Hi,

What is your framework for finding the total number of gas station, McDonald, Starbucks questions?

Thanks

expert

Expert with best answer

Hi,

What is your framework for finding the total number of gas station, McDonald, Starbucks questions?

Thanks

14 answers

Originally answered:

Hi,

I am not completely sure **whether you are talking about the general calculations or a market sizing quant problem.**

If it is the **general math**, you follow these steps:

- Take a minute to think about the problem
- Provide the formula / your approach to calculations / equation
- Check with the interviewer whether it's correct
- Continue calculations, coming back to the interviewer with some intermediate numbers
- Come up with a final number and check it with the interviewer.

If it is a **market sizing problem**, there are several things you should know:

1) First of all, there are **two ways to calculate market size:**

- From supply side (e.g. vending machines spots)
- From demand side (e.g. vending products consumption)

2) There are **2 ways to structure market sizing:**

- Formula - basically a math formula to come up with a solution. The problem with the formula is that it is easy to forget something or get lost.
- Tree - same as with regular cases you build a tree. A very simple example: you need to calculate the number of dogs on manhattan. A number of dogs = share of households having a dog * # of households. # of households = population / average household size. In the end, you'll have a pyramid where you have to fill the numbers on the base of the pyramid. This approach is much easier and help you track all the numbers

3) You should **use the key market sizing techniques:**

- Making assumptions based on personal experiences (Use the example of your house where out of 100 apt-s 10 have dogs)
- Adjusting numbers (NY is a busy city thus fewer people have dogs)
- Sanity check - try to apply your calculations to real environment
- etc.

4) You should **learn and use the key tools:**

- Using age even age split (suppose life expectancy is 80 years. Assuming even age split we have 4 mln people in US of each age group)
- Using 80/20 split (suppose 20% people earn 80% wealth and the average salary is xx...)
- Using approximations (Length of NY-SF flight and plane speed to calculate US length)
- etc.

5) **Learn and use the key numbers**: populations, gas price, gas consumption, Boeing speed and nmber of seats, average salary, # of gates in the airport, GDP growth rate, inflation, etc.

Good Luck!

Originally answered:

Hi,

assuming that with the term „guesstimating question“ you are referring to market sizing questions, please find bit of advice below:

- Most market sizing questions can be
**solved in more than one way**. What does this mean for you?**Just think it through by drawing a tree:**"What are the main factors determining the final answer?" "How can I split each of this factors further?" Go into as much detail as possible. Ideally, you have time to check your final answer by quickly applying a different approach ("do the two approaches give approximately the same answer?") **Make proper assumptions:**Once you have drawn the tree, take a look at the most detailed level of your tree. Are you able to quantify the factors on that level? You do not need to know the precise number, but most of the time, you can relate them to something you have seen / experienced, e.g. you need to determine the number of people living in a given city. How does this city relate to one you have lived in and for which you know the number of inhabitants? Is it of equal size, is it approximately twice as large etc.? The biggest advantage of this approach is that you show your interviewer your reasoning: You do not just come up with a number, you try to derive it. And even if this approach does not work (because you have no idea how large that city you are talking about actually is) your interviewer is a lot more likely to help you.**Practice, practice, practice:**As it is the case with other case types, the more you practice, the easier it gets: You will see that drawing the tree will become a lot quicker and more intuitive. Do as many cases as you think are useful and until you feel more secure.- This will then get you to the point where you will
**lose your fear**and this is probably the most important part. Once you lose your fear and you start to like market sizing cases, you can show your interviewer that you in fact enjoy solving the problem at hand. Not only will this help to solve the case at hand but it will also contribute to the interviewer's positive overall impression.

Hope this helps!

Originally answered:

Hi Anonymous,

these are the steps I would suggest to be able to correctly structure a market sizing upfront:

**Learn the 4 key market-sizing frameworks**. Facing a market sizing case without structure is like facing a business case without it – it becomes almost impossible to solve it. At the bare minimum, the interviewer will feel you are unstructured. For further information on market sizing structures, please feel free to look at my profile or just send me private message.**Take one minute of time to structure your thoughts.**Many candidates think about market sizing as something different from a case – in the sense that it is not compulsory to take time to structure. That’s not the correct approach, since you should always take time at the beginning to structure. If you don’t take time upfront and immediately present a structure, even if your structure is correct, the interviewer will (i) think you have already done such case and (ii) wonder if you are able to correctly structure cases you are not familiar with. While if your structure is wrong, he/she will just think you are too impulsive.**First lay down the structure, then start using numbers.**Numbers are the last step in the process – your overall structure is always the priority. Before using numbers, you should also align with the interviewer on the fact that the structure is fine.

Best,

Francesco

(edited)

my approach would be to match supply with demand.

for the gas station case, on the demand side:

-Start with the total population of the area you're looking into

-get the number of households

-then estimate the number of cars per household (on average 1.5 cars for example)

-now you have the total number of cars

-get the consumption in liters of fuel per car on average

-then you can add an extra 20% for the commercial vehicles (buses, trucks, and so on)

- now you have your total demand of fuel per day or week or month or year.

Supply Side:

-start by sizing the gas station

-estimate the number of pumps (5 usually)

-the time the vehicle take to fill in the tank

-the average quantity in liters each cars fills in

-now you can get the total number of liters that can be filled in an hour, for example, if the pumps are operated at full capacity

-assume a certain capacity utilization of the gas station (say 50%)

-now you got your total supply of fuel per gas station per day or week or month

final answer: divide the total daily demand BY the total daily supply per gas station

Originally answered:

Totally agree with Vlad - it's not at all about the true number, as long as your result is within the realm of possibility. If you say there are 7 or 70,000 gas stations in Paris I would question your judgment. But saying 120 or 300 and basing that on a solid set of assumptions is totally fine.

