tried solving this case, let me give you some detailed feedback:
a. "The total length of roads in India is around 4.4 million km..." - How are you so sure? During an interview, Google isn't going to come to your rescue.
b. Your solution assumes that the population density, density of paved roads, and distance between street lights in India are uniform throughout the country. Those are huge assumptions that takes you away from practicality.
c. "I am assuming the average population of large cities in India to be around 8 million..." - How? Why?
You could have avoided a lot of the apparent "error" in your final solution (~1 million) by making fewer assumptions. Two ways in which you could have done so are:
1. Why start from the country level and then drill down to a city? Why not start with a city?
2. Why follow the paved roads / person approach? Why not a more conventional / alternative approach?
Think simple, execute fast - that will get you to a much better solution.
The good thing is you have arrived at a final number (~1 million solar panels). Hopefully you were able to do so in 15-20 minutes.
All the best.
Re: Consulting Case Interview Preparation
Posted: February 7th, 2013, 12:53 pm
Thank you for your feedback Amit. To maintain the flow of this conversation, I will give you the detailed analysis of the first question followed by my response to your feedback for the second question.
Now that you have given me two objectives (in other words, I just shot myself in the foot :D ), I will give my estimate for the sales, and then I will answer the second part of your question by describing how would things change if the objective was to maximize profitability instead.
Part 1.a - Target Segment
For the first part, since this is a market estimation question I am going to assume that the population of India is 1.2 billion. As our product is a premium fountain pen, my hypothesis is that our target segment is the affluent upper class. In any economy, the upper class comprises of about 2% of the entire population. That gives us (1.2 bn * 2%), which is 24 million out of the total population.
This is the sum of all the people in the upper class segment. India is a young population, which means that there are more people living in the working age (18 to 45). I will assume that 60% of this population is in the working age. Therefore we get (24 mn * 60%), which is 14.4 million. I will round this down to 14 million.
My hypothesis is that our target segment can be narrowed down to focus only on the white-collar segment (the ones who are actually working) as they are the ones who are most likely to buy and use our pens. I will assume that they make up 60% of the upper-class segment. This is because I am considering 40% households will have dependents like housewives and students. Thus we get the total number of our white-collar segment to be (14 mn & 60%), and that is 8.4 million. We will take this figure to be 8.5 million.
As our marketing resources are limited, it is not possible to reach out to all the 8.5 million white collar workers. For the first year, let us limit our marketing activities to the top-tier cities alone. To calculate how many potential customers live in these cities, I will assume the following break-up of the target segment population, based on locations -
Tier 1 Cities - 50%
Tier 2 Cities - 30%
Towns - 15%
Villages - 5%
My reason is that some of the upper-class segments do live in villages, towns and tier 2 cities. They are usually the politicians, rich landlords or the business people. Since we are focussing only on the tier 1 cities for the first year, our potential customers will be (8.5 mn * 50%), which is 4.25 million. To make our calculations easier, we will round this down to 4 million.
Therefore, to answer part (a) of your question - Based on my initial analysis, our target segment is about 4 million. They have been segmented based on four factors - income group, age, profession and location. Firstly, we have considered only the upper class population, which is about 24 million. Secondly, we have looked at those who are in the working age of 18-45. They are about 14 million. Thirdly, we have focussed on the white-collar professionals, and they number about 8.5 million. Finally, we have considered those who are living in tier 1 cities and that gives us our target segment of 4 million people.
Two key issues that I would like to point are two of my assumptions - the income group and the location. If we include the upper-middle classes, then our target segment would expand. Also, if we decide to focus on other locations, then we will have a bigger number.
Part 1.b - Marketing Mix
Now, I will move on to part (b) of your question - to suggest the appropriate mix of marketing channels to employ, given a budget of Rs. 50 lakhs. I will first state the channels I am considering, and then give my reasons for the same. We should use three channels for marketing our product - in-store advertisements, in-flight magazines and the internet. Firstly, we should look at advertising in premium pen stores. These stores are usually located in airports and in malls which house premium brands and they attract a lot of our target segment. For advertising in these stores, we need to invest in posters and in visual merchandising. That way we can convert many shoppers into prospective buyers. Secondly, in-flight magazines can be considered because a considerable proportion of our target customers will travel by air for either business or personal reasons. If we choose the right airline magazine and the right route, we can create maximum awareness, and thereby generate more sales. Finally, I will also consider the internet since internet-sales are growing annually and many in the younger age-groups use the internet a lot.
