Guesstimates – An Introduction

A guesstimate is a combination of guess and estimate. Solving a guesstimate involves using a structured approach to reach the solution while making
sound assumptions, performing back-of-the-envelope calculations and communicating the process effectively during an interview.

Types of Guesstimates Splits & Filters

Market Sizing Unconventional When, guesstimates are solved by using top-down approach, following
This type of guesstimates involve This type of guesstimates may not are various splits commonly used to drill down –
estimation of size of a market or have a defined structure. Such
revenue of a company. This is guesstimates are typically asked in • Rural-Urban Split: Geography based approach
solved by both estimation of interviews to test the approach • Gender Split: Split between male and female
quantity and selection of price for rather than the specific
• Age Split: Split the population into various age ranges
each type of product. calculations.
• Income Split: Based on individual or household income

Methodology Demand & Supply Approach

Top-Down Approach Demand Approach

This approach starts with a large set such as total This approach for market sizing is based on
population and then works on adding different splits estimating the number of customers in the target
and filters making reasonable assumptions to reach to market and units consumed per customer in a given
an estimate. time period.

Bottom-Up Approach Supply Approach

This approach starts with forming an estimate for a This approach for market sizing uses the supply value
specific representative unit and later extrapolating chain to estimate production quantity based on raw
with reasonable assumptions to reach to an estimate. materials or number of factories and further
extrapolating.

(C) Consult Club, IIM Ahmedabad 2023-2024 Page 230

 Guesstimate 1

Estimate the annual demand for Gold Flake cigarettes in Mumbai

Interviewee Notes Facts/Assumptions Approach/Framework

• Begin by estimating the total • Market Share for Gold Flake is 20% Population of
market for cigarettes in Mumbai in Delhi and 22% in Bangalore Mumbai (20 Mn)

• Market Share of Gold Flake in • Assume that 50% of Adults smoke

Mumbai would be like the Market cigarettes
Of Smoking Age Not of Smoking Age
Share in other metro cities such as (25%) (5 Mn)
(75%) (15 Mn)
Delhi and Bangalore • Assume average cigarette
consumption of 1 Pack/Day
• Segment the population into those
7.5Smokers
Mn (50%) Non-Smokers (50%)
of Smoking Age (Adults) and those • Each pack has 20 cigarettes (7.5 Mn) (7.5 Mn)
not of Smoking Age (Children)

1 Pack/Day 365 Days 20 Cigs/Pack

7.5 Mn * 365 * 20 = 54,750 Mn Cigs/Year

Approx. Market Share of Gold Flake = 20%
Demand for Gold Flake = 0.2 * 54,750 Mn = 10,950 Mn Cigs/Year

Observations/Inferences
• This has been solved using the Top-Down approach

(C) Consult Club, IIM Ahmedabad 2023-2024 Page 231

 Guesstimate 2

Estimate the monthly residential electricity consumption in India

Interviewee Notes Approach/Framework

• Clarify which segments are to be
Electricity
included (Residential or
Commercial) as their consumption
patterns are very different. In this
case, focus only on residential. Residential Commercial

• Within the Rural segment,

Urban Population Rural Population
interviewee may consider adjusting (300 M) (700 M)
for homeless people and areas
which are not electrified
Urban Households Rural Households
(300 M/4) (75 M) (700 M/6) (120 M)
• Weighing monthly consumption by
% of households in the income type
and multiplying this by # of Low Income Mid Income High Income Low Income Mid Income High Income
60% 30% 10% 70% 25% 5%
households will give monthly units
consumption
300 1200 3000 100 300 1000
Units/Month Units/Month Units/Month Units/Month Units/Month Units/Month
• Average Household Size is 4 in
Urban areas and 6 in Rural areas 75M x (60% x 300 + 30% x 1200 + 10% x 3000) = 63,000M units 120M x (70% x 100 + 25% x 300 + 5% x 1000) = 23,400M units

(C) Consult Club, IIM Ahmedabad 2023-2024 Page 232

 Guesstimate 3

Estimate the duration of ads streamed on YouTube daily

Interviewee Notes Facts/Assumptions Approach/Framework

• Users are divided based on • Assuming 2 Bn monthly active Monthly Actives
whether they will encounter ads. users on YouTube. 50% of these (2 Bn)
use YouTube daily Daily Users
• They are further categorized them (50%) (1 Bn)
based on the length of videos • Videos with length less than 8
Users Encountering
streamed. mins (Short Videos) only has one Ads (50%) (500 Mn)
ad while Long Videos have 2 ads
Short Videos Long Videos
• The ads are bifurcated into 3 types:
(60%) (300 Mn) (40%) (200 Mn)
Non-Skippable (50%), Skippable and • Assume that 50% users streamed
Skipped (40%), Skippable and Not video with an ad
Non- Skippable Non- Skippable
Skipped (10%) 50% * 1 * 15s = 2250 Mn S 50% * 2 * 15s = 3000 Mn S
• Users prefer Shorter Videos
• Total Users will also include (Less than 8 mins) (60%) over Skippable & Skipped Skippable & Skipped
premium account users Longer Videos (More than 8 40% * 1 * 5s = 600 Mn S 40% * 2 * 5s = 800 Mn S
mins) (40%)
• Day Agnostic - Weekend and Skippable & Not Skipped Skippable & Not Skipped
10% * 1 * 20s = 600 Mn S 10% * 2 * 20s = 800 Mn S
Weekday treated similarly • Assume that each user watches
only 1 video per day
8050 Mn Seconds

Observations/Inferences
• First create the structure, and then put numbers to it
• If the users could be bifurcated into premium and non premium, then different
• Explore if the ad length could vary as per video length and whether a person is a premium account-holder

(C) Consult Club, IIM Ahmedabad 2023-2024 Page 233

 Guesstimate 4

Estimate the number of pairs of shoes sold in Mumbai daily

Interviewee Notes Facts/Assumptions Approach/Framework

• Estimate by finding the • Only covered shoes are Avg. Pairs Life of
Distribution Distribution by Gender & Pairs of Shoes
total number of shoes considered; not including by Income Age
of Shoes Shoes
bought per year
footwear like slippers, flip Owned (Years)
worn by people in
Mumbai flops, crocs, etc.
Men (35%) - 0.7mn 5 1.17mn
Upper
• Segment the total • Shoes can be categorized Class (10%) Women (35%) - 0.7mn 5 3 1.17mn
based on usage: Office 2mn
population based on Kids (30%) - 0.6mn 4 0.80mn
income, gender and age wear/formal, school shoes
etc. Every population Men (35%) - 2.1mn 4 2.10mn
Middle
• Consider appropriate category will have a Class (30%) Women (35%) - 2.1mn 4 4 2.10mn
Total
number of pairs of shoes combination of these based Populatio
6mn
Kids (30%) - 1.8mn 3 1.35mn
owned by people in each on income and need n of
Mumbai Lower Men (35%) - 2.8mn 3 1.68mn
category and their
• BPL population has the (20mn) Middle
expected life Class (40%)
Women (35%) - 2.8mn 2 5 1.12mn
lowest number of shoes per
8mn Kids (30%) - 2.4mn 2 0.96mn
• Assume an average day of person and highest life of
shopping (not affected by shoe Below Men (30%) - 1.2mn 1 0.20mn
weekends, holidays, sales, Poverty
Women (30%) - 1.2mn 0 6 0
etc.) • Kids are the population <15 Line (20%)
years 4mn Kids (40%) - 1.6mn 0.5 0.13mn

Pairs of shoes bought in 1 year = Population * Avg. Pairs of shoes owned / Life of shoes
Total number of shoes bought in Mumbai in 1 year = 12.78mn
Total number of shoes bought in 1 day = 12.78mn/365 = ~ 35k

Observations/Inferences
• Life of shoe has been assumed to be constant in an income category. It is dependent on the desire and need (due to wear and tear) to change
• Sanity Check from supply side: Number of shoes sold by a shop in a day ~5-10 (avg. 8) and area of Mumbai ~600 sq km
• Number of footwear shops in Mumbai per sq km = 35000/8/600 = ~ 7 (reasonable)
• On an average, an upper-class person buys a shoe in ~8 months, middle class person in ~13 months, lower middle class in ~26 months
• Out of the 12.78mn shoes bought in Mumbai in a year, 3.13mn (25%) is bought by the upper class which shows the market size for high end branded shoes in a city
like Mumbai

(C) Consult Club, IIM Ahmedabad 2023-2024 Page 234

 Guesstimate 5

Estimate the size of the Home Renovation Market in Hyderabad

Interviewee Notes Approach/Framework

• Population of Hyderabad is 10M
Population (10M)
• Only considering the residential
market where the customer is
directly getting the work done Households (2.5M)

• Average Household Size = 4

BPL (20%) LMC (35%) MC (30%) UMC (10%) UC (5%)
• Target Segments: Middle Class 500K 875K 750K 250K 125K
(MC), Upper Middle Class (UMC)
and Upper Class (UC) MC UMC UC

• Renovation Cycle: Years after Number of Houses 750K 250K 125K

which renovation takes place is 15 Renovation Cycle (Years) 15 12 10
for MC, 12 for UMC and 10 for UC
250K/12 = 125K/10 =
Number of Houses for Renovation Per Year 750K/15 = 50K
20.8K 12.5K
• Average House Price: Rs. 25L for
MC, Rs. 75L for UMC and Rs. 150L Average House Price Rs. 25L Rs. 75L Rs. 150L
for UC Renovation Spend % 6% 8% 10%

Renovation Expenditure Per House Rs. 1.5L Rs. 6L Rs. 15L

• Renovation Spend as a % of House
Price: 6% (MC), 8% (UMC) and Renovation Market Rs. 750 Cr Rs. 1250 Cr Rs. 1880 Cr
10% (UC)
Market Size = Rs. 3880 Cr

(C) Consult Club, IIM Ahmedabad 2023-2024 Page 235

 Guesstimate 6

Estimate the size of the Maternity Wear Market in India

Interviewee Notes Approach/Framework

• Single pregnancies only – no twins/ triplets Number of Children Born Per Year
28M
• Product used only for 1 birth cycle – no
reuse from the previous births
Urban Rural
• 28M children born per year 33% (9M) 67% (19M)

• 33% babies born in Urban and 67% born in

Rural Upper Income Middle Income Lower Income
Market Size
10% (0.9M) 50% (4.5M) 40% (3.6M) 2% of Urban
• Very limited Rural demand. Assumed as 2% ₹108M
of Urban demand 50% Penetration 20% Penetration 0% Penetration
(0.45M) (0.9M) (0M)
• Urban Upper-Income Group = 10%, Urban
Middle-Income Group = 50% and Urban
Lower-Income Group = 40% 5 Units Per Delivery 2 Units Per Delivery

• Penetration of 50% for Urban UIG, 20% for Total Market Size
Average Garment Average Garment (Urban + Rural)
Urban MIG
Price ₹1,600 Price ₹1,000 ₹5,508M

• Assume only 1 variety of garment in the

market. Average price is Rs.1600 for UIG Market Size Market Size
and Rs.1000 for MIG ₹3,600M ₹1,800M

(C) Consult Club, IIM Ahmedabad 2023-2024 Page 236

 Guesstimate 7

Estimate the Annual Revenue of a Multiplex

Interviewee Notes Approach/Framework

• Three Main Drivers of Revenue: Ticket Revenue Per Screen
(Weekday)
Sales, Parking Charges and F&B orders

• Assume 4 Screens Per Multiplex Ticket Sales Parking Food & Beverages

• Assume 300 Seats Per Screen

Footfall * Ticket Price * Footfall * Parking % * Footfall * F&B Buyers % *
#Shows #Shows * Parking Ticket #Shows * Avg F&B Price
• Occupancy on Weekdays is 60% and (180 * 250 * 4) (180 * 0.5 * 50 * 4) (180 * 0.5 * 150 * 4)
Occupancy on Weekends is 90% (1.5x) Rs. 1.8L Rs. 0.18L Rs. 0.54L

Weekday Weekend
• Equal Ticket Price of Rs. 250 for all shows
Occupancy 60% 90%
• Parking Charge of Rs. 50 for all shows Revenue Per Screen Per Day Rs. 2.52L Ra. 3.78L

Number of Screens 4 4
• Average Order Value of Rs. 150 for F&B
Total Revenue Per Day Rs. 10.08L Rs. 15.12L
orders
Number of Days 260 105
• Assume 50% of customers use parking Total Revenue Per Day Type Rs. 26.2 Cr Rs. 15.9 Cr
facility and 50% buy F&B
Total Revenue Rs. 42.1 Cr

(C) Consult Club, IIM Ahmedabad 2023-2024 Page 237

 Guesstimate 8

Estimate the Monthly Revenue of Tata Play subscribers in Delhi

Interviewee Notes Approach/Framework

• Begin by understanding total Delhi Population (20M)
market for TV viewership in Delhi

• Understanding the demographics of Households (4M) Office Subscribers

Delhi (income/affordability)
High Income (20%) Middle Income Low Income BPL
• Making sure that we incorporate
(0.8M) (30%) (1.2M) (30%) (1.2M) (20%) (0.8M)
penetration of TV among income
groups
100% TV Penetration 90% TV Penetration 50% TV Penetration
(0.8M) (1.08M) (0.6M)
• Market share of Tata Play assumed
as 40%, 30% and 20% for High, 40% Market Share 30% Market Share 20% Market Share
Middle and Lower Income groups (0.32M) (0.324M) (0.12M)

• Understanding of Tata Play Low Middle High Total

subscribers as households and Income Income Income Revenue
possibility of multiple connections
in a household No of Subscribers 0.32M 0.324M 0.12M
No of Connections Per Subscriber 1 1.25 1.5 Rs. 59.25
• Only residential subscribers to be
Monthly Revenue Per Connection Rs. 250 Rs. 500 Rs. 750 Cr
considered
Total Revenue Rs. 36 Cr Rs. 20.25 Cr Rs. 3 Cr
• Average household size is 5

Observations/Inferences
• Assume a top-down approach to solve such questions. Start with the basic equation and break into smaller equations.
• Exhaust all possible scenarios and make reasonable assumptions for the same. Clarify with the interviewer at each stage.
• Don’t need to be too precise. Interviewer looks for the assumptions called out and adaptability with Math.