If I was the interviewee, with this specific case, I would actually try to steer the conversation into a different direction - how the number is changing due to the changes in urban transport (public transport, ride sharing, e-mobility...). Because that is a **strategy** discussion. Trained monkeys can count gas stations, and elementary school children can do the math for your assumptions. But only smart people can have a thoughtful strategy discussion. If you're face to face with the right interviewer, that might score you some points. If he just wants you to do basic math, then you can still go back to doing that...

BTW: Paris, in particular, is tricky, because the 2.2 Million refer to Paris "inter muros", so the actual city. The agglomeration is of course vastly larger, so much of the supply for the city might just be outside of the walls of the city (cheaper real estate, lower taxes etc.). Just like you don't have that many gas stations in Manhattan or the City of London...

(edited)

Originally answered:

Initially, most of my peers approached guestimates using the "general to particular" approach. For example, estimating the market size of Lipsticks in Country A will require to start with the number of women in the country, and then break it down by age, purchasing power, preferences, and so on. In each level of your tree, you will "plug-in" assumptions which will be used to make the calculations and arrive to a final answer.

This approach tends to be very safe since it makes easier to structure the information and allows you to start with some information you might now (population of your country or city, distribution in terms of genre and purchasing power, etc.). The problem is that it usually takes a lot of time and will require you to do a lot of calculations.

Later during my preparation, I had the chance to tackle guestimates with one of my McKinsey buddies and he gave me the following recommendations:

**Focus on the drivers:**Instead of going from the general to the particular, try to understand what would be the 2-3 factors that could determine the 80% of your answer (remember the 80/20 principle) and start breaking down from there. In the lipsticks example, the drivers could be the number of women between 15-60 years old, the average number of lipstick purchase per year per women (this is usually driven by culture, not by income), and the distribution and average price between mass, mass-premium, masstige, and premium lipsticks.**Close enough is more than good enough:**Remember that you don't need to provide an exact number. Usually, you have a margin of error that can go up to a few orders of magnitude (2x-3x). The key in here is to showcase your thinking process and your mathematical abilities, not how good you are at memorizing numbers. These are the skills you will use in your day-to-day as a consultant.**Leverage your interviewer:**Some interviewers play the bad cop and do not give any feedback during the interview. Nevertheless, most of the interviewers will respond to your rhythm. Share your thinking process and the logic behind your assumptions. If you are on the right path, the interviewer can give you a little push to refine your assumptions or can even give you the concise information.

Once you start approaching guestimates keeping these three things in mind, you will find yourself solving guestimates with ease and precision. I hope it helps and let me know if you want to practice some!

All the best,

**Patricio**

(edited)

Originally answered:

Hi Anonymous,

firstly, you should be aware that **it is okay to take a moment** in order to lay out the calculation steps. Just let the interviewer know that you are first taking a moment to outline the necessary calculatory steps, then sharing and aligning these steps with him, and finally executing the calculation (while taking the interviewer along).

This is a crucial skill that consulting interviewers look for in a candidate, becaus guess what - this is exactly how you need to walk clients through your thinking on a day to day basis.

Cheers, Sidi

First, understand what is the "ocean" (e.g. assess the main dimensions of plane, etc.)

Second, understand the unit with which you deal (e.g. ball, ppl, table)

Third, understand the limitations & bottle-necks (e.g. has to be a space for the people to move back & force)

The last, do the math

Important: don't try to be precise. The way you think is much more important than the figures. But don't diminish meaning of figures as well.

Originally answered:

Hi there,

My experience is that you need to practice and in everyday life there are plenty of opportunities to do so.

when you are traveling to work/institute: how busses are there in the city, how many bus stations, how many bus lines, etc....

Out in the evening in some bar/cafe: how many people visit that place every day, every week every month, how much bewereges are consumed, how many similar palces are there in the city, etc...

etc. etc.

First time it will be kind of strange and you will think 'what the hell I know' but you will get better with practice. If you would do it with somebody else will be funnier and more effective :)

In a market sizing exercise, there are 2 broad ways of starting: either top down, or bottom up. __ Top down__ will be more high level, “baby math” (my former boss, an ex-McK, used to run a $5B business w/ back of the envelope / baby math calculations -> very powerful actually).

The other way is ** bottom up**, which can quickly become very complex due the multiple assumptions and calculations. Then again, you will mostly do multiplications here, whereas top-down would largely be divisions.

My own recommendation onmarket sizing exercises is usually to start w/ the top down, and only do the bottom up if your interviewer makes it clear that’s what’s expected. Doing the bottom up from the get go might take longer than you are supposed to take, and also quickly become very tedious / error prone / precisely wrong vs. roughly right. Put it another way… don’t give me a 2x4 to beat you up with, trust that your interviewer is smart enough to make your life miserable

PS: To address your question more specifically now, and assuming you do indeed have to do a bottom-up exercise. I would typically start from the demand (how many people drive how many miles how a vehicle with what kind of efficiency; how many people eat how many burgers / drink how many overpriced lattes); I would then move on to the supply (how many people / cars can one gas station/restaurant serve), before ultimately dividing one with the other

Hope this helps

Originally answered:

Hi,

Usually students struggle because no one has previously explained the basic principles. It makes you panic every time you see a arket sizing question and feel like it is a lot worse than it actually is.

The basic premise of these types of questions is: what is the £/$ value so demand or population size of using a specific product. I like the toothbrush example.

Let's say there are 10mn people in London and 90% use a toothbrush. That is 9mn toothbushes being used, now let's say at a price of £2 per toothbrush, London market size is £18mn.

The above is very basic and there are lots of factors to add. But hopefully this helps as the starting point. The rest is additional info and 'colour' to the story (growth, etc.).

Hope this helps, let me know if you want to practise these.