Guess Estimates Compilation PDF
Guess Estimates Compilation PDF
A Primer
Complied By:
Strategist.
The Strategy & Consulting club of MDI
Table of Contents
GUESSTIMATES – A Primer ........................................................... 3
Few Examples: ............................................................................. 8
1. Surf Excel Detergent Usage in India in a day? .............................. 9
2. How many passengers travel daily in Delhi metro? ..................... 10
3. What is the average number of people at any given time in the
Mumbai airport on a weekday? ........................................................ 10
4. How would you estimate the number of weddings in India in a
year?............................................................................................... 11
5. How would you estimate the number of fruits sold in India in a
day? ................................................................................................ 12
6. Estimate the number of office chairs sold in India? .................... 12
7. How many total gmail users are there in india? .......................... 14
8. Guesstimate for the number of laptops sold in Delhi on an average
routine day? .................................................................................... 14
9. How do I estimate number of ambulances on the road?.............. 15
10. How can the number of (pairs of) shoes sold online in Delhi
through Flipkart in one day be estimated? ....................................... 16
11. What is a guesstimate of YouTube's revenue on the weekends?
16
12. How many t-shirts do you think e-commerce companies selling
in India per day? ............................................................................. 17
13. How many gallons of white house paint are sold in the U.S.
every year? ..................................................................................... 18
14. Estimate the market for bicycles India. ................................... 19
15. Estimate the litres of cola sold in Delhi per day ....................... 20
16. Estimate the number of denims sold in India in a year. ............ 20
17. Estimate the amount of petrol used in Delhi per day ................ 21
18. Estimate the market size of paint industry in India .................. 21
19. Estimate the number of red lights in Delhi ............................... 23
20. Estimate the number of leather balls sold in India in a year ..... 23
21. Estimate the number of one rupee coins that would fit in a
normal size room............................................................................. 25
22. Estimate the number of vehicles crossing the road in front of
MDI. 25
23. Estimate the number of number of subscribers of Economist in
India 26
24. Estimate the number of cars sold by Maruti in India. ............... 26
25. Estimate the number of beers sold in Goa every day. ............... 27
26. Estimate the market for air conditioners in Bangalore in 1 year.
27
Additional Examples ....................................................................28
GUESSTIMATES – A Primer
What?
Why?
• The approach to the answer is more important than the answer itself
So, you walk into the interview room, exchange greetings and introduce yourself.
You talk about your work and your hobbies and there you go, you get your first
guesstimate. What follows below is an example of an actual guesstimate
presented during summer processes.
Often, they are so off the wall to shake up an otherwise calm and collected
candidate. In our previous experiences, a guesstimate is often related to your field
of work or hobbies. The candidate is expected to keep calm, analyse the situation
and conclude while thinking out loud.
Once you get your question, politely confirm the same with your interviewer. Take
a minute to compose your thoughts and jot down points neatly onto the notepad.
Once you feel you are ready to go, start involving the interviewer. Start with
explaining your approach run him through all the factors you have already listed
down. These factors should contribute to the final answer and also be clear to
provide your assumptions for the same. Take the interviewers’ approval and clarify
the questions he poses. Once done, you may start assigning values, figures for
each and arrive at the final solution.
Key success criteria will be to show how broadly you can think in terms of number
of factors and to arrive at a solution at the end making appropriate assumptions.
Let us look at a few illustrations to get a further perspective –
How?
The target metric is usually spread over different categories. Split the categories
into various sub-categories: up to a level where you can make an informed guess
about the individual contribution of each of the sub-categories towards the target
metric. Combine the contributions of all the categories to get to the final number.
Overall there are three main type of approaches to solve a guesstimate:
Consumption side/Demand side
Production side/Supply side
Random approach
1. Consumption side
Here the question is usually solved using end consumer’s data. For instance, for
calculating number of cars in a city it would be better to start from population of
the city and then following a top to bottom approach. Below are some examples
of where and how this approach is used:
Estimate the total consumption of water every day within the MDI campus
You may start by segregating the consumption into various categories such as
individuals, families, mess & eateries maintenance-related, etc.