(C) Consult Club, IIM Ahmedabad 2023-2024 Page 238

 Guesstimate 9

Estimate the total amount of toll tax collected annually across India

Interviewee Notes Approach/Framework

• Split of Small, Medium and Large Number of Toll
Vehicles in the traffic is as given in Plazas in India
Table A.
• Toll tax for different types of Approximate distance between consecutive toll plazas ~80km
vehicles is as given in Table B.
• Processing time at a plaza against Land area served by a single toll plaza = 3.14*40*40 sq. km
the time of the day is as given in (Assuming 80km length is shared by 2 plazas)
Table C. Total land area of India ~ 4 million sq. km
Assuming 70% roadways coverage. Serviced area ~ 2.8 mn sq.
Note: The processing time used here km
is including the idle time for the plaza.
Total toll plazas in India = 2.8 * 10^6 / 3.14 * 40 * 40= ~ 560
At night when the load is lower idle
time will be high so average Table A: Load distribution of vehicles types Vs Time of the day Table B: Average toll Tax for each vehicle type Table C: Vehicle Processing time
processing time is higher.
Vehicle 10AM – 6PM – 2AM – Vehicle Type Toll Tax Time of the Day Processing
Type\Time 6PM 2AM 10AM Time
Small ₹100
Small 50% 40% 10% 10AM – 6PM 30 seconds
Facts/Assumptions Medium 30% 50% 20%
Medium ₹200
6PM – 2AM 60 seconds
• Daily toll collection from 1 lane of Large 10% 30% 60% Large ₹400
2AM – 10AM 90 seconds
a toll plaza ~ ₹3.6L
• On an average there are 4 active Table 1: Toll collection in a single lane of a single toll plaza
lanes in a plaza.
Time of the No of Vehicles Total Toll
• Daily toll collection from 1 plaza ~ Day Processed
Average Toll Price
Collected
₹15L
10AM – 8 hrs / 30 sec = ₹(0.5*100 + 0.4*200 + ₹170*960 ~
• Daily toll collection from 560 6PM 960 0.1*400) = ₹170 ₹1.6L
plazas ~ ₹85 Crores
8 hrs / 60 sec = ₹(0.3*100 + 0.5*200 + ₹210*480 ~
• Annual toll collection ~ ₹30,000 6PM – 2AM
480 0.2*400) = ₹210 ₹1.0L
Crores
2AM – 8 hrs / 90 sec = ₹(0.1*100 + 0.3*200 + ₹310*320 ~
10AM 320 0.6*400) = ₹310 ₹1.0L

Total Toll Collection ~ ₹3.6L

(C) Consult Club, IIM Ahmedabad 2023-2024 Page 239

 Guesstimate 10

Estimate the revenue from consumption of Nestle Coffee on a Weekday in Mumbai

Interviewee Notes Approach/Framework

• Four variants of coffee Total population of Mumbai
1. Chicory + coffee = Rs. 2/cup (200L)
2. Only coffee = Rs. 4/cup
3. Freeze-dried coffee= Rs.
Low Income (20%) Middle Income (40%) High Income (20%)
20/cup
40L pax 80L pax 40L pax
4. Coffee pods= Rs. 50/cup
• At-home coffee drinkers – Age<15 and
Age<15 and >60 Age 15-60 (60%) Age<15 and >60
Buyers of coffee sachets/ jars Age 15-60 (60%) >60 Age 15-60 (60%)
(40%) Coffee (5%) (40%)
& coffee pods (40%)
• Roadside stalls– Sales made
to small owners who buy at At home Stalls At home Shops At home Shops
retail price. % drinkers 50% 50% % drinkers 60% 40% % drinkers 20% 80%
• Commercial includes sales for # cups/day 2 2 # cups/day 2 2 # cups/day 2 2
owners of commercial coffee
Price/cup 2 2 Price/cup 2 2 Price/cup 50 50
machines & cafes (franchises
and others) Market Share 1/2 1/2 Market Share 50% 50% Market Share 50% 10%
• Prices for commercial factor Total Revenue 1.2L 1.2L Total Revenue 11.5L 9.6L Total Revenue 120L 96L
quality & implicit service
Total Revenue on Weekday= Rs. 240 L

Observations/Inferences
• This guesstimate estimates revenue from the final consumption of soluble coffee (sachets/ jars/ pods) by Nestle. This includes retail sales through traditional channels
as well as e-commerce. This guesstimate excludes sales of coffee machines, milk, sugar, installation services, etc. Costs are not considered at this stage.
• For age range and coffee preference: the age range 15-30 and 31-60 has been clubbed and the weighted average based on population and preferences is taken
• For roadside stalls, revenue is Re.1/ cup as such stalls generally use ½ the amount of standard serving size. It is assumed that 1 pod = 1 cup.

(C) Consult Club, IIM Ahmedabad 2023-2024 Page 240

 Guesstimate 11

Estimate the number of snapchat streaks shared in a day

Interviewee Notes Facts/Assumptions Approach/Framework

Urban Population
• Estimate the number of daily • A person would send streaks to 520M
snapchat users. someone in India only.
• Consider the propensity to • Only snapchat streaks and no other 0-12 years 12-60 years 60+ years
~90M ~355M ~75M
maintain streaks. snaps is considered.
• Segment the urban population by • 0-12 years age bracket is discounted 355M
age brackets and further by income as they are expected to not have Lower Class Middle Class Upper Class
class. phones and 60+ as they won’t use (20%) (60%) (20%)
• Apply various filters to arrive at the phones much. 71M 213M 71M
guesstimate of snapchat streaks • Life Expectancy = 70 years, No. of
% using phones 60% 80% 95%
sharing in a day people in each bracket is distributed
% using smart phones 30% 80% 100%
proportionately as per life
% using social media 50% 75% 75%
expectancy.
Calculation =159M
• Urban households include 4 people:
% using snapchat 30%
parents and children, In middle and
% mantaining streaks 33%
upper class households, ¾ Calculation 15.9M
members would use social media
# of streaks sent in a day 8
whereas in lower class households
% using % using %

x
% using
x
# of streaks
½ would use it. (basis age and
income).
Formula used Population smart
phones x social
media x snapchat
maintaining
streaks x sent in a day
~140M

Observations / Suggestions
• Urban population = Population of India (1300M) * 40% = 520M
• Number of snapchat users in rural India and those who maintain streaks would be comparatively insignificant, therefore a multiple (1.1) is taken at the end for the sake
of simplicity.
• A person who maintains streaks on Snapchat sends a snap everyday to fulfil the requirements of a streak. Ignoring breaking of streak anytime.

(C) Consult Club, IIM Ahmedabad 2023-2024 Page 241

 Guesstimate 12

Estimate the litres of paint used annually in India to paint private cars

Interviewee Notes Approach/Framework

• Used cars included in calculation to
Total Households in India
account for repair work conducted
1.4 Billion/4 per family = ~350 Million
to make car resalable

Low & Lower Mid Income Upper-Mid Income High Income

• Proportion of new car to used car 75% (~250M) 20% (~70M) 5% (~20M)
sales assumed to be higher for
higher income levels % households
2% (5M) 20% (14M) 50% (10M)
owning cars

• Liters of paint per car is higher for # of years car

7 6 5
high income due to larger sized is used
vehicles and better-quality finishing
Annual
replacement 5/7 = 0.7 M 14/6 = 2.3 M 10/5 = 2 M
• Totals may not match due to
Demand split New: 0 M New: 0.3 M New: 0.5 M
rounding
by new/used Used: 0.7 M Used: 2 M Used: 1.5 M

Liters of paint New: 5 Lit New: 5 Lit New: 5.5 Lit

per car Used: 1 Lit Used: 1 Lit Used: 1.5 Lit

Total litres = (0*5 + 0.7*1) + (0.3*5 + 2*1) + (0.5*5.5 + 1.5*1.5) = 9.2 M

Litres

Observations/Inferences
Additional refinements possible:
• Expanding scope to include commercial vehicles, taxis, buses etc. to gauge size of the larger market
• Segregating rural and urban demand for greater accuracy
• Estimating use for general maintenance and repainting of already running cars
• Accounting for high income families with more than 1 car

(C) Consult Club, IIM Ahmedabad 2023-2024 Page 242

 Guesstimate 13

Estimate the number of bullets fired in India annually

Interviewee Notes Facts/Assumptions Approach/Framework

• India’s population is rounded off to • On an average, 60 shots are fired in
1.4 billion one hour Bullets fired by

• Only bullets used in rifle and pistols • Hours in training are higher for new
are included. No other ammunition recruits and lower for existing Non-civilian Civilians

employees for both police and

• Non-material avenues like firing at armed forces
Police Force Armed Forces Licensed Gun Unlicensed
ceremonies, are ignored 1.4M (1 in 1000) 1.4M (1 in 1000) Holders Gun Holders
• Live fire is higher for existing
Sportspersons:
employees as compared to new New: 120 shots New: 300 shots 28K (0.02%)*120
recruits for both police and armed x 250 days x 250 days shots * 150 days Criminals:
Training
Current: 60 Current: 120 Security: 2.8M Negligible
forces shots x 50 days shots x 100 days (0.2%)* 50 shots
* 50 days

• A sportsperson’s single match Sportspersons:

New: 50 shots New: 300 shots 28K (0.02%)* 4 Criminals:
includes 60 shots for a match + matches * 100 700M * 10% gun
annually annually
Live Fire
practice shots Current: 200 Current: 400 shots holders* 1 shot
shots annually shots annually Security: on average
Negligible
• Not calculated figures where
numbers were too small to be Total 8,239 Million 25,718 Million 7,515 Million 7 Million
material

Observations/Inferences
• Based on the 80-20 rule, only non-civilians need to be focused upon extensively
• Non-civilians would include police force and the armed forces
• We can further segment this as new-recruits and old employees

(C) Consult Club, IIM Ahmedabad 2023-2024 Page 243

 Guesstimate 14

Estimate the total number of chess moves played in a day in India

Interviewee Notes Approach/Framework

• Differentiate among the professional,
intermediate and amateur as per the
strength of a chess player.
Total chess games
• Among all chess games approximate played
the proportion of Classical, Blitz, and
Bullet games.
• Estimate the average number of Professional (0.5%) = Intermediate (15%) = Amateur (84.5%) =
5,000 150,000 845000
games of each variant played by each Active players @ 25% =
Active players @ 50% = Active players @ 10% =
type of chess player and also the 2500 37,500 84,500
total number of moves in each types.