Now, we need to consider if each of these options are viable and if so, how much do we need to invest in them. I would like to first consider in-store advertising. My knowledge is limited but I am assuming that we have two costs - cost for putting up posters and buying merchandise space. Also, I understand that the cost for advertising in stores in malls and in airports are different, and they also differ by location. I would like to assume that the costs are same for all tier 1 cities. Hence I will consider an average cost for a mall-based store and an airport-shopping centre.
A mall-based store has the following costs -
* A poster of 3' x 2' (i.e. 3 feet high by 2 feet wide) is Rs. 30,000
* Shelf-space in higher visibility area is Rs. 10,000 per sq. ft.
* Shelf-space in lower visibility (i.e. towards the back of the store) is Rs. 5,000 per sq. ft.
An airport-shopping centre would charge higher -
* A poster of 3' x 2' is Rs. 60,000
* Shelf-space in a highly visible area is Rs. 25,000 per sq. ft.
* Shelf-space in a less visible area is Rs. 15,000 per sq. ft.
I will consider that there are 5 tier-1 cities each having 1 premium pen store in a high-end shopping mall.
Each store will have 1 poster (3' x 2') = 30,000 x 1 x 5 = Rs. 1,50,000
Shelf-space in high-visible area (20 sq. ft.) = 10,000 x 20 x 5 = Rs. 10,00,000
Each airport shopping centre will have 1 poster (3' x 2') = 60,000 x 1 x 5 = Rs. 3,00,000
Shelf-space in high visible area (20 sq. ft.) = 25,000 x 20 x 5 = Rs. 25,00,000
Total spend = Rs. 39,50,000/- That will leave us with Rs. 10,50,000 to spend for the other two categories.
Next, we will consider advertising in airline magazines. I will allocate a budget of Rs. 8,00,000/-. Although I am not aware of the costs for ad-space in these magazines, I feel that this figure is a little low for advertising in all routes and airlines. Hence, I suggest that we look at only a few routes or only one airline, whichever fits the budget. Should it still be expensive, we can drop this alternative and consider dividing this amount between the 5 premium pen stores as our marketing budget. This would work out to Rs. 1,60,000 and the stores can spend this amount on buying advertising in newspapers or in fashion magazines. This means that our allocated fund for option 1 will now be revised to Rs. 47,50,000.
For the last channel, the Internet, I think a budget of Rs. 2,50,000 will be sufficient. I will assume a one-time cost of Rs. 1,50,000 for designing and creating a website. Using adwords and other social-media marketing tools will cost us about Rs. 80,000 and we will have to spend Rs. 20,000 on an annual contract for maintaining the website.
Thus to summarize my answer to part (b) of your question, although there were 3 options initally, I think we should look at only two of them, given that our marketing budget is only Rs. 50 lakhs. The two options are - in-store advertising and social media marketing. For in-store advertising, our estimated budget is Rs. 47,50,000. This will cover 5 premium pen stores and 5 airport-shopping centres. Next, I feel that we should spend Rs. 2,50,000 towards social media marketing. This would involve creating and maintaining a website and using efficient marketing tools.
Part 1.c - Estimated sales in 2013
I would like to close my answer by estimating the sales in 2013. Assuming that we have gone ahead with my recommendations for marketing, I will proceed with my estimation. In a year, there are two shopping seasons - Jan to Feb and July to August. I will assume that the average monthly sales during the shopping season will be twice the average monthly sales in the non-shopping season. That means, the total sales will be 4(2x) + 8(x) = 16(x). I will explain this - the shopping season extends for 4 months in a year, and the sales are twice that of the non-shopping season. 'x' is the average monthly sales, which we will calculate next.
To calculate the average monthly sales, I will assume that all our sales has happened only through the stores. My reasoning is that, given that this is the first year of launch our customers will not purchase through the Internet. Next, I will estimate that the stores in shopping malls will sell 1 pen on average daily. Also, the airport-shopping centres will sell 2 pens on average daily. Before I calculate, I understand that what you meant by 'each pen costing Rs. 5,000' is that each pen is priced at Rs. 5,000. That means, the average monthly sales at all premium pen stores will be = Rs. 5,000 x 1 x 30 x 5 = Rs. 750,000. The average monthly stores at all airport-shopping centres will be = Rs. 5,000 x 2 x 30 x 5 = Rs. 1,500,000. The total average monthly sales is Rs. 2,250,000. This is the 'x' that we wanted.