Then split the categories, say individuals, into sub-categories such as residents
and visitors. You may split residents further into students and support staff to get
a better idea of the numbers in each sub-category.
Then make an informed guess about the consumption of each individual, say 5
litres for drinking, 30 litres for bathing, etc.; multiply this by the total number of
individuals (say 600 students and 50 support staff) and you’ll get the consumption
of individuals. Similarly, make an estimate about the other categories and add
them up.
Can you tell us the number of basketball in the city of Gurgaon right now?
Here it seems easier to use the The Start Big Approach. So it is easier to look
at various avenues where basketballs could be used and then drill down to each
variable and assign values to it. This is how the candidate ended up solving the
question. Please note there is no right or wrong way to go about it.
Major avenues for basketball –
One could have made it complicated by taking different types of school or colleges
and then assigning a rate of number of balls etc., but as mentioned earlier, idea
is to quickly arrive at a solution. So, we recommend mentioning the factors, but
simplifying the steps at each stage taking the interviewers’ approval.
Production side or the supply side approach is better when making estimations
using demand of a certain product or service is difficult for example estimating
revenue of a restaurant in Cyber Hub, Gurgaon. In this case, we can assume the
total capacity of the restaurant say 100 and then make estimations of the
occupancy rate at any time of the day. To see the method in detail, below is an
example:
How many passengers fly out of Delhi in a day?
Total 102200
As you can see, here if we had started with Start big approach, say population
etc., it would have been very difficult to drill down.
3. Random approach
Here both demand and supply side approaches would not serve the purpose and
you need to devise a random approach according to the question. For instance,
calculating the number of trees in MDI Gurgaon or calculating weight of a big
substance. Below is an example of the approach:
The first step is to clarify certain elements or just make assumptions (informing
the interviewer about them) regarding certain peripheral factors:
Are there passengers on board? No.
Any baggage? No
Are the fuel tanks full or empty? Full.
Any food or beverages on board? No.
Now, just go ahead and calculate the weight of each part of the plane
8 full fuel tanks: I’ll assume the plane can fly 6,000 miles and uses 10
gallons to the mile. So that’s 60,000 gallons at 2 pounds a gallon equals
120,000 pounds.
18 tires: I’ll assume that the tires weigh 200 pounds each — that’s 3,600
pounds
4 engines: I’ll assume 2,500 pounds each, which adds another 10,000
pounds
2 wings: 200 feet long by 30 feet wide is 6,000 square feet, times a square
foot weight of 5 pounds times 2 wings equals 60,000 pounds.
Interior: 75 rows of seats times 4 feet per row equals 300 feet. Add on the
cockpit, bathrooms, etc. — let’s say around 400 feet long. I assume that
the average weight per foot is 10 pounds, which equals 4,000 pounds
The seats: They number, say, 500 and weigh 10 pounds each, so that’s
5,000 pounds.
Air in the cabin: It’s captured air so we need to add one ton for the air in
the cabin — 2,000 pounds.
The Aluminium exterior: It’s pretty thin and lightweight. If the plane is 400
feet long by 25 feet high, then about 10,000 exterior square feet at 1 pound
per foot equals 10,000 pounds.
Miscellaneous materials: The tail, overhead bins, carpet, stairs, wiring, and
bathroom fixtures add on, say, another 2,000 pounds.
1. Surf Excel Detergent Usage in India in a day?
People at any point in time can be either those who are waiting at the airport for
arrival of the flight or for the luggage to come after departure of a flight.
There are 7airlines and 15 flights suppose on an average per airline making total
no. of flights at any point of time =15*7=105.
Frequency of departure is half an hour, let’s say, this means when a flight arrives,
some people arrive from the flight to the airport waiting for the baggage and some
other passengers at that time are waiting at the boarding point. At the same point
in time, we have both types of passengers waiting in airport. Each flight has a
capacity of 350 people suppose. So, the total no. of waiting passengers will be
105*(350+350) =70k(approx.). Also, we have 10–20 people (4–5 at the boarding
counter,4–5
at the security counter+4–5 at the boarding point out gate) of the airline staff
making it 20*7=140(max).