Facts/Assumptions Bullet games @ 10 games Total moves @ Bullet games @ 10 games Total moves @ Total moves @
• Total chess players in India is taken per day per active player 40 moves/match per day per active player 40 moves/match 40 moves/match
= 25000 = 1M = 25000 = 1M = 1M
as 1M
• In Bullet game time control is <=1
Blitz games @ 2 games Total moves @ Blitz games @ 2 games Total moves @ Total moves @
min; Blitz >1 & <= 5 min and in
per day per active player 50moves/match per day per active player 50moves/match 50moves/match
classical game >5min. = 5000 = 250,000 = 5000 = 250,000 = 250,000
• Games of higher rating category
players are assumed to be more Classical Games @ 0.1 Total moves @ Classical Games @ 0.1 Total moves @ Total moves @
complicated and have more number games per day per active 60 moves/match games per day per active 60 moves/match 60 moves/match
of moves in general. player = 250 = 15,000 player = 250 = 15,000 = 15,000

• As the strength of player decreases,

they tend to play with higher time
control. Total Moves Played = 8,206,500

• Case counts moves of games only

played in 3 variants i.e. Bullet, Blitz
and classical. Other variants,
puzzles and game with computer is
neglected while doing these
calculations

(C) Consult Club, IIM Ahmedabad 2023-2024 Page 244

 Guesstimate 15
Estimate the number of Injections administered in Ahmedabad
Estimate the number of injections administered in Ahmedabad

Interviewee Notes Approach / Framework

Types of
• Senior citizens are expected to visit Administration
the doctor more often
Professionally
• Children and seniors are most likely Self Administered
administered (by
(by frequent users)
doctor/ pathology)
to get medicines administered
through injections Total Population of
AMD (80 lac)
• Senior citizens would require blood
tests most frequently Children (1- 14 yrs) Adults (15-60 yrs) Senior Citizen (60+)
25% = 20 lac 65% = 50 lac 10% = 10 lac
• Calculations for total injections :
Instance Lower Middle & Instance Lower Middle Upper Instance Lower Middle Upper
[weighted avg. of income dist. and income income income income income income
income upper
injections/ individual * age population] (30%) income (30%) (60%) (10%) (30%) (60%) (10%)
(70%)
Visits to 1 2 4 Visits to 1 4 6
Visits to doctor 1 2 doctor doctor
Facts/Assumptions
% for blood 10% 10% % for blood 10% 10% 10% % for blood 20% 20% 20%
• Population of Ahmedabad is ~80 lac tests required tests required tests required
• Average visits to doctors is 2.5 per % for injections 30% 40% % for 5% 10% 15% % for 20% 30% 40%
year by children (1-14 yrs) in US administered injections injections
administered administered
(non-vaccine). Estimating 40% for Injection per 0.4 1.0
lower income and 70% for upper individual Injection per 0.15 0.4 1.0 Injection per 0.4 2.0 3.6
individual individual
income India. Additional 0.5 1
injections for Number of % needing % diabetic population
• Average visits to doctor by adults Vaccines injections insulin Insulin
(15+) is 2 per year. Average visit is 100 10% 5% 80 lac
patients

4 by senior citizens.
• Diabetic population is 5%. Total Injections: Children (Normal + Vaccine) + Adults (15-60 yrs.) + Senior Citizens + Insulin Patients
= 1.67 * 20 + 0.385 * 50 + 1.68 * 10 + 0.4 * 80
= 101 lac (101,00,000)

Observations / Suggestions
• A normal scenario is assumed, where vaccinations are only applicable to children from 0-14.
• In case of children, injection per individual and vaccine injection are added together.
• % for blood test required and % for injections administered are mutually exclusive and can be added since we are considering the total number of injections.
(C) Consult Club, IIM Ahmedabad 2023-2024 Page 245
 Guesstimate 16

Estimate the number of Electronic Voting Machines required for the 2024 Lok Sabha Elections

Interviewee Notes Approach/Framework

• Estimate the number of Eligible Voters in India
Total Population of India
• Segment Population basis Rural/Urban divide & age (1400 Mn)
• Factor in the propensity of individuals to cast vote
Rural (70%) Urban (30%)
• Find utilization of EVMs basis voting time required ~ 1000 Mn ~ 400 Mn

Assumptions Made 0-18 years 18-60 years 60+ years 0-18 years 18-60 years 60+ years
• Votes cast through ballot paper have not been ~ 200 Mn ~ 700 Mn ~ 100 Mn ~ 80 Mn ~ 280 Mn ~ 40 Mn

considered for the sake of simplicity % with

75% 55% 85% 70%
• Avg. Life Expectancy of an Indian ~ 70 Years valid Not eligible
(~ 525 Mn) (~ 55 Mn)
Not eligible
(~ 240 Mn) (~ 30 Mn)
Voter ID
• Age Brackets (Based on Demographic Dividend)
➢ 0- 18 Years ( 20% of the Population)
Parameter Calculation Methodology
➢18 – 60 Years ( 70% of the Population)
➢60+ Years (10% of the Population) Time taken to cast vote per indv. Around 4 minutes on an average Assumption ( 2-6 mins. to cast a single vote)

• An average Poll Booth operates for 10 hours daily # votes casted per EVM per phase 150 votes (600 mins / 4 mins/vote) Polling Booth Op. Hours / Time taken per indiv.
(8AM to 6 PM) => 600 minutes
# votes per EVM – based on 750 votes per EVM # Election Phases * # Votes Casted per EVM in
• # of Election Phases (Days) – Assumed to be 5 phases (5 Election phases * 150 votes) a single Phase
➢ Depends on population density of a state & # EVMs required ~ 1.13Mn (~ 850 Mn / 750 votes) Population Eligible to Vote / # Votes per EVM
political volatility – Bihar needed 7 phases in 2019
# EVMs (with buffer stock of 10%) ~ 1.25 Mn (1.13 Mn + 10%) # EVMs required + 10%
Elections v/s Goa which needed 2 phases only

Observations/Inferences
• The above figure has been computed assuming a voter turnout of 100%. In reality, Lok Sabha Election voter turnout is generally around 70%. This
implies a voting population of around 600 Mn, which translates into an actual EVM requirement of 800,000 EVMs
• Therefore, the capacity utilization of EVMs in India is (0.8/1.25) => 64%
• The time taken by an individual to cast a vote depends on several hygiene factors such as availability of election officers at polling booth, time taken by
them to complete verification checks and sensitivity of the EVM.

(C) Consult Club, IIM Ahmedabad 2023-2024 Page 246

 Guesstimate 17

Estimate the weight of fries sold by McDonald’s in Delhi-NCR in a week

Facts/Assumptions Approach/Framework 1 McD Store

(in a week)
Geography
• Delhi-NCR Surface Area: Weekday Weekend
50,000 sq. km.
• Equal distribution of outlets
Offline Online Offline Online
(1 store in 5 sq. km. of area);
one outlet represents all others
Rush Hours : 14 Non-Rush Hours : 3
Store & Orders Rush Hours : 4 Non-Rush Hours : 10 Rush Hours : 11 Non-Rush Hours : 6 Rush Hours : 11 Non-Rush Hours : 3

• 100 offline orders in rush (4 X 100) (10 X 50) (11 X 150) (6 X 75) (11 X 100) (3 X 50) (14 X 150) (3 X 75)
hours, 50 in non-rush hours 400 Orders 500 Orders 1,650 Orders 450 Orders 1,100 Orders 150 Orders 2,100 Orders 225 Orders

• 150 online orders in rush Total Orders on

hours, 75 in non-rush hours Weekdays
3,000 Orders x 5 Days 15,000 Orders Total Orders on
3,575 Orders x 2 Days 7,150 Orders
Weekends

• Duration of rush hours and

non-rush hours are different on
Orders Containing Fries 22,150 Orders in a Week
weekdays and weekends, but
(80% of all Orders)
the orders serviced during
Small (30%) - 5,316 Small : 5,316* 80g = 4,25,280g
these windows are the same Medium (50%) – 8,860
Free Fries (5% of all Weight Free Fries : 1,108 * 80g =
17,720 Orders Medium : 8,860*120g = 10,63,200g 88,600g
Orders): 1,108 Orders of Fries
• 80% of all orders contain fries, Large (20%) – 3,354 Large : 3,544*150g = 5,31,600g
within which: 30% small fries,
Total Weight of Fries from 1 outlet : ~2.11 Tonnes
50% medium fries, 20% large
fries
Total Outlets in Delhi Total Area : 1Outlet every Total Weight of Fries from all 10,000 * 2.11
10,000 Outlets Tonnes 21,100 Tonnes
NCR 50,000 sq. km. 5 sq. km. Outlets in Delhi

Observations
• Operating hours for online and offline orders vary, thus for offline orders the outlet is active from 9AM - 11PM (14 hrs) whereas for online orders the outlet is active
9AM – 2AM (17 hrs) Online consist of orders from Swiggy, Zomato and McDonald’s app.
• A lot of orders above a certain order value are eligible for free fries. We have thus assumed free fries for 5% of our total orders (all free fries are small)
• Weight of small fries = 80g, medium fries = 120g & large fries = 150g
• The estimate of the number of McDonald’s outlets in Delhi and the number of orders serviced in an hour could have been separate guesstimates in themselves, however
for the scope of this exercise we’ve taken the aforementioned numbers based on certain assumptions
(C) Consult Club, IIM Ahmedabad 2023-2024 Page 247
 Guesstimate 18

Estimate the number of soldiers living in trenches at any point in WW2

Interviewee Notes Facts/Assumptions Approach/Framework

• Estimate the number of people • Only men are involved in this case Total Population
2.3 Bn
who fought in the war at any as women did not fight in trenches.
point of time • Only 18–45 year olds are enlisted in Female Male
1 Bn 1.3 Bn
the army. Beyond this, people are
• Apply various filters to arrive at usually not drafted.
0-18 (30%) 18-45 (50%) 45+ (20%)
people active engaged in trench • There is a time lag between battles, 390M 650M 260M
warfare but considering it is WW2, this lag
is not much. This time is spent in
From peaceful From countries with From Allies or Axis
• Use more filters to arrive at the planning, preparation, sourcing countries (25%) some involvement (50%) Powers (25%)
162.5M - 325M 162.5M
final answer weapons, recovery etc.
• WW2 was fought through trench Percentage Drafted 10% 50%

• Sanity check can be done warfare initially and later, countries Engagement in active warfare 50% 80%

through area of battlefield, realized that mobile warfare and Persons engaged in warfare 16.25M 65M
distance between trenches, utilization of technology would
taking 10 soldiers per trench as make the war strategy much more
81.25M
an average number of people, efficient.
and ~90% occupancy at any given • Not all battles (within infantry as Infantry (60%) Naval (30%) Air Warfare (5%) Miscellaneous (5%)
point of time well) are trench warfare. There are
other forms of warfare too. Mountain (10%) Trench (30%) Armoured (20%) Jungle (10%) Others (30%)
~14.6M
• Not all soldiers live in the trenches,
Time Spent in
they have rotating shifts (Total = 3). Preparation(10%) Battles (60%) Recovery (20%) Others (10%)

Inactive (2/3) Active Shifts (1/3) ~ 3 million soldiers

Observations/Inferences
• The fighting age is based only on the majority of nations. During the later periods of war, any male teenager capable of handling a rife was enlisted in the army
• The deaths are not accounted for, it is assumed that those who die would have been replaced by the Homefront

(C) Consult Club, IIM Ahmedabad 2023-2024 Page 248

IIMB Guesstimates
2022-23

ICON, IIM Bangalore 146

 Squash balls Guesstimate | Easy | Bain (Buddy)

Estimate the number of squash balls sold in India 1 year, we get, 24 *52 * 3900/2 = 24,33,600 games per year. Does the calculation seem
correct?
Thank you for the problem statement, I would like to ask a few preliminary questions. What
is the time frame we are considering? Yes, you can proceed.
Okay, so we will calculate for casual players. For them, we can take that they play on
You can take over 1 year
average 3 games a week which translates to 156 games for every 2 players, therefore, total
Okay, so squash balls generally are of 3 types, single dot, double dot and others. For number of games played = 156 * 1,00,000 = 1,56,00,000. Thus, the total number of games
simplicity, I will assume a uniform ball for our analysis. Does this assumption seem fair to played would be equal to 1,56,00,000 + 24,33,600 = 1,80,33,600 games. Does this number
you? seem fine?
Yes. Yes, you can proceed with this number
I would start with the population of India i.e., 130 Cr, and segment that into rural (70% = Now, generally 1 squash ball can last for close to 30 games. Thus, the total number of new
91cr) and urban (30% = 39 Cr). For simplicity, I would assume only the urban population for balls needed would be the total number of games by 30 which computes to be 6,01,120 or
my analysis. Does this seem fair? roughly 6L balls. Does this number seem fine to you?
Yes. You may proceed.
Yes, that seems to be a fair number. We can wrap up the guesstimate here.
In the urban population, I would like to divide the population into professional, casual and
non players with the fact the non players will not be relevant to our analysis. Does this seem
reasonable to you?
Yes.
As squash is not a very common sport, we can assume that the number of professional
players (0.001% = 3,900) and number of casual players to be 0.05% ~ 2,00,000. Do these
numbers seem reasonable?
Yes, you may go ahead.
Now, let us assume that the number of games played by professional players (i.e 2) in a day
be 3. Also, they would have a practice component to it which can be estimated to be
equivalent to 1 game. Therefore, the number of games played in a week (6 days of practice)
would come out to be 4 * 6 = 24 games for every 2 professional players. Translating this to

ICON, IIM Bangalore 147

 Squash balls Guesstimate | Easy | Bain (Buddy)

Case Statement • Number of squash balls sales in 1 year

Interviewee Notes Structure/ Framework

• Population of India = 130 cr Population

• Break down into rural (70%)–

urban (30%), further segment
into professional and un-
professional and based on the Rural Urban
number of games played and
the average life of 1 ball, we
can estimate the sales of
squash balls. Professional Casual (0.05% =
Non player
(0.001% = 3,900) 2,00,000)
• Compute the total number of
games played in 1 year and
divide by the life of a ball (30
games = 1 ball) Number games Number of games
per week = 24 per week = 3
• 1 game will require 2 players

• Keep the framework easy and take buy ins from the interviewer
Key Takeaways
• Ensure there are no calculation mistakes

ICON, IIM Bangalore 148

 Fantasy Sports App Guesstimate | Challenging | Bain (Partner)

Can you estimate the total potential market size for a fantasy sports app in India? Following Interesting analysis. Let’s assume the smartphone and mobile internet penetration to be
which, can you also list down a few potential risks to such an application? 100%. Can you investigate the other factors?