Substituting this in the earlier equation we get, 16 (x) = 16 x 2,250,000 = Rs. 22.5 million + 13.5 million = Rs. 36 million approximately. This is the estimated total sales in the year 2013.
Finally, to summarize my answer to part (c) of your question, I have estimated our total sales in 2013 to be Rs. 36 million. I would like to point out that this figure will change if we use other distribution channels.
Part 2 - Maximizing profitability
If our objective was to maximize profitability, we have two options - maximize revenue or minimize cost. Given that we can only generate a maximum of Rs. 36 lakhs in the first year, we have to look at minimizing our costs to achieve maximum profitability.
For this, we will use the value chain model to analyse where we can cut costs. There are 5 components to it - Inbound logistics, operations, outbound logistics, marketing and servicing. Here are some suggestions that are worth considering -
* As we manufacture our pens, we should consider our operations costs. I would like to look at our P&L statement. Are there any areas where we are costlier than the competition? Could we outsource those areas?
* Although the internet advertisements hardly generate any sales in the current year, we should not drop them. This is because we will lose out on the opportunity to generate sales for the future years through this channel. I think the value of lost-sales in the future will be higher than the Rs. 2,50,000 that we have allocated for this channel. So I will not recommend dropping this channel even though it may not contribute much to this year's sales.
* How about servicing? How much do we spend on after-sales support and attending to service complaints? How much are our competition spending on the same? Can we outsource any part of this component to further minimze our costs?
Kindly give me your feedback on this answer. I will respond to your feedback on the second question in a new post.
Re: Consulting Case Interview Preparation
Posted: February 7th, 2013, 11:03 pm
To respond to your feedback on my second question -
After posting my answer, I was reading 'Case In Point', where Marc Cosentino points out the aspect of practicality. I had not considered that in my earlier solution. I hope you would not mind if I revise my answer using your feedback.
I will re-structure my approach to start from the city-level itself. That way I can avoid many of the glaring errors that came up because of my impractical assumptions. I will adopt the following method to calculate the number of solar panels required -
* Estimate the total road length for the 4 types of roads - small roads, main roads, arterial roads and highways.
* Estimate the total number of street-lights in each of these categories.
* Calculate the number of solar panels to be manufactured.
First, I will estimate the total road length in a typical large city. There are four types of roads in a typical large city - highways, arterial roads, main roads and smaller roads. I have assumed that highways have 4 street lights for every 50 m - 2 in the centre and 1 on either side, arterial roads and main roads have 2 street-lights for every 30 m and smaller roads have only 1 street light for every 30 m.
Next, we will calculate the total length of each of these roads. For this, I will use certain assumptions based on my daily observations. I stay in a residential area in large city, where every small road is about 100 m long and two small roads are connected by a junction. My suburb extends for a total of 6 sq. km. and has 20 small roads, each 500 m long. As there are numerous intersection roads, we will add an extra 5 km. This gives a total road length of 20 * 500 m + 5 km = 15 km.
My suburb has four main roads that span the length and breadth of the area, with each main road being 2 km long. The total road length is 2 km * 4 = 8 km.
Also, there are two arterial roads that connect my suburb with other suburbs. The average length of this road is about 10 km. Therefore the total length will be 2 * 10 = 20 km.
For practical purposes (and to make my calculation easier), I will assume that large cities have a total area of about 600 sq. km. Since a suburb of 6 sq. km. has 15 km of small roads, 8 km of main roads and 20 km of arterial roads, the total length of these roads in a large city will be -
Small roads: 600 * 15 / 6 = 1500 km
Main roads: 600 * 8 / 6 = 800 km (approx)
Arterial roads: 600 * 20 / 6 = 2000 km (approx)
Add to this there will be two national highways spanning the length and breadth of the city. Assuming the city to be 20 km by 30 km, we get the length of the national highways to be 50 km.
Finally, there are state highways. These are usually the expressways. In my city there are 5 highways, each about 30 km long. Their total length will be 150 km.