Cleaning staff of airport, security personnel at the airport and maintenance
technicians would add to another 100. Also, you have restaurants/bars within
airport where people would be serving food. This would be around 50.
I am not taking people waiting at the bar or having food in restaurant here
separately as that has already been considered.
Total=70k+140+50=70.3k(total)
In case where people are involved always start with the country's population.
India's Current Population = 1.285 Billion
Assume that people are linearly distributed from 1 year to 80 years old. People
usually tend to get married at 20-30 years of age. I have taken this window
targeting the rural landscape as well.
So, no. of people in 20-30 years of age will be = (10/80) * 1.285 Billion =160
million
But as we already know that India has comparably large youth population and
there is a tiny
fraction that gets married beyond 20-30 years of age. I would increase this figure
to 180 million.
Since marriage is dependent on the female population and India professes to have
an
average sex ratio of 900 in 1000 males.
Calculating by this approach (0.9/ (1+0.9)) *180 million = 85.2 million female
population that gets married
I have to assume that each person gets married only once and then after reaching
the appropriate answer introduce and error of 2% in the calculation since few
people India go for second marriages.
Also, it will be safe to assume a linear distribution in age people get married since
it is greater than 26 in urban areas and less than 26 in rural areas.
So, the total marriages will be 85.2 / 10 = 8.52 million marriages. Now adding an
approximate error of 2% as mentioned above = 8.7 million marriages in a year.
5. How would you estimate the number of fruits sold in India in a day?
Let’s start by the current Indian population= 1.285 billion. From this figure, we
have to eliminate the population which is unable to afford eating fruits.
According to the figure stated in Wikipedia, 23.6% people are below poverty line.
I would approximate this figure to be 25% which comes out to be approx. 320
million.
So, now net population = 1285 million -320 million = 965 million
Let us categorize people into age groups and assuming a linear distribution of ages
across people (average life expectancy 80 years)
1. 1-15 = 181 million ~ 200 million
2. 15-50 = 423 million ~ 465 million
3. 50-80 = 362 million ~ 300 million
Since our population is has a fair majority of youth population, I have shifted the
figures appropriately.
Now, we have to assume the average consumption of fruits in these age groups.
1. 1-15 = 1 fruit per day = 200 million fruits
2. 15-50 = 2.5 fruits per day = 465 * 2.5 = 1162.5 million fruits
3. 50-80 = 1.5 fruits per day = 300 * 1.5 = 450 million fruits
Total fruits = 1812.5 million
Since this figure is totally based on individual consumption and does not focus on
commercial aspects, it would be fair to multiply this figure by 1.2!
So, net consumption of fruits in a day= 2175 million
We can approach this problem in two ways 1) Demand side 2) supply side I am
going to use demand side:
Total population of India: 1 bn (approx.)
Reach to internet: 40% =400 Mn
Reach of ecommerce companies to deliver product: 3/4th = 300Mn
Assume 50% are male and 50% are female
Male = 150 Mn Female = 150 Mn
Let’s solve for male: Now we will divide on the basis of age group because demand
for each age will be different for example college students will have more t shirt
compare to corporate person & old person.
0–15 yr. = 45 Mn, on an average individual have 4 t shirt -> 4*45=180 Mn
16–22 yr. = 23 Mn, on an average individual have 4 t shirt -> 4*23 = 92 Mn
23–50 yr. = 65 Mn, on an average individual have 3 t shirt -> 3*65 = 195 Mn
50 - 80 yr. = 18 Mn, on an average individual have 2 t shirt -> 2*18 = 36 Mn
Total t shirts owe by men: 180 + 92+195+36 = 503 Mn ~ 500 Mn.