Sure. Can I begin by asking a few clarifying preliminary questions? Alright. For an app like Dream11, only those income categories which can afford to pay the
entry fee will be willing to play the game, in addition to determining the frequency of
Go Ahead. entering a competition. Additionally, only specific age groups like teens and young-adults
will be more inclined to compete in these event and even among these, a larger proportion
Just wanted to clarify that by ‘fantasy sports app’ , do you mean apps like Dream11?
of males will compete than females.
Precisely, you can take Dream11 as your benchmark for the further process if you want.
Fair enough. Can you get into a little detail as to how will you analyse into the income
Got it. I also wanted to clarify the business model of Dream11. I know for a fact that in such segments and incorporate it into the analysis? Let’s consider the cost of entering a game to
apps, there is a small upfront fee demanded for entering a contest/event, which requires be INR 200.
selecting players & making a team for which points, which can later be converted to money,
For this, I would consider the average cost of signing up to a contest on Dream 11 (given as
will be awarded based on the performance of the team in real matches. Is that correct?
INR 200) and compare it with savings of an average family, computed on a single day basis,
Yes, absolutely correct. Dream11 has various sporting events for which contests are which will be calculated as a percentage of monthly income of the family. Based on the ratio
conducted, based on real match events. For our analysis, let’s just consider cricket matches. of entry cost to the daily savings amount, I would qualitatively determine the frequency
(average no. of games played per day). Below is the table explaining this process:-
Sure, understood. Finally, I just wanted to know if we have any information regarding entry
fees for a match or the pot money? Also, on what factor should the market size be found? Income Savings as a % of daily
Avg. Monthly Daily frequency of
Hmm…let’s just focus now on the various factors you will consider for market sizing. We can Segment income
Income (INR) entering a contest
incorporate the entry fees cost at a later stage. (Urban areas) (1month = 30 days)
Also, let’s find the potential market size based on number of game entries per day. Lower & Lower =133 (20% of
20k 0.2
Ok, sounds good. I will begin my analysis by dividing India’s population on urban-rural basis. Middle income/30)
Out of a total population of 1.2 billion, about 1/3rd reside in urban areas and the rest in rural =500 (30% of
areas. I will be considering the urban population only for my analysis. Is that appropriate? Middle 50k 0.6
income/30)
Yes, that’s alright. You can proceed ahead. Upper Middle =1300 (40% of
1L 1.2
& Upper income/30)
Sure. Next, I will be looking into the smartphone & mobile internet penetration in urban
Another factor to consider in the above steps is the winning pot money, which will also
India. This would depend on different income classes as well as age & gender categories.
influence the frequency of entering games. I can incorporate that too if you want.

ICON, IIM Bangalore 149

 Fantasy Sports App Guesstimate | Challenging | Bain (Partner)

No, that’s all right. Can you now have a quick quantitative look into how many target
individuals can be there for Dream 11?
Alright, so as I mentioned before, I will be looking into urban India and factoring it on the
basis of income, age-groups and gender. I will be taking the base to be urban India’s
population (1/3 of 1.2B = 400M)

Target age group as a % of Cost-free potential target Frequency factor Total daily potential
Income Class Population Gender factor
total population market (done before) target market
Lower =160M (40%) 100% M, 0% F =20M (160M*0.25*0.5) 0.2 =10M (0.2*20M)
=15-35 age group (25% of total =14.4M
Middle =160M (40%) 100% M, 20% F =24M (160M*0.25*0.6) 0.6
population) (0.6*24M)
Upper =80M (20%) 100%M, 40% F =14M (80M*0.25*0.7) 1.2 =16.8M (1.2*14M)
We get a potential market size of nearly 41 million game entries in a day.
Great. I think we are done with the market sizing part. Can you look into some factors that
can drive the market size up?
Sure. Some of the ways to increase potential market size includes – promoting the app in
rural locations, decreasing of the contests/event entry fees, increasing the winning money
pot, adding more sports/events available as competition.
Nice ideas. Could we also quickly look into some risks for fantasy sports apps?

Absolutely. Some risks associated with the app include the negative perception surrounding
online money-making apps in Indian society, app usage during off-season (for ex., during
cricket calendar's lean period). Additionally, recent court cases filed to declare such apps as
‘betting platforms’ and subsequently ban them, can also negatively affect their image.

Great. That would be all from our side. Thank you.

ICON, IIM Bangalore 150

 Fantasy Sports App Guesstimate | Challenging | Bain (Partner)

• A fantasy sports app (such as Dream 11) which wants to determine its potential market size
Case Statement
• Also need to investigate factors that can drive-up the market size and risks associated with the same

Interviewee Notes Structure/ Framework

Average monthly incomes per person
• Understand the business Population
model of fantasy sports apps,
especially the concepts of
entry fees and winning money
• Use top-down approach, start
from national population and Rural : 2/3 Urban : 1/3
Upper class : 1
then apply filters in a nested Lower class : 20k Middle class : 50k
manner – urban-rural, lakh
0-15 : 15%
smartphones, internet,
income group, age group and
finally gender. 15-35 : 25% Savings Rate : Savings Rate :
• Using the concept of savings
Savings Rate :
20% 30% 40%
for different income
categories and comparing it 35-60 : 50%
with the entry fees for
estimating playing frequency 60+ : 10%
is a very important step.

• Understanding that such apps will primarily be used by the youth in urban areas
Key Takeaways • Comparing savings with entry fees & pot money value is a very useful method to find out the frequency of entering an event across income categories
• Keeping up to date on news helps in understanding potential risks associated with such apps, such as recent court rulings on Dream 11.

ICON, IIM Bangalore 151

 EV Market in India Guesstimate| Moderate | McKinsey (Partner)

Your client is an automobile manufacturing company and wants to enter EV market in India Yes, these numbers looks okay. What do you think the EV market penetration of each of
in next 2-3 years. They have approached you to estimate EV market demand of India in these 3 segments will be in next 3 years?
2025. Please help them in doing this.
Okay, given the overall EV market penetration is very low currently ( less than 2%), I expect
Okay, I have a preliminary question before I begin solving this. When our client is talking these numbers to increase given the infrastructure and acceptability of EV cars is growing.
We can assume that for economy segment, EV penetration will increase to 20% as EV cars
about EV industry, do they have any specific vehicle segment in mind like 2-wheeler, 4-
will become more fuel efficient and affordable with improving battery technology, for luxury
wheeler passenger cars or commercial vehicles?
segment, this number will remain low to around 5% as not much investments are
Let’s focus only on 4 wheeler passenger cars market. happening in this high-end space, also people are looking for style and comfort in this
segment rather than fuel efficiency which is difficult to emulate in short span of time. For
Okay, I would like to state my approach first. I will first estimate the number of cars sold in mid range SUV segment, I will take 10% penetration, as a number in between these 2
India in 2022. I will further project this number to 2025, divide cars sold into different categories. So, the number of EV sold in 2025 will be 6.7 Lakhs units.
categories and apply a EV market penetration multiplier by 2025 to each of those categories Okay, these figures looks okay. How will you verify that these are correct numbers? You
and get a final number. don’t need to do exact calculations, explain this qualitatively.
Okay, your approach sounds good. What do you think is the number of cars sold in India in To verify the correctness of these numbers, we can take benchmarking against other
2022? A rough number will work so do a quick calculation. countries where EV market is more developed than India like China and USA. We can see
how EV cars’ market share grew in different categories in these countries in past 2-3 years
For this, I will start will population of India which 140 crores. Taking average family size of 4, and check if our assumptions are reasonable. While doing this analysis, we will have to be
this give number of families to be 35 crores. Since India is still a developing market for 4 careful about the difference in category distribution and purchasing power numbers and
wheeler passenger cars, I will say around 1% of families buy car every year, which gives an make suitable adjustments in calculations.
estimate of 35L cars sold in this year.
Great, now that you have estimated these numbers, give me 3-4 factors which can
Okay, sound good. Now go ahead with your approach. increase/decrease this number.
Okay, I would like to first divide my answer in 2 buckets namely industry level factors and
Okay, taking 5% CAGR growth rate of the automobile market, I estimate 4W passenger cars
external factors and list couple of points under each bucket. Under industry factors – 1) If
market in 2025 to grow to approx. 40 Lakhs. (= 35*(1.05)^3) . Now I would like to divide this
the industry is able to agree on and set ‘EV standards’ in practice for battery charging and
market broadly in 3 categories on the basis of car price – 1) economy and affordable sedan,
replacement, roadside charging infrastructure will be be built at a faster pace and electric
2) middle range SUVs and 3) premium luxury cars segment. Since Indian market sells mostly
adoption will increase. 2) Investments in battery technology by automobile companies to
economy segment cars and luxury cars are far and few, I estimate the percent market share
reduce apprehensions like high-upfront costs, range anxiety, heating and charging issues will
of these 3 car categories to be 70%, 25% and 5% respectively. Is this a fair estimate?
increase EV penetration.

ICON, IIM Bangalore 152

 EV Market in India Guesstimate| Moderate | McKinsey (Partner)

Moving on to external factors – 1) If the government decides to push for cleaner vehicles to
meet its 45% reduction in carbon emissions by 2030 goal, it may provide more incentives to
buy EVs by increasing subsidies, which can result in a higher number than the estimate. 2)
We have to keep cognizance of economical factors in our calculation, since inflation is on a
rise globally and many experts are predicting a recession in 2023. This can affect India as
well and cars sell take a major hit in recessionary environment. So, this can lead to
reduction in our calculated numbers.

Very good, your analysis was good and succinct. We can close the case now.

ICON, IIM Bangalore 153

 EV Market in India Guesstimate| Moderate |McKinsey (Partner)

• Estimate 4-wheeler passenger EV cars market in India in year 2025

Case Statement • Identify factors which can lead to a positive/negative change in the estimated EV market

Interviewee Notes Structure/ Framework 5% CAGR growth No. of 4-W

passenger cars in
No. of 4-W 2025 (40 L)
• In EV markets, penetration
passenger
will be different across
cars in 2022
different car categories.
Economy segment Mid-range SUVs Luxury Cars
• After estimating any number, ( 70%) (25%) (5%)
its always a good idea to
benchmark and do a sanity 140 Cr
20% 10% 5% EV Penetration multiplier
check. Population
• Try to show your general 40*0.7*0.2 = 40*0.25*0.1 = 40*0.05*0.05 =
knowledge in answers by ÷4
5.6 L 1L 0.1 L
giving example whenever you
are making assumptions on 35 Cr Families
market share, penetration etc. Benchmarking against other markets
6.7 Lakhs
1% buy car
each year Industry related
factors
Factors which can
35 L cars sold increase/decrease EV
in 2022 External Factors
(PESTEL)

• Rather than starting every guesstimate with population, you can think of a closest data that is easy to estimate. For example, I chose number of cars
sold in current year as a starting point in this case.
Key Takeaways • Be succinct in you answers when asked about factors that can increase/decrease any number. First write your answer in the form of framework, then
pick 2-3 points out of it and answer briefly.
• Ask clarifying questions in the beginning to know what exactly is meant by EV market and which vehicle segment is our client focused on in India.
ICON, IIM Bangalore 154
 Snow Melting Liquid Guesstimate | Moderate | McKinsey (Buddy)

Our client has invented a new Snow Melting Liquid Chemical and we need to estimate the Let me help you with some data here. You could assume 60% of the area of US where
amount of annual consumption to get a total addressable market. chemical will be used, this includes your area by climatic conditions and inhabited area.
10mL of chemical is used to melt 1 cubic meter of snow and its typically used for days
I have a few preliminary questions before I jump into the guesstimate, where exactly is the
experiencing snowfall over 50cm which is about 40 days in the year.
liquid used and which geography are we talking about?
Sure. So, this gives us about 1400 billion litres of chemical as the total addressable market
The chemical is used to melt accumulated snow. We would like to look at US.
for the chemical on an annual basis. (Calculations in chart on next page). Is there anything
Sure, I’ll take a minute to come up with an overall approach. else you’d like me to consider.
That sounds comprehensive. We can close the case now.
Sure.
Thank you!
So, the way I’d like to estimate this is based on 1st the amount of snow that needs to be
melted multiplied by amount of chemical needed per unit snow. For amount of snow, I
believe this chemical will not be used for smaller amounts of snowfall, but only when we
see a certain significant amount of snowfall happen. So, we would see how many days on
an average in winter months experience such heavy snowfall, and the inches of snow that
need to be melted multiplied by the share of area in US experiencing heavy snowfall. There
would also be the factor of percentage of area where snow needs to be melted as a lot of
area will be buildings or uninhabited and would not require melting snow. Does that sound
ok?
That sounds like a good approach. Let’s calculate this.
Sure. I am not aware of the area of US. I would guess it is about three times the area of
India. For estimating the area of India, I say we have an approximate kite or rhombus shape
with N-S and E-W lines as diagonals. For N-S distance, I’ve flown from Delhi to Mumbai in
~2hours and with a speed of 500km/hr for an airplane, that is approximately 1000Km. This
should be about one-third the N-S distance between top of J&K and Kanyakumari in the
south. The flight from Mumbai to Kolkata was about 3 hours. The E-W distance would be
1.5 times this. Hence, we have diagonals of approximately 3000Km and 2200Km. This gives
area of India as (1/2)*3000*2200 = 3.3Mn sq. Km. Area of US = 10Mn sq. Km. I’d say about
2-3rd of US experiences heavy snowfall, is that correct? Also, I’m not sure what amount of
snow warrants the use of chemical and how often that much snowfall would happen.
 Snow Melting Liquid Guesstimate | Moderate | McKinsey (Buddy)

Case Statement • Estimate the amount of Snow Melting Liquid Chemical used annually

Interviewee Notes Structure/ Framework

• US area ~ 3x India
% area where Product per unit Required
Area of US chemical Snowfall Snow # Days in yr Answer:
10Mn sq Km required when chemical
• Snow melting chemical used 50 cm 10mL per m3 of is reqd. = 40 1200 Bn Litres
40 days in an year when 60% snow per Year
snowfall > 50 cm
Area of India
3
3.3 Mn sq Km
• % area where chemical
required x %area experiencing
snowfall x %area where snow N-S distance E-W distance
needs to be melted 1/2
(3000 km) (2200 km)

6Mn * (Sq. km = 106 sq.m) x 0.5 m = (3 x106) Mn m3

• Use proxy and ask for more information when not aware of geographic or any other specifics.
Key Takeaways
• Assumptions like “chemical not being used for smaller amounts of snowfall” are good for brownie points, but always take interviewers buy-in for such
assumptions.