The total number of street lights in these roads is =
Small roads: 1500 km / 30 m per street-light = 50,000 lights
Main roads: 800 km * 2 lights / 30 m per street-light = 55,000 lights (approx.)
Arterial roads: 2000 km * 2 lights / 30 m per street-light = 130,000 lights (approx.)
National Highways: 50 km * 4 lights / 50m per street-light = 4,000 lights
State Highways: 150 km * 5 / 50 m per street-light = 15,000 lights
The total number of lights required will be = 50,000 + 55,000 + 130,000 + 4,000 + 15,000 = 254,000 lights approximately. As every street light will have a solar panel, the target-city will require about 254,000 solar panels.
While my earlier estimation was 1 million, I have now corrected it to about 250,000 solar panels. That's a reduction of about 75%! I hope this is a more realistic figure. Any feedback would be greatly appreciated.
With regard to timing myself, I take about 30 minutes to solve a candidate-led case during live interviews with practice partners. For interviewer-led cases (the McKinsey type), it varies depending on the difficulty level. I have realised that estimation questions are taking me longer to complete (this one took me about 40 minutes), so I will focus on halving the time.
Re: Consulting Case Interview Preparation
Posted: February 25th, 2013, 10:05 pm
Thanks for the detailed analysis. Here are my comments:
1. Why is 2% of the population affluent? Why not 2.3% or any other number. We are taking about a base figure of 1.2 billion here. A simple "error" might cause large deviations.
2. You have used many assumptions - 60% working class, 60% upper class, breakdown of target segment population across cities/towns, etc. Fewer assumptions is always better (less the "perceived error").
3. Coming to the marketing mix section, which is part 1b in your post, the intent behind this question is to understand the thought process behind choosing the right mix of channels. There is a science behind it - its just not a question of picking the 3 that seem to be most appropriate. The science, in short, is that effectiveness of channels can be measured, and subsequently, the "optimal" mix can also be determined. And there is a lot of math involved behind all this. For an interview situation, I would typically expect the interviewee to think about a way to determine effectiveness of a channel(s). Very rarely I have also found some that can even speak a line or two about "optimal conditions" - that's a huge bonus. All in all, what you have done in part 1b is fine, though I would be more interested in some science/math being behind your choice of 3 channels.
4. "Value chain model" - Consultants talk about these models all the time, but what works finally is some improvisation / modeling that fits the situation in question. If you are absolutely sure that you can lecture a consultant with 20 years in the job about a model, bring it up. If not, don't refer to it. Consultants being extremely aggressive people (the nature of the job makes them aggressive), they would love to surgically dissect any such terms used by an interviewee for an entry level position. Same thought applies to comments such as "...we have two options - maximize revenue or minimize cost..." If you have a consultant in the room with an accounting / finance background, you've had it; he is going to jump right in and start the bi/dissecting process.
5. Now on to your revised post. I am not aware of any case books, though if Marc, whoever that is, wants your solution to be practical, he has made an obvious point.
6. Approach - Remember that you have maybe 20 minutes for a case (the remaining 10 minutes of the half hour being for questions, discussion, etc.). I still feel that your approach - two roads connection by a junction, arterial roads, small roads, etc. - could be simpler. Following your approach, too many permutations and combinations are possible. Typically you should formulate a solution that is simple, does not have too much scope of different possibilities, and can be executed in a very short period of time.
A few recommendations:
1. Simplicity - Keep things very simple, use as few assumptions as you possibly can.
2. No fancy terms - Better not to use fancy terms. The moment you start talking about a model, there is a good chance that someone in the room might know a lot more. Typically this is a situation I have faced with statisticians / mathematicians that we have invited to the panel. The moment some sort of an equation is discussed, they jump right in!
3. Backup with numbers - Everything that you assume or propose should be backed up with numbers.
Let me know if you have more comments.
Re: Consulting Case Interview Preparation
Posted: February 25th, 2013, 10:42 pm
Thank you for your feedback. I have two doubts with regard to your feedback.
1) Is it acceptable to base my assumptions on simple statistics? Like, if India has a young population, can I assume 70%? Or, if it is a common statement that in any economy the top 1% or 2% are the most wealthy, can I state this and then assume it for my calculations? I mean to ask, will assumptions based on common perceptions be considered reasonable during a case interview?