Let’s solve for female: also, device them on the basis of age groups
0–15 yr. = 45 Mn, on an average individual have 2 t shirt -> 2*45=90 Mn
16–22 yr. = 23 Mn, on an average individual have 4 t shirt -> 4*23 = 92 Mn
23–30 yr. = 15 Mn, on an average individual have t shirt -> 3*15 = 45 Mn
30 - 80 yr. = 67 Mn -> we can neglect this section
Total t shirts owe by women: 90+92+45 = 227 Mn ~ 230 Mn
Total t shirts owe by men + women = 500+230=730Mn
Average life of a t shirt = 2 year
Demand per year = 365 Mn ~ 360 Mn
Online platform provides coupons and offers but because of trust factor and fitting
issues, people still in India prefer to buy offline.so we are assuming 30% of people
buy t shirt through ecommerce portal and 70% buy using shops & showrooms.
Sell through ecommerce platform = .3*360 = 108 Mn ~ 100 Mn per year per day
= 100 * 10^6/365 ~ 27,000.
THE “START BIG” APPROACH: If you’re not sure where to begin, start with the
basic assumption that there are 270 million people in the U.S. (or 25 million
businesses, depending on the question). If there are 270 million people in the
United States, perhaps half of them live in houses (or 135 million people). The
average family size is about three people, so there would be 45 million houses in
the United States. Let’s add another 10 percent to that for second houses and
houses used for other purposes besides residential. So, there are about 50 million
houses. If houses are painted every 10 years, on average (notice how we deftly
make that number easy to work with), then there are 5 million houses painted
every year. Assuming that one gallon of paint covers 100 square feet of wall, and
that the average house has 2,000 square feet of wall to cover, then each house
needs 20 gallons of paint. So, 100 million gallons of paint are sold per year (5
million houses x 20 gallons). (Note: If you want to be fancy, you can ask your
interviewer whether you should include inner walls as well.) If 80 percent of all
houses are white, then 80 million gallons of white house paint are sold each year.
(Don’t forget that last step!)
THE “START SMALL” APPROACH: Take a town of 27,000 (about 1/10,000 of the
population). If you use the same assumption that half the town lives in houses in
groups of three, then there are 4,500 houses, plus another 10 percent, which
makes 5,000 houses. Painted every 10 years, 500 houses are being painted in
any given year. If each house has 2,000 square feet of wall, and each gallon of
paint covers 100 square feet, then each house needs 20 gallons – so 10,000
gallons of house paint are sold each year in your typical town.
Perhaps 8,000 of those are white. Multiply by 10,000 you have 80 million gallons.
Reference:
http://www.bc.edu/clubs/bcfa/docs/vault/Brainteasers%20and%20Guesstimates
.pdf
• India has a population of 1.2B; India has ~30% population living below the
poverty line. It would be fair to assume that a negligible proportion of poor
population owns bicycles.
• Of the remaining 0.84M, approximately 50% fall in the age group of 8-40,
the core bicycle using population. This amounts to ~0.42M people.
• Of these, assuming equal distribution between sexes, there are ~0.21M
males and ~0.21M females.
• Here we assume that 75% of men buy at least single cycle in their lifetime
and at least 20% women buy single cycle at some point. That makes a total
of ~0.2M.
• Let’s divide the population further into high users, middle users and low-
income groups. Assume high income is 10%, middle income is 60% and low
income is 30% Assume that the high-income group buys a cycle every 5
year; middle usage buys a cycle every 10-year, low usage buys 1 during
their lifetime
• Calculating, cycles bought in a year = 0.2 (0.1/5 + 0.6/10 + 0.3/30) = 0.2
(0.02 + 0.06 + 0.01) = ~0.02M
• Assume the average price per denim to be 1000Rs. Thus, market size = Rs.
20B
India has a population of 1.2B; India has ~30% population living below the
poverty line. It would be fair to assume that a negligible proportion of poor
population wears denims.
Of the remaining 0.84M, approximately 50% fall in the age group of 6-35, the
core denim wearing population. This amounts to ~0.42M people.
Of these, assuming equal distribution between sexes, there are ~0.21M males and
~0.21M females.
Here we assume that 80% of men buy at least single denim in a year and at least
50% women buy single denim.
Let’s divide the population further into high income, middle income and low-
income groups. Assume high income is 10%, middle income is 60% and low
income is 30% Assume that the high-income group buys 3 denims per person per
year; middle income buys 2 denims per person per year, low income buys 1 denim
per person per year.
Calculating, male denims bought = 0.8*0.21*(0.1*3 + 0.6*2 + 0.3*1) =
0.8*0.21*0.18 = 0.03M; calculating female denims bought = 0.5*0.21*(0.1*3 +
0.6*2 + 0.3*1) = 0.02M. Thus, total = 0.05M
Assume the average price per denim to be 1000Rs. Averaging for branded
products in the 1800Rs pus range and non-branded in the ~500Rs. range. Thus,
market size = Rs. 5B
Answer: Market size of denims bought in India is ~Rs. 5B
17. Estimate the amount of petrol used in Delhi per day
India has a population of 1.2B, India has 70% population living in rural India. It
would be fair to assume that a negligible proportion of rural population uses
paints.
Given an urban population of 36M and an assuming an average family size of 5,
number of households in urban India are 7.2M
Assume that only 75% of urban houses use paint or related products. Some
houses in
India still use lime (chuna) to paint walls. Thus, relevant households are 5.4M
Assuming an average wall space in every household to be 5000 sq. ft, the amount
of wall space painted in Indian urban households is = 27*109
Further, one gallon of paint can be used to approximately paint 200 sq. ft area.
Thus, the number of gallons required to paint these urban household walls is ~
14*107 gallons of paint
Using this figure to represent 70% of the volume of paint used, we can estimate
the remaining volume of paint used by industrial sector, this would amount to
~6*107 gallons of paint
Considering that a household gets painted once in every 2 years, annual volume
of paint required = 10*107 gallons of paint
Considering industrial sector gets painted once in 4 years (average), annual
volume of paint required = ~3*107 gallons of paint
Hence, the total annual consumption is approximately 13*107 gallons of paint
across all segments.
Adding 10% to this for rural consumption and another 10% for painting of newly
constructed households, we get a total consumption of 16*107 gallons of paint
Assuming an average price of Rs 1000 per gallon of paint, the market size of paint
annually in India is approximately Rs. 16*109
Let us assume the length of the plane to be ~70m, internal width of the plane can
be calculated taking into consideration the seat and gallery space, given a 3*3
configuration of seats, each set of 3 seats would approximately be 2.5m wide.
Thus, internal cabin width is approximately 6m. Similarly, the height of cabin can
be estimated – the cabin allows a 6ft. man to stand comfortably, cabin height is
approximated at 7ft. or 2m
The cabin space represents a cuboid with rounded edges. Estimating the volume
of the cuboid, we arrive at 840 cu.m
Correcting for loss of volume due to rounded edges, decrease this by 10%, arrive
at cabin volume of ~760 cu.m
Assuming cockpit, food gallery, washroom areas cumulatively account for ~10%
of cabin area, estimated volume of plane is ~830 cu.m
Assuming cargo space to be of dimension of 2m (height), 8m(width) and 10m
(length), cargo space volume is ~160 cu.m
Adding up, approximately 1000cu.m of water can fit in a Boeing 747
Working assumption – Number of red lights refers to the traffic signals on the
roads of Delhi, not any other lights which are red in colour!
We know that Delhi is approximately 1600 sq.km in area. This is pretty much
evenly divided between North, South, East and West
Imagine Delhi to be a square area with dimensions 40*40 as length and breadth.
We divide the entire city into blocks of equal area, hence, 20 blocks from east to
West and 20 blocks from North to South (with each block representing 2 sq. km)
Every intersection of four adjacent blocks would give us a cross road. Each
crossroad would have a traffic signal. Hence, the number of signals at crossroads
would be 400 across Delhi, averaging out for different areas.
Since there are 400 crossroads in Delhi, the number of red light at each crossroad
is 4.
Hence, the total number of red lights at crossroads in Delhi is 1600
Let us assume that there are 20% additional red lights to the ones found at
crossroads that are meant to control traffic on straight roads. These would amount
to ~300 red lights.
Answer: 1900 red lights in Delhi
India has a population of 1.2B. Assuming almost equal distribution between both
sexes, India has 0.6B males. Here we assume that a very small fraction of girls
play cricket and can be ignored for this calculation.
India has 50% population below the age of 25 years. Applying this ratio to male
population, there are 0.3B men in India below 25 years of age.
Assuming equal distribution of people in the age 0-24, there are approximately
0.2B males in the age group 10-25.
According to the World Bank, ~50% of Indian population lives on less than US 2$
per day, or approximately less than 100Rs per day. Given the approximate cost
of a cricket leather ball is 150-200 Rs, we can assume that a very small fraction
of leather balls is bought by this segment.
This reduces our target set of potential customers to 0.1B males in the age group
10-25
In a cricket crazy nation like India, let us assume 75% of these have played cricket
at some point in time. This gives us 75M cricket players in India.
Of these, let us assume 10% have played with leather balls and 5% have bought
a leather ball ever in their lives. Thus, approximately 3.8M Indians have ever
bought a leather ball.
Of these, we can further assume that 50% are one-time purchasers, 40% are
sporadic purchasers and 10% are regular purchasers.
Of the one-time purchasers (~2M), distributing over 20 years of life at an average,
yearly purchase is equal to 1,00,000.
Of the sporadic purchasers (1M), assume they buy a single leather ball every 2
year. Thus, every year they buy 5,00,000 leather balls.
Of the regular purchasers (0.8M), assume they buy 2 leather balls per year. Thus,
they buy 16,00,000 leather balls yearly. (These include the professional players
and team members of various cricket clubs – assume, at an average, the
association/club buys 2 balls per payer per year)
Accounting for all kinds of purchasers, total leather balls bought in India in a year
= 22,00,000
Adjustments:
Adding 5% of this number to account for balls used in national/international
cricket matches = 1,10,000
Also, adding 2% for balls bought by those with income below $2 per day and for
girl cricket enthusiasts = 44,000
Total = 22,00,000 + 1,10,000 + 44,000 = ~ 23,50,000
Let’s assume the dimensions of a normal size room to be as follows, Height = 3m;
Length = 4m, Breadth = 5m Volume of the room = 60 m3
Consider a one-rupee coin to be a square of area 2cm * 2cm. (Since when two
one-rupee coins are kept adjacent, the space in between will remain unused and
effectively, the space taken up by each coin would equal the space taken up by a
square of side equalling the diameter of the coin) o Let the thickness of the coin
be 2mm = 0.2 cm = 0.002m o Volume of the coin = 0.02 * 0.02 * 0.002 =
0.0000008 m3
Number of coins that would fit into the room = volume of room/volume of one-
rupee coin (incorporated as a square) = 60/0.0000008 = 7.5 * 107
Answer: Number of coins that would fit a normal size room are 7.5 * 107
Working assumption: A normal working day in DU north campus with all colleges
open. People using vehicles are Students, Teachers, Staff, Residents and Visitors
while types of vehicles are Cars, two wheelers, autos/ rickshaws, Buses, cycles
Estimating the traffic based on time:
o 0000 - 0600 hours: 5 vehicles per hour (2 cars, 2 two wheelers and 1
cycle/rickshaw/ auto) buses won't run at that time
Total vehicles: 30 o 0600 - 0900 hours: 10 vehicles per hour (3 cars, 3 two
wheelers, 3 cycles/rickshaw/ auto and 1 bus)
Total vehicles: 30 o 0900 - 1800 hours: 100 vehicles per hour (30 cars, 30 two
wheelers,
35 cycles/ rickshaw/ auto, 5 buses)
Total vehicles: 900 o 1800 - 2200 hours: 50 vehicles per hour (10 cars, 20 two
wheelers, 15 cycles/ rickshaw/ auto, 5 buses)
Total vehicles: 200 o 2200 - 2400 hours: 10 vehicles per hour (5 cars, 5 rest) no
buses.
Total vehicles: 20
Answer: Total number of vehicles crossing the road in front of MDI in the day:
1200 approx.
Answer: Total Maruti cars sold in India in a year= 60,000 + 72,000= 1,32,000
Additional Examples
http://www.census2011.co.in/states.php.
http://www.indiademographics.com/default.aspx