ICON, IIM Bangalore 156

 Tennis Balls Consumption Guesstimate| Moderate| Strategy& (Director)

Estimate qualitatively, the total number of tennis balls sold per year in Bangalore How would you proceed with the frequency of matches played?
To estimate that figure, we can take 2 approaches, supply-side or demand-side. Can we go Professional players play at least thrice a week, while amateurs play only during the weekends,
ahead with the demand-side estimation? so a maximum of twice a week. Based on this estimate, we can calculate the number of
matches played per week and extrapolate it to a year. Then estimating the average life of a
Why not estimate from the supply side?
tennis ball, we can arrive at the number of tennis balls sold per year in Bangalore
From the supply-side, there are multiple variables such as the various brands, price range,
varying quality of the tennis balls according to brands and product offerings. Also, Good. Now focus on the demand from cricket.
production of tennis balls takes place at various locations from which they are exported to We can take a similar approach by estimating the number of cricket matches played and
Bangalore, which is relative difficult to accurately estimate. dividing it by the average life of a tennis ball Tennis ball is predominantly used for street
Ok, go ahead with the demand based approach. cricket, i.e., cricket that is played by enthusiasts and amateurs. We can ignore professional
matches since they are played with a leather cork ball.
We can estimate the number of tennis balls sold by first estimating the number of tennis
matches played per year and dividing it with the life of a tennis ball. Ok, how does the life of a tennis ball for cricket match differ from that of a tennis match?
But first, what are the various uses of tennis balls? While for tennis matches, the life of a ball is determined by the average time taken for the ball
Yeah, so apart from tennis, tennis balls are also used for playing street cricket and also used to lose its bounce and surface properties due to wear and tear, but for cricket matches, along
as a play toy for dogs. We can take a similar approach to estimate for cricket and pet uses. with the same reason of wear and tear, there is also the possibility of the ball getting torn or
getting lost. Considering these reasons into account, we can calculate a different life span for
That sounds great, let’s focus on tennis matches first. tennis balls used for cricket matches.
To decide on the number of matches played, we first divide the type of matches as Thanks, we can conclude the case here.
amateurs and professional matches. Professional matches involve officially regulated
matches, played by professional players, while amateurs are those who play as a hobby.
How will you differentiate professional players from amateurs on paper?
There are 3 ways by which we can identify professional players apart from amateurs.
1. Check whether the players are registered with the official tennis association of
Karnataka and/or India.
2. Check the statistics such as number of aces, total points, faults, etc. of all matches
played.
3. Check the frequency of matches played per a standard time period (week/month/year)

ICON, IIM Bangalore 157

 Tennis Balls Consumption Guesstimate| Moderate| Strategy& (Director)

Case Statement • A qualitative guesstimate for the number of tennis balls sold per year in Bangalore

Interviewee Notes Structure/ Framework Number of tennis balls sold per year in Bangalore
• Chose demand-based
approach rather than supply-
based approach Number of tennis balls used
• Number of tennis balls sold =
(Number of tennis matches
played / Average life of tennis Tennis Cricket Pets
ball) + (Number of street
cricket matches played /
Average life of tennis ball for Professional Amateurs Street cricket Professional cricket
cricket)
• Number of tennis matches
played = Number of
(Professional + Amateur) Frequency 3 to 7 Frequency less than 3
matches played matches per week matches per week
• Frequency of matches: Life of tennis ball:
professional = 3-7 times a 1. until wear and tear leads to damage
week, amateur <= 2 times a Life of tennis ball: until wear and tear 2. ball gets torn
week. leads to damage 3. ball gets lost

Key Takeaways • The interviewer was more focussed on the approach taken rather than using quantitative data to arrive at the actual estimate.
• It is better to ensure that problem is broken down exhaustively at each step.

ICON, IIM Bangalore 158

 Global Aircrafts Guesstimate | Moderate| BCG (Partner)

Estimate the number of aircrafts in the air at this moment. Pretty accurate estimate. How will you go about estimating flights in the air at the moment?
This sounds Interesting. Before jumping into the guesstimate, I would like to clarify a few I will first calculate total no. of flights flying in a day from all airports in the world as per the
aspects of the case. By aircrafts should I consider commercial, military, cargo, personal etc. equation (no. of runways) * (flights on peak days+ flights on non-peak days). Can I take few
Planes? Should I focus on aircrafts flying globally? Can I look at it from a supply side lens? seconds to calculate?
Great observations. You can focus on commercial planes flying worldwide. Sure, go ahead.
The world has 7 continents and approximately 10,000 cities in it. I will look at tier 1 and 2 For all countries under consideration the total number of flights in a day comes out to be ~6
cities for my analysis and further classify types of airports into big, medium and small. I will Lacs. One way to calculate number of flights at any moment is to consider domestic flight
then estimate number of flights in air according to the equation =No. of airports* No. of avg. duration to be 1.5 hours and International flight avg. duration to be 3.5 hours and
runways per airport* No. of planes flying in a day. I will then estimate flights in the air at any obtain a total average duration of 2.5 hours. So, for entire day it will be equal to 24/2.5 =
moment by dividing no. of planes by number of times a plan leaves or lands on the runway. approximately 10 instances where the flight is either landing or flying from the runway. To
I will base my analysis to pre-covid era with no travel bans. Does this approach seem viable? obtain avg. no. of flights in air currently I will divide no. of flights in a day by 10. Does this
seem correct or am I missing something?
Yes, all the assumptions seem fair. You can proceed.
Interesting. What is the final figure?
To calculate the number of airports in the world, I will limit my scope to developed and
some developing nations and take about ~30% of the world with an average of 3000 cities At any point in time Flights are either landing, flying or are held in the hangar/ track. So one-
(tier 1 and 2) or ~50 countries approximately in the world that have airports with 100% third will be flying. Hence result is ~20000 flights. If I factor in for climatic and traffic
utilization. Further these countries have close to 6 big, 25 medium and 45 small airports. Is conditions and take about 80% of these end up flying, I will obtain ~16000 flights.
this a feasible assumption?
Great Job. Can you quickly estimate number of passengers in these flights?
Yes, your understanding is correct. Go ahead.
To arrive at this I will multiply no. of flights* seats per flight* occupancy rate
I am assuming the big, medium and small airports to have 4, 3 and 2 runways respectively.
Sure. Good approach.
So total number of runways is 6*4+25*3+45*2 =189 runways. Can I go ahead with this?
Can I assume an average figure of 115 seats in a Boeing-Airbus plane and occupancy rate on
You are on the right track. Keep going.
an average to be 70%?
There is a huge variation between number of airplanes depending on the time of the day. Is
Yes, you can work with that.
it okay to assume peak hours from to 9 am to 9 pm for ease of calculation? Also, in peak
hours can I assume planes take off every 10 minutes Vs. 20 minutes during the non-peak Total number of passengers in a commercial flight on air at any point in time globally is
hours? That gives me 6 and 3 planes per hour during peak and non-peak hours respectively 115*0.7*16000 = approx. 13 Lacs.
and I obtain total no. of flights as 6*12+3*12=108.
Thank you. I think this is a comprehensive analysis covering all aspects. All the very best!

ICON, IIM Bangalore 159

 Global Aircrafts Guesstimate | Moderate| BCG (Partner)

Case Statement • Estimate the number of aircrafts in the air at this moment

Interviewee Notes Structure/ Framework

• Scope of estimate global

• Exclude under-developed
countries
• 6 big, 25 medium and 45
No. of airports No. of runways /
airport
No. of planes / day
= Total planes per day

small airports in a country

• Airport Peak hours from 9 AM
to 9 PM Big Medium Small No. of planes in No. of planes in non-
• Domestic flight average peak hours peak hours
duration 1.5 hours
• International flight average
duration 3.5 hours Total peak hours in a Total non peak hours
• At any point flight either lands day*60/fly in a day*60/fly
or takes off from the runway behaviour behaviour
• Average occupancy rate of

=
flight 70%
Number of Capacity of Occupancy Number of
flights Flight rate % passengers

• It was tricky to estimate number of flights at any point in time, Interviewer was looking for the correct approach instead of accurate numbers.
Key Takeaways • Deciding on solving the case from supply/demand side analysis helps in robust structuring and later helps in sanity check using the other side.
• Do not hesitate to ask as many preliminary questions as possible to avoid obtaining a wrong estimate.

ICON, IIM Bangalore 160

 Perfume Retailer Guesstimate | Challenging | Auctus Advisors (Manager)

You are the owner of a retail outlet which sells luxury perfume. How do you plan the Yes, you are right it will hamper the brand image of the luxury brand, so another option is to
inventory for this perfume ? return it back to the manufacturer, is that a possible option ?
Just to confirm my understanding we are a retailer who just resells the branded perfume Yes that’s possible. Now how will you arrive at the optimal quantity to order ?
through our retail outlets. Obviously, our primary objective is reducing the different kind of
inventory cost and be more profitable. Sure, we can use the newsvendor model to arrive at the optimal quantity to be ordered. Do
we have any data on cost structure and price of the product ? Also, we will need
Exactly, what all kinds of inventory costs can you think of ? distribution of demand, assuming it’s normal distribution can you tell me the mean and
standard deviation of the demand.
There is fixed material cost which is charged irrespective on the amount of inventory
ordered, apart from that there might be an ordering cost, opportunity cost of capital The perfume costs us Rs. 80, and we sell it at Rs. 150. As per the contract with
blocked in inventory, warehouse and inspection cost. manufacturer any unsold perfume is taken back at Rs. 50 after 1 year. The marketing
Sounds good, let’s go ahead and find out the optimal inventory. department has estimated that demand is not normal and there is 20% chance that
demand is 1000, 50% chance that demand is 1500 and 30% chance that demand is 2000.
Since this is not a high demand good, one option is to use periodic order inventory system,
Having a discrete demand makes the problem easier, now we the optimal quantity to order
where we find the optimal time to place an order using EOQ (Economic order quantity)
model and based on the service level required we can find the correct quantity to order. is either of 1000, 1500 or 2000. For each of the case we can find expected profit and choose
the one with highest expected profit. Does the approach sound fine to you ?
Let’s take a step back, can you tell me what are the components of inventory planning ? Yes, this looks great please go ahead with the calculations.
In my opinion inventory planning means , what is the amount of inventory to order and Scenario 1 (1000) : 1000*(150-80) = 70K, Scenario 2 (1500): 1500*(150-80)*0.8 + 1000*
when to order. So basically, we are trying to answer when and how for the inventory. Am I 70* 0.2 - 500*30*0.2 = 95K, Scenario 3 (2000): 2000*(150-80)*0.3 + 1500*70*0.5 –
missing something ? 500*30*0.5 + 1000*70*0.2 – 1000*30*0.2 = 95K. Since scenario 2 and scenario 3 both give
highest profit that is 95K we can go with either of them.
Sound perfect, so let me give you some more information regarding the planning
parameters. We can order perfumes only at the beginning of the year and perfumes go out Ok, if I force you to choose either one of them which one will you choose ?
of trend in next year and cannot be sold. Before we go ahead can you tell me what all can
we do with the unsold perfume ? II will choose 3rd option that is order 2000 as we don’t want to loose our customers to other
retailers as retailing is a highly competitive business.
Next year we can sell the perfume on sale or through other marketing campaigns.
Thank you! we can close the case now.
You do understand that this is a luxury brand so is running sales and discount on these
perfumes a good idea ?

ICON, IIM Bangalore 161

 Perfume Retailer Guesstimate | Challenging | Auctus Advisors (Manager)

Case Statement • Inventory planning for a retailer who sells luxury perfume

Interviewee Notes Structure/ Framework

Inventory
• Plan inventory such that Planning
different kinds of inventory
costs are reduced.
• Planning horizon is 1 year.
• Order can be made only at the How much to
beginning of the year. When to order?
order?
• Unsold perfume is returned to
the supplier at a loss of Rs. 30.
• Selling the perfume earns
Rs.70. Overstocking
Beginning of the Understocking At-par with
• Demand pattern for the (Leftover
year (Shortfall) Demand
upcoming year is given. Inventory)

Loss on Unsold
Lost Revenue
Product

• This case can be solved without any knowledge of inventory planning models. It was a simple probability based case.
Key Takeaways • More clarifying question like planning horizon should be asked before moving into solution.
• Case could have been structured from beginning.

ICON, IIM Bangalore 162

GUESSTIMATES
Guesstimates are estimates based on a mixture of guesswork and calculation. They can be given independently or as a part of a case. The
interviewer is generally looking to assess:
• Structure of your approach
• Comfort level with numbers and quick calculations
• Ability to make back of the mind checks to validate the numbers

As a part of the guesstimate you will required to make reasonable and logical assumptions, it is important to state them clearly and seek
confirmation from the interviewer. You will not be expected to calculate the exact value, but a rough number.

Approaches to solving guesstimates:

Top-Down approach
Involves starting with an entire population (in other words, the “top” level) and then breaking it down until you arrive at an answer.

The major elements of a top-down approach are:

• Identifying the correct starting universe
• Applying the right set of filters and relevant conditions. Segmentation can be carried out as follows:
- Demographics (age, sex, income, ethnicity, etc.)
- Psychographics (attitudes, behaviour, values, e.g. smoker vs non-smoker, dog vs cat person, etc.)
- Geography (urban vs rural, Tier I/II/III cities, country, continent, etc.)
- Channels (offline vs online, mobile app vs website, etc.)

169
GUESSTIMATES
Bottom-Up approach
For this approach, rather than starting from the “top” with a high-level figure such as population, the best approach is to start from the “bottom”—
some low-level statistic, such as Revenue per customer, and build your way up to the answer.
Some common bottom-up approaches:
• Household approach
- Example: For a guesstimate involving the number of cars, washing machines etc driving calculation using a household approach is easier
• Population approach
- Example: For estimating the number of mobile phones, chocolates consumed, apparel bought a population approach can be used
• Bottle-neck approach
- Example: For estimating the number of flights per day in a busy airport the number of available runways would serve as a handy
bottleneck to base your calculations. There could also be a limitation on the production capacity of the manufacturer in certain cases

Bottom-Up approaches

Household approach Population approach Bottle-neck approach

170
Residential Electricity Consumption
Problem Statement Guesstimate the monthly residential electricity consumption in India

Case Facts Population of

India (1,400M)
• Calculation to be
done in terms of
units of consumption Urban Population Rural Population
(400M) (1,000M)
• Average family size
for both Rural and
Urban
# of households = # of households =
400/4 = 100M 1000/4 = 250M
• The # of units per
month for each of
the income group
can be assumed High Income Middle Income Low Income (50% High Income (5% Middle Income Low Income (70%
basis some (10% = 10M) (40% = 40M) = 50M) = 12.5M) (25% = 62.5M) = 175M)
calculation such as
the monthly
electricity # of units / # of units / # of units / # of units / # of units / # of units /
consumption of your month = 600 month = 300 month = 150 month = 500 month = 250 month = 150
home
= 6B Units = 12B Units = 7.5B Units = 6.25B Units = 15.63B Units = 26.25B Units = ~75B Units

• While guesstimates are summarized here as a flowchart, even guesstimates discussion flow like any other case discussion
Summary / Takeaways • You are supposed to ask for details or make & clarify assumptions with the interviewer as the discussion progresses

171
Kolkata Airport
Problem Statement Guesstimate the number of daily passengers on Kolkata Airport

Case Facts # of passengers # of flights landing

on Kolkata or taking off
• Number of runways Airport
=2
# of flights
• Domestic / landing / taking
Domestic Flights International
International Flight (70%) Flights (30%)
off
split = 70/30%

• Time between two Depends on the

flights on one Average Capacity = Average Capacity =
# runways = 2
180 300
runway = 6 mins i.e.
3 mins between 2
flights altogether Total operational
since there are two hours of runways Average Average
runways = 20 hours Occupancy Rate = Occupancy Rate =
70% 70%
Time b/w 2 flights =
6 mins (10 flights /
hour)
Total domestic passengers per day = 400*70%*180*70% = ~35,300
Total International passengers per day= 400*30%*300*70%= 25,200
Hence total flights handled in a day= 2*20*10 = 400 flights Hence total daily passengers = ~60,000
• This is based on the bottleneck approach. Here bottleneck was the number of flights that can land/takeoff on one day and basis that daily number of
Summary / Takeaways passengers are calculated
• Here, the implicit assumption is that no passenger travels twice in a day. It is better to clarify this explicitly
172
Electrical Vehicles
Problem Statement Guesstimate the electrical vehicles market size in Kolkata for the next one year

Case Facts Electric Vehicle

Market Size
• The EV market size
of Kolkata is
required to estimate Urban Areas Rural Areas

• Assumed market
penetration of 3%
for EV’s Tier I City Tier II City Tier III City

• Population of major
Tier I cities of India is
Kolkata (1.5 Cr
1 Cr to 2 Cr
Population)

Can Afford Cars Cannot Afford

(40%) Cars (60%)

Average Total number of cars in Kolkata = (1,50,00,000/4)*0.4 =15,00,000

Household Size
Assuming a market penetration of 3% = 45,000 electric vehicles
=4

• Always use a structured approach while solving for the Guesstimates and use clean numbers for assumption for easy calculation
Summary / Takeaways
• Approximating the car affordability can be done on the basis of income split refer to case License Plates (Pg No ) for calculations
173
Fast Food Chain
Problem Statement Guesstimate the daily revenue of a typical fast-food chain in Kolkata

Case Facts
# of items sold
Total Revenue in Dine-In
• Focus on Dine-In
orders only
# of operational Capacity (# of Average # of items per / Average time a
Average Price # of items sold hours seats) occupancy person seat is occupied

Average Price ₹100

Dine-In
Operational Hours 12
Capacity 50
Takeaway Average Occupancy 70%
# of items per person 2
/ Average time a seat is occupied* 30 mins = 0.5 hours
Delivery
Total Daily Revenue ₹1.68L

*The average time every person eats at a fast food restaurant =30
min, therefore 2 (60min/30min) people occupying a seat per hour]

• A similar mathematical approach can be taken to estimate the total revenues via Takeaway and Delivery business
Summary / Takeaways • For Delivery / Takeaway orders, you can include factors such as average order value, any special discounts / cashbacks offered, number of orders in a
hour basis demand
174
License Plates
Problem Statement Guesstimate the number of license plates in the state of Maharashtra

Case Facts Population of

Maharashtra
(120M)
• Only personal
vehicles need to be Assumption: Average /4 /4 /4
considered that is people per household - 4 High Income Middle Income Low Income
ignore commercial Households Households Households
vehicles (10% - 3M) (40% - 12M) (50% - 15M)

% Vehicle % Vehicle % Vehicle

Assumption: Average # of Ownership (90% Ownership (50% Ownership (30%
vehicles per household - - 2.7M) - 6M) - 4.5M)
*2 *1.5 *1

No. of vehicles No. of vehicles No. of vehicles

(5.4M) (9M) (4.5M)
Number of license
plates per vehicle - 2 *2 *2 *2
Number of Number of Number of Total Number of license
license plates license plates license plates plates = 10.8 + 18 + 9 =
(10.8M) (18M) (9M) 37.8M

• It is always better to clarify with the interviewer whether you are missing out on any other place where the license plates could be used
Summary / Takeaways
• Once interviewer confirms, delve deeper. Sometimes, it may happen that the interviewer might choose to not go into depth in certain segments
175
Dinner Plates
Problem Statement Guesstimate the number of Dinner Plates in the state of West Bengal

Case Facts Other areas that have dinner plates:

Population of West
Bengal (100M) 1. Restaurants
• No specific case 2. Commercial areas (Offices)
facts were given
Approach - Restaurant
• Restaurants & 1. Try to estimate the no. of restaurants basis
Commercial Places restaurant either basis demand for restaurants or
need not be delved High Income People Middle Income Low Income People using the restaurant density per sq.km
into. Just specify the (10% - 10M) People (40% - 40M) (50% - 50M)
2. Similarly try to estimate the number of offices
approach basis percentage of population employed
3. Then make a rough assumption for the number of
plates in each of restaurant / office spaces
% Crockery % Crockery % Crockery
Ownership (100% - Ownership (80% - Ownership (60% -
10M) 32M) 30M)
Assumption:
Average # of *8 *5 *3
dinner plates
per person No. of dinner plates No. of dinner plates No. of dinner plates Total Number of dinner
(80M) (160M) (90M) plates = 80 + 160 + 90 =
330M

• Always clarify each step with the interviewer

Summary / Takeaways
• In most cases, interviewer will guide you with data as and when required
176
Decision Tree – Movie Production House
Problem Statement Using Decision Tress, estimate whether a production house should produce and release a movie

Case Facts Produce & Release

Movie?
• All the probabilities
were given as case
facts
High Quality Movie Low Quality Movie
• You just need to Chances (0.6) Chances (0.4)
calculate Bayesian
Probability and
hence the net profits Box Office Hit Box Office Flop Box Office Hit Box Office Flop
Chances Given High Chances Given High Chances Given Low Chances Given Low
• Total Costs = $10M Quality Movie (0.7) Quality Movie (0.3) Quality Movie (0.3) Quality Movie (0.7)

Potential Revenue = Potential Revenue = Potential Revenue = Potential Revenue =

$20M $5M $10M $2M

Total Expected Revenue = ($20*0.7 + $5*0.3)*0.6 + ($10*0.3 + $2*0.7)*0.4 = $11.06M

Since Profits > 0, produce &
Total Costs = $10M (Given) release the movie
Total Profits = $11.06M - $10M = $1.06M

• This is essentially Bayesian probability. Some firms tend to ask cases based on Decision Trees hence this case is meant to give you an idea of that
Summary / Takeaways • This is a very simple decision tree. It might be possible that longer decision trees may be given in the interview
• Such interviews progress just as any other case with the interviewer giving intermittent data during the discussion.
177
APPENDICES

178
APPENDIX A
GUESSTIMATE STATISTICS
Following are some important statistics to keep in mind for solving guesstimates for market sizing and otherwise.
Population Distribution in India (in Millions) Population and Area of Major Cities
Income (in Mn) %Total Cities Population (in Mn) Area (in sq. km)
Total Population 1400
Below poverty line 280 20% Mumbai 13 600
Gender (in Millions) %Total Low income 560 40%
Males 742 53% New Delhi 11 1500
Middle income 420 30%
Females 658 47% High income 140 10% Kolkata 5 200
Sex Ratio (M/F) 1.12 Bangalore 5 750
Area wise Distribution of India (in '000 sq. kms)
Chennai 4 400
Geography (in Millions) %Total Total Area 3,300 Hyderabad 4 600
Urban 560 40%
Rural 840 60% Terrain (in '000 sq. km) %Total
Land 2,310 70% Tiers Population (in Mn) % Urban
Age Groups (in Millions) %Total Water 330 10% Tier 1 (Top 10) 60 15%
0-14 420 30% Forests 660 20% Tier 2 (Next 50) 60 15%
15-24 280 20% Tier 3 (Next 100) 45 12%
Population Distribution by Faith
25-34 210 15%
35-44 210 15% Faith (in Mn) %Total
45-54 140 10% Hindu 1,040 80%
55+ 140 10% Muslim 195 15%
Note: 50% of the population is under 25, Christian 26 2%
65% under 35 Sikh 19.5 1.5%
Others/Unspecified 19.5 1.5%
179
*Note: These are only rough figures (and not actual) that are meant to help you in solving the guesstimate with simple calculations
Key Data Points for Guesstimates
1. Population 3. Rural v/s Urban 6. Sector- wise (%age of GDP)
A. Rural: 65% A. Agriculture: 20%
A. India: 1.4 Bn (140 Crores)
B. Urban: 35% B. Industry: 25%
B. Lucknow: 3.5 Mn
C. Top 10 cities – 7% C. Service: 55%
C. Uttar Pradesh: 230 Mn
D. Delhi : Mn
E. Mumbai : 20 Mn 7. Occupation
4. Age A. Agriculture – 45%
F. Kolkata : 14 Mn
A. 0-6 years: 15% B. Industry – 25%
G. Bangalore: 12 Mn
B. 7-14 years: 15% C. Services – 30%
H. Growth Rate: 1.2%
C. 15-25 years: 20%
I. Birth Rate: 20/1000
J. Death Rate: 7.5/1000
D. 25-35 years: 15% 8. Others
E. 35-65 years: 30% A. Literacy Rates
K. Life Expectancy: 70 years
F. 65+ years:5% A. India – 78%
L. Sex Ratio: 950/1000
G. <35 years: 35% B. Youth (15-24 years) – 92%
H. <25 years: 50% C. Male (85%), Female (72%)
2. Area (In Sq. Km.)
A. India: 3.2 Mn B. Religion
B. Lucknow: 600 5. Income A. Hindu – 80%
C. Uttar Pradesh: 0.25 Mn A. BPL: 25% B. Muslim – 15%
D. Delhi: 1500 B. Low: 30% C. Others – 5%
E. Mumbai: 600 C. Lower Middle: 30% C. Car Penetration – 3internet 0/1000
F. Kolkata: 205 D. Upper Middle: 10% D. Internet Penetration – 50%
G. Bangalore: 700 E. High: 5% E. Mobile Phone Penetration – 80%
F. Smart Phone Penetration – 54%
GUESSTIMATES
Pension Investment BFSI | Easy