2) Since you suggested that my approach could be simpler, I would like to know at what point should I draw the line between practicality and simplicity, for my solutions. Sometimes the interviewer may expect a little more practicality, whereas some other times (like in this case), the interviewer may be expecting a simpler answer. So how do I know what is the interviewer expecting from me? Also, I am a little unsure on how to avoid the apparent errors that you had pointed out in my solutions.
I have been working with a 30 minute time-slot during my practice interviews. As of today I have had about 40-50 practice interviews and barring a few tough ones (like, typical McK 2nd round cases), I have managed to complete the others within the time frame. I would greatly appreciate it if you could offer me more of these live cases so that I can evaluate myself and know where I stand with respect to the company's standards.
Finally, I would like to acknowledge that I have enjoyed the two cases you have given me. You have challenged my solutions with your feedback and suggestions, and that has helped me realise where I can improve. I hope that you would continue to spare some time off your busy schedule to give me your guidance.
Re: Consulting Case Interview Preparation - End to End Secre
Posted: November 7th, 2014, 4:14 pm
A good page to learn about Case Interview McKinsey
Candidate-led cases are interesting but also very difficult especially for beginners. I personally really like these cases because it represents very well how an actual consulting project works and the role of the engagement team in it.
So rather than talking specifically about how a candidate-led case works, let me introduce you to the LOGIC behind management consulting “Problem Solving Test”. It’s ironic that many experienced candidates, having practiced 20 30 cases, actually don’t grasp this very well. That’s why even with that much practice, they can still struggle in certain cases and still feel like they need even more practice.
Well, practice is good, but only when you have mastered the basics. So here we go: the core logical foundation of how management consultants solve problems. We talked a little bit about this in the Case Interview 101 video but here is the much more in-depth explanation. To make this as easy to follow as possible, I divided the whole concept into bite-size elements with numbers.
Element No.1: The problem has to be defined
Element No.2: To solve a big problem, we – management consultants don’t look for solutions right away. Instead, we try to find the ROOT-CAUSE. This ensures us to completely eradicate the problem and to have a long-lasting impact.
Element No.3: There can be millions of possible root-causes. To effectively and efficiently find the right one, we use a top-down and MECE approach. This is called an “issue tree”. See our MECE and Framework video for more details.
Element No.4: In order for each branch to exist in the issue tree, there HAS to be a chance that the Root-cause is in it. Or in other words, for every branch, there must be at least one hypothesis associated with it.
Element No.5: Now assume we have a structured, MECE, and hypothesis-based issue tree. Pick the best hypothesis, or, in other words, choose a big branch to begin with. Then test if the root-cause is in there.
Element No.6: Depending on each case, different methods are used to test hypotheses. But generally a powerful tool is to use benchmarks. Two main types of benchmarks are historical and competitors’ data.
Element No.7: If data suggests that the root-cause is indeed IN the testing branch, go down one level deeper and repeat the same process: breaking down the branches into sub branches in a MECE, hypothesis-driven way and test each sub-branch using data. See case interview examples and answers
Element No.8: At any point where the data suggests that the root-cause is NOT in the testing branch or sub-branch, move to a parallel branch or sub-branch on the same level.
Element No.9: Keep doing this until the whole issue tree has been covered or until the interviewer would like you to switch gears.
Element No.10: Lastly, once you have identified one or many root-causes, think of solutions to fix them!
In theory, the above approach always works. But the following two conditions must be met.
Condition No.1: each and every single part of the issue tree must be perfectly MECE.
Condition No.2: the issue tree, or in other words, the breakdown must somewhat properly isolate the root-cause.
A few tips to meet those conditions:
(1): Make sure you understand the concept of MECE really well. Please refer to our MECE video for more detail.
(2): Try to improve your business intuition in order to be able to pick good frameworks or correctly draw spot-on issue trees. We devoted a whole eBook in our End-to-end program for this.
(3): Most importantly, develop the habit of aligning with the interviewer. No matter how good you are with the two tips above, there will always be cases that are hard to be MECE inside out and hard to draw frameworks that are spot on. The interviewer is actually a great resource you can use.
You should use Case Interview End-to-End Secrets Program for exclusive practice cases and insightful tips!