Your client is the Government. They are promoting investments in pension schemes, especially for youngsters
facilitated through UPI. The current value is roughly 10% of GDP and they have approached us to estimate how
much the value can reach at the end of the next five years
Clarified the objective and the clear definition of pension schemes that are being discussed. Can I take a minute?
Sure, no specific external factors are to be considered, estimate based on the current market situation and other
relevant factors you think are important. Let me know how you'd like to proceed.
I would like to look at the potential areas of growth, by analyzing the four major buckets of the customer journey as
they evaluate and invest in pension schemes – Awareness, Accessibility, Experience, and Feedback.
Sure, also keep in mind we need to quickly move on to the numbers as well.
Alright. Quickly discussed various subjective ideas in each of these buckets, which could potentially increase the
investments- largely revolved around increasing internet penetration, UPI access, and financial literacy through
online platforms. (Gave personal references based on my experience to make it relatable and practical)
These are some interesting ideas, can we estimate the numbers?
Solved this as a guesstimate. Starting from the population of the country is divided into urban & rural and further
into income groups to estimate the number of potential members who would invest. Considered reasonable
percentages for factors like internet penetration, financial literacy, and amount of investment as a proportion of
their income. This gave the total value of current investment which is 10% of GDP. To evaluate the future value,
relating to the qualitative points discussed, estimated reasonable percentages for an increase in the considered
factors ( like 1.2x/ 1.5x) to arrive at the new value of a total investment and also calculated the increase in GDP over
5 years. The final value was approximately 20%.
So we are estimating that the value can be expected to double, can you summarize the case by mentioning the
main factors that were key to this increase?
Sure, the key factors would be
• increase in digital platforms for investment,
• access to the internet
• improving financial literacy
• Increase in disposable income and
• tax benefits
can be a factor that can be key in increasing investments in long-term instruments like pension schemes
Thank you, we can end it here.

Table of Contents Consulting Club Casebook 23-24, IIM Indore 141

Fiction Books Media & Entertainment | Easy

Estimate the market size of fiction books in India. Makes sense. How would you quantify the mobile approach now?
Sure, will take a few moments to come up with a strategy. I used the top-down approach first and tried to divide the I started with the population data for the cities (told my assumption that I’ll be looking at metro cities for the
sales segments into the following – example). Broke it down into the working population based on age demographic data and further broke it down
1. Stores like Crosswords (For the time-rich and money rich) into salary classes. The interviewer stopped me in between.
This looks fine, we can close this part here.
2. E-Commerce platforms like Amazon and Flipkart (For money rich)
3. Local stores – Stationery shops, vendors
4. Black Market (First and second copy books sold on street / local markets)
Went on to explain the approach in the same order as above.
Interesting, can you think of any other approach for this?
Sure, so what we can do is that we identify the majority user persona that reads a fiction book and then try to
multiply that by a plausible frequency to come up with a number.
Great so this is a bottom-up approach. Let us go back to the first approach. You mentioned the black market, what
do you mean by that? (The interviewer was sitting in his car while taking the interview)
Sir, for instance at the next red light, there will be someone knocking on your window and bringing to you the top
15-20 books that are grossing in the market. Those are mostly second copies and can be purchased at a price much
lower than that of an original copy. Similar vendors can be found near major railway stations, bus stations, and local
markets in cities.
This sounds interesting, why don’t you neglect the other three and proceed with estimating the market size for the
black-market segment?
(Started solving it out loud) So, I’ll divide this into mobile and stationary points of sales. Mobile is the one that’d
come to your car for sales and stationery would be the vendors that have a cart near the bus station, railway
stations, and local markets.
How would you estimate the mobile part?
I would assume that the majority 80% here would include people who travel on a work day for their jobs and
among those, the majority would be those who have a car. So, I’ll break it down based on the working population
age (22-40) and a salary parameter assuming that those with 15L+ annual salary are the ones that will be traveling
with their vehicles. (I was focusing only on the majority segment here, multiple assumptions could have been used
here but given the pace with which the interviewer was driving it, I was just thinking on the spot)
But I have also seen old-aged people buying books on airports, which is not getting covered here.
So that’d be covered in the first bucket of the segmentation where I talked about stores like Crosswords for Time
Rich and Money Rich People. Here we would include only hard copy sales and ignore any digital sales on platforms
like Kindle.

Table of Contents Consulting Club Casebook 23-24, IIM Indore 142

Fitness Bands Retail | Easy

Estimate the Market size of fitness bands in India.

Sure. So, I was thinking that first I would share my approach to this problem with you and if that would be fine with
you, I will go ahead with numbers. Is that okay?
Sure
(Took a minute & came up with the structure). I initially segmented the population into rural-urban, then income
divide, and then age divide. I will use a penetration percentage per age group under all income levels. And then
sum up to come up the demand and then multiply it by the average price of fitness bands to finally come up with
the market size. Is that fine or do you want me to consider something else as well?
No, this is fine.
Okay. I will start with the numbers now. Do you want me to do the numbers with you only or should I do it first and
then explain that to you?
You can do the numbers by yourself and then we can discuss.

Sure. (took may be around 3-4 min to calculate) Explained the numbers, percent divide, penetration level, etc.

Don’t you think that the penetration number for rural India you assumed is on the higher side?
I don’t think so. The major targets for fitness bands are sports persons, health-conscious people, etc. I am a
national-level player. And based on whatever interactions I had with other teams (from different states) what I
observed is that most of the chunk belongs to rural areas only. So I believe that this is not on the higher side.
(Interviewer smiled very positively and convincingly after this)
Okay. What do you think that the number you came up with is on a low side, high side, or is fine?
I think the number is on the lower side which may be because of the penetration level that I took. I was on the
conservative side which might have led it to this.
Okay, so can you think of some other approach to cross-check your answer?
Sure. (took a couple of minutes). I can’t think of any concrete approach but currently, I can think that we can try to
compute the number of gym goers and sports persons to get the guesstimate.
Okay. That is fine with the case.

Table of Contents Consulting Club Casebook 23-24, IIM Indore 143

Plant-Based Meat FMCG | Easy

Estimate the market size (by number of potential customers) for the plant-based meat in India
That sounds interesting to work out. How about I walk you through an approach and then figure out the numbers
to drive the estimate?
Sure, go ahead!
So, I would start by dividing the population of India into vegetarians and non-vegetarians, Assuming the population
of the country is 140 crores, I would assume around 100 crore Indians consume non-vegetarian food items. This is
not taking into account the frequency of the consumption by an individual. So, for estimating for a potential market
purpose, I would like to go forward with a 100-40 split in favor of nonvegetarians. Does that sound good?
Your rationale sounds okay for the split, but do you think using the current eating habit will determine exhaustively
the inclination to consume plant-based meat?
I agree that there will be a percentage of non-vegetarians as well as vegetarians who would consume plant-based
meat, but determining those percentages will involve numerous assumptions, some of which may be unqualified.
So, you have successfully identified how your approach falls a little short of arriving at the final number easily. Let
me help you with an analogy that might help you tweak your approach. I want to buy a Rolex. I like Rolex. So, there
is a willingness to purchase, but it is not backed by my ability to pay for it. Hope this should help.
Thank you so much for the analogy. I’d like to use the income level of the Indian population as a segmenting
variable to calculate the market size. I guess that would work sufficiently.
Perfect, go ahead.
At current price levels and industry nascency, plant-based meat is expensive and I would like to consider a price of
Rs. 500/meal for 4 people. 4 people to account for the average household size estimate for the country. Assuming
that a plant-based meal is consumed 10 times a month, that would indicate an expenditure of Rs. 5000/month on
plant-based meat.
That sounds good to me! What next?
Yeah, so a household would consume a total of 90 meals a month. So, the total food expenditure would comprise
plant-based meat expenditure for 10 meals and an average expenditure of Rs. 250/meal for the remaining 80
meals. So that would bring the total monthly food expenditure to Rs. 25000.
That sounds great. You can skip the quantitative part, I’m in a bit of hurry. Just walk me through what you would do
next to arrive at the final estimate.
Sure, I would assume a certain percentage of monthly expenditure is spent on food by an average Indian household
in a weighted manner for different income segments. Those households having monthly incomes which can support
these figures will be the potential market. Having arrived at the number of households this way, I can also estimate
market size by revenue.
That sounds pretty good to me! Thank you.

Table of Contents Consulting Club Casebook 23-24, IIM Indore 144

Flights in Air Airline | Easy

Estimate the number of Flights In the air in India in a day.

Is it good to assume it to be a normal day because during special occasions Air carriers will increase the flight to
meet the demand?
Yeah, It is a good assumption.
When we say flights, there are many use cases for it – Passenger Carriers, Cargo carriers, Private jets, Government
Officials, military jets, and Training academies.
Good. Let’s stick only to the passenger carrier.
I will start by dividing cities into metro and non-metro. There are four metro cities with an average runway of 2
each and there are 29 states with an average of 4 nonmetro cities with airports and 1 runway in each state which
gives us 120 runways approximately. The runway is the bottleneck and so we are following this method.
Aren't 4 airports in each state on average too high? Let’s take Maharashtra & check how many airports are available
Pune, Nagpur, Mumbai (covered in metro). I have read that sometimes there are around 150 -160 airports in India
and that’s why I have taken an average of 4 airports.
Okay. Now, let's assume 3 airports on average in each state.
Sure. Then, we will round up to 100 airports in non-metro cities. Now, we will have divided the day into peak and
non-peak hours. And we assume for take-off/ landing it takes 15 minutes on average and during peak hours, the
runway will be at 100% occupancy level in metro and 75% occupancy in non-metro. During non-peak hours, 75%
occupancy in metro and 50% occupancy in non-metro cities.
Sounds good. Please proceed.

Calculated the number of flights landing or taking off.

Is it the final number?

No, we have to divide this by 2 because the same flight will take off and land and so we would have calculated
twice. We can split x% to be domestic flight and 1-x% as international flight so that only the domestic flight count
will be divided by 2 and not the international flight.
Good. You are missing something really important in this guesstimate.

I was thinking and gave a lot of answers, but could not find it.

It was an international flight that took off and landed in other countries but flew above India. (Example)

Yeah.

We can close.

Table of Contents Consulting Club Casebook 23-24, IIM Indore 145

Annual Paper Demand FMCG | Easy

Estimate the number of trees shed for meeting the annual demand for paper.

Is there a target geography we have in mind?

Mumbai
Sir, I see that the problem statement has 2 parts – an estimation of the annual demand for paper in Mumbai and
the number of trees shed to meet the same. The latter part is a simple conversion, we can find the number of trees
shed to make one sheet and multiply it by the number of sheets required.
Are you sure that the number of sheets is the best approach?
A better approach may be tonnage. So, the equation will be the number of trees shed for making 1 ton of paper *
annual demand for paper (in tons)
Yes, continue

I’d like to bucket the annual paper demand in terms of functionality. Paper is used in various forms- in schools and
universities, as paperwork in corporates, as paper bags in Mumbai since there is a state ban on the use of plastics.

Sounds okay
Sir, I also have a slightly unconventional bucket which I believe would be significant. Cigarettes use paper as rolling
paper and for packaging purposes. A few years back, ITC had recognized the paper as the highest cost incurred for
making cigarettes and decided to invest in manufacturing their own paper. This same paper is what is scaled multi-
fold and is now known to us as the brand Classmate. Shall I use this also as one of my buckets?
Hadn’t thought of this, you can use it. How will you go about calculating each of these buckets?
Each bucket is a mini guesstimate. For the first bucket-> Schools and colleges- estimate the number of educational
institutions in Mumbai * an average number of students per institute * an average number of books used by each
student in a year * conversion factor (to tons) Does this approach look okay?
Yes, continue
For the second bucket-> Corporates - I’ll consider a project-based approach and mark up the miscellaneous paper
usage as 20% since every stage of the project requires very extensive paperwork, with certain wastages. So,
(number of corporates in Mumbai * average number of projects taken up in a year * average requirement of paper
per project * conversion factor (to tons)) * Markup. Do you want me to delve into more details?
No this looks okay, next?
For the third bucket-> Paper bags - The ban on plastic in Mumbai is very stringent. People usually need bags when
they for shopping. Indians are usually price sensitive and there is an ingrained mentality to reuse bags. So, for this
guesstimate, I’ll only consider the paper bags given at retail stores and assume that the bags used for grocery
shopping are just reused retail store bags. So the equation would be, number of retail stores in Mumbai * number
of bags/order * average orders/day * 300 days * paper required for 1 bag * conversion factor (to tons)

Table of Contents Consulting Club Casebook 23-24, IIM Indore 146

Footfall of Food Store Hospitality | Easy

Estimate the footfall of food store at IIM Indore Do you think these many people visit the shop?

Are we talking about packaged food? No, this is because there is a hospital, barber shop, Mars, and other shops there, So the crowd disperses there. So,
suppose 120 people go to other shops. So, around 200 people visit the shop, and since two shops keep packaged
Yes. food items. Footfall is around 100 per shop.
Also, are we talking about weekday or weekend? Okay, that looks good.

It is a Wednesday that we are talking about.

I will like to divide the day into peak and non-peak hours. Peak hour is from 5 PM to 9 PM. I will ignore the non-
peak hours since all of us are busy with classes and other things, and we do not move out. Also, I will like to
calculate the number of students who move out for a purpose during peak hours.
Go on.
We have different courses running here, the Majority of them are PGP & IPM. So, I will divide the student hostels
into PGP1, PGP2, and IPM. PGP 1 and 2 have ~500 students each and IPM has ~300 students. PGP1 - There are 2
boys’ hostels. Each hostel has 4 floors of around 35 rooms each. Since the hostel is on the lower side of the hill and
the shop is far away. Also, the work pressure does not allow us to move out. We do not go out, rather we ask
someone else to bring the stuff for us. So, when I look in the evening, I see only 7-8 rooms locked. So, I assume that
20% of people from each floor go out on a purpose.
PGP1 (Boys)
People from each floor = 7, No. of floors = 4, No. of hostels = 2, Total no. of PGP1 guys moving out = 56
PGP1 (Girls)
There are 33% of girls on campus. Benchmarking it with the boys, since the work pressure is same. Total no. of girls
moving out = 56/2 = 28
Total PGP1 = 84
IPM
IPM hostels are also nearby PGP1. And they face similar constraints. So, I will like to benchmark it with the PGP1
since I do not have any idea about their floor plans.
Yes, Go on.
Total IPM Students moving out = (2/3) * 84 = 54.
PGP2
The PGP2 students live nearby the stores. Mostly move out for some or other purposes. I will assume that 30% of
them move out.
Total PGP2 students moving out = 0.3 * 600 = 180
Total Number of students = 180 + 54 + 84 = 320

Table of Contents Consulting Club Casebook 23-24, IIM Indore 147

Two-wheelers in India Automobile | Easy

Calculate the number of two-wheelers sold in India in a year.

Repeated the guesstimate statement to get it clarified. Asked if it was okay.

Sounds correct.

Started with clarifying questions like new / old two-wheelers, personal or business (Zomato and other delivery), etc.

New Two Wheelers for personal use sold in a year.

I want to take the demand-side approach, starting with demographics (Age, Region, and finally income level) and
then dividing the estimated number by the average life of a two-wheeler. Does this approach sound appropriate?
Yes
So, considering the population between the age of 18-60 who would frequently use two-wheelers and hence
demand two-wheelers. Using the recent Covid vaccination data available we can find the population in this age
group. Gave some numbers. Does this seem fair?
Yes, move ahead.
Okay so now dividing the population into urban and rural (40/60 split for convenience of calculations) and then into
income levels (Low, Middle, High) (40/40/20 urban) (50/40/10 rural region). Does this number seem fair?
Yes.
Now since the poor cannot afford owning two-wheelers and rich would not want to own two wheelers, we can
discount them. Focusing on the middle-income segment, we can assign proportions of how many people would
want to own a two-wheeler vehicle.
Okay now?
Now dividing by the average life of a two-wheeler vehicle which from my personal experience is easily around 8
years. Arriving at a final figure, however, I think this is a bit higher than what I had in mind.
Can you tell reasons why this number would be a bit higher? You can take a few minutes to think about out it.
I feel I included second-hand vehicles as well as some vehicles which might be used for private as well as business
purposes. For the hygiene check, I would like to multiply this number by a factor of 0.7 (excluding all other factors,
listed the reasons for this assumption).
Yes, this seems like a better figure. Can you just enumerate some reasons why two-wheeler use is growing in India?
The convenience of parking and gender neutrality, Ease of availability of finance, Increasing traffic in metro cities
etc.
Okay I think we can close the case now.

Table of Contents Consulting Club Casebook 23-24, IIM Indore 148

Masks used in Delhi FMCG | High

Please size the number of masks used in Delhi between the months of October 2019 and September 2020. Right, that makes sense. Please go ahead with the COVID period.
Clarified the problem statement. First, I can see that this would be divided into two periods. COVID hit India around Right, let me sip water. Great so the method I’ll take is similar to adding together the commercial and domestic
March-April. requirements. Let me get into the commercial side first. A large part of this period was a very rigid lockdown so
Right so let’s divide it into 6-month periods of non-COVID and COVID. industrial activity was anyways shut. Things opened towards the second half of this period, and the second wave
also hit us which was very intense. This makes me think. Would you like me to further divide this period into two,
Let me attempt the non-COVID period first. Let me start with the population of Delhi, I am assuming that is around
three months periods?
1 crore. Does that sound fair?
Go ahead with assuming common assumptions for both periods. You can take figures that will average out effects.
The correct population is around 2 crore but let’s go ahead with 1 crore for this calculation.
Alright then, so the number of hospitals that were there before will remain. There was some capacity addition, but
Alright, so you’re taking all of NCR, either way, I can double the number at the end if I need. Let me take a minute to
I’ll ignore that and after this take a sub-segment of new hospitals and Covid care centers added by the government.
structure my approach. (Takes 1 min) Right, so we can divide the mask utilization into domestic and commercial. Let
(Asks for her approval and gets it) Alright so the number of doctors won’t change suddenly but they might have had
me size the commercial side first. Commercial will be hospitals and industrial. Since Delhi is not famous for any
to change masks every 6 hours and doctors worked longer shifts, I’ll assume 12 hours each. That brings me to 2
major chemical manufacturing or hazardous work like tanneries or steel plants, I shall assume only hospitals to be
masks per doctor. The nurse’s capacity was expanded, and I will assume to 1.25X, further the number of masks used
the commercial requirement. Does that sound reasonable?
by them will be the same as those by doctors. Furthermore, the government expanded beds with temporary
Yes, please go ahead. facilities and covid care centers. I will assume 100 such facilities were set up with 100 beds per facility. Does that
Great, so Delhi is a large city, around 1,500 square km. I will divide that into localities of X square km to give me Y sound reasonable? (Nods in approval) In that case, I will have another X number of doctors and Y number of nurses.
localities. I will assume every locality is serviced by one hospital with an average bed size of Z. Given this I would That brings me to N number of masks used for medical purposes. Now let me get to the domestic side. With a
like to assume there are Several doctors per hospital and B number of nurses and ward staff, etc. For simplicity, I population of 1 crore. 10% are from the richer segment who shall buy branded cloth masks. One of these masks
will assume that every one of them uses 1 mask per day on average. Now, that I have my per-day requirement, I shall easily last for the entire 6 months period. These shall be 10L masks. Another 10-20% of the population will not
need it to be for 180 (6x30) days. (Calculates) Here, is my final number for the commercial side. wear masks at all. Let me assume 20% for a city-like Delhi, I would have taken lesser for a city like Chennai or Kochi
with higher education levels. Does that sound fair? That leaves me with 70% of the population. I will assume that
This seems okay, now how would you like to proceed?
every mask will last 1 day (as it is a single-use disposable and then I have it for 180 days).
On the domestic side, we’ll need to include a small population segment who is using it for respiratory disorders and
Does that sound like a fair number to you?
another small segment like bikers who use it for basic protection or aesthetic reasons. (Takes percentage ratio
assumptions of the population for each segment) Further, I would assume that the bikers use cloth masks, which No, this seems like a very large number, and I don’t think this much demand could be satisfied. This number I have
should easily last an entire 6 months, and the health patients would use N96 masks. The health patients I assume calculated assumes perfect demand.
will need 1 mask per week. Alright so domestic and commercial together gets me to this number. What changes would you like to do?
Alright, this is a very small number, don’t you think we can ignore this? I think the medical side is fine, what I would like to change is the number of days a mask is used and the number of
days a mask is needed. Even though every mask is disposable, I believe people stretch each mask for 2-3 days.
That’s true, but I wasn’t sure about the size prior, so I went ahead and calculated it.
Secondly, masks will not be needed for all 180 days. There was a strict lockdown for most of the days. Let me take
That’s alright, never hurts to be extensive. Is there any segment you are missing out on? (hints me towards saying that down to 25 assuming 1 will be needed per week.
its Delhi specific, finally I am unable to reach the idea) Its people wearing masks to avoid pollution. Would you like
That sounds like better numbers. What is the final reasonable number you would suggest for Delhi?
to consider that too?
Right, Delhi is famous for its smog levels. I don’t see the point of considering it. The number of people who will be Around 25-30 crore masks.
using masks will be small and restricted to the richer segments, most other people will either not use a mask at all,
or fashion a handkerchief as a mask. The number will be insignificant compared to the commercial demand. That sounds reasonable. Thank you!

Table of Contents Consulting Club Casebook 23-24, IIM Indore 149

Fantasy Games Media & Entertainment | High

Estimate the market size of online fantasy gaming players of football in India for the first 2 years of entering in
Indian Market.
The factors I would consider are - Smartphone penetration in India, Income available for Investing/Spending in the
application, Time available, Legality/Barriers from the Government, People’s general interest in sports and
particularly in football, Awareness of online fantasy sports, Competitors…
Ok, could you now estimate the market size for the first 2 years?
Sure, I would divide India’s population broadly in two segments – Tier 1 (30%) & Tier 2/3 cities (70%). I will first
estimate the market size for Tier-1 cities and then extrapolate the number obtained to Tier 2/3 cities. Would that
be alright?
Sure, Go ahead
Now for Tier 1 cities, I would assume 80% of the population would be equipped with smartphones. I would further
segment the population age-wise. I will divide the Tier-1 population (who own smartphones) into four age groups –
0-18, 18-30, 30-50, and finally 50-70 age. Now, for each age group – I will list factors to arrive at a realistic
percentage of actual people interested in online football fantasy sports. I will allocate a percentage to each factor
for each age group (I made a matrix that has the factors listed in rows and the age groups in columns – and
populated % in the matrix). The factors I will consider for each age group would be - people interested in sports,
people interested in football, people aware of fantasy sports, income factors, and time availability. Now, I will
neglect the age group of 0-18 years, as my understanding of Fantasy Sports Apps is that they require users to make
payments to participate in tournaments. Since they do not earn and may end up spending too much, they are not
allowed to participate. Does this approach seem fine? If it does, I shall start assigning numbers to each segment.
Sure, this seems quite exhaustive, can you give me a final number now?
Yes, I will now assign percentages for each segment and arrive at a final number of Tier 1. (Performed all
calculations) – arrived at 3 million for Tier 1 (2 million for 18-30 and 1 million for 30-50). Now to extrapolate the
number to Tier 2/3 cities, the percentage age for various factors for each age group will differ. The income
availability would reduce & time availability increase. I would take 70% of the Tier-1 estimate, therefore 2.1 million
for Tier-2. The total market size for online fantasy football sports would be 5.1 million.
Sounds good, can you tell me how would you convince real football fans, who love playing football, to start using
the online fantasy football app?
We can convince real football fans in the following ways:
• The online fantasy football app follows real-world tournaments. Hence, they would get a chance to form their
teams to display their skills in the game. The app follows all major leagues and tournaments
• A medium to connect and compete with friends and create an online community
Great, I think we are mostly done with the interview.

Table of Contents Consulting Club Casebook 23-24, IIM Indore 150

Cash withdrawal from ATM BFSI | Moderate

Estimate the total amount of cash withdrawn from ATMs in Bhubaneswar in a year.

Repeated the guesstimate statement to get it clarified. Asked if it was okay.

Sounds correct.
Asked him if it is okay to take the supply-side approach to the problem. He asked me to elaborate on what I meant.
I explained that considering the supply side meant estimating the average amount of cash withdrawn from a single
ATM by considering the cycle time of cash withdrawal per person as the bottleneck process time.
Okay do that!
Considered the average cycle time is 5 mins per person, thus 12 withdrawals per hour in an ATM during full
occupancy. Asked him if the ATM is open 24*7.
Yes!

Ok. Let’s take 8 hours as peak hours and 16 hours as non-peak hours.

How can you classify 8 hours as peak hours? And what do you mean by peak and non-peak?
Told him that we can consider morning 7-10 am and then evening 5-10 pm as peak hours which operate at full
occupancy. The other hours are considered to be non-peak for which we can take an average of 30% occupancy.
Okay go on!

Asked him if we could consider the average cash withdrawal per person as ₹ 1000.

Can you arrive at a number for average withdrawal by thinking about a few factors?
Okay. Considered the income levels of people who use the ATM. Excluded super-rich and low-income groups and
arrived at a weighted average of around 1300 for medium and rich-income groups. Thus, roughly ₹ 15,000 per hour
during full occupancy and ₹ 5,000 during other hours. Arrived at the number for a day. Considered a downtime of
10% in a year to arrive at the amount for a year.
Would the amount of cash withdrawn be the same across all days in a year, for example, during festive periods?
Told him that I have taken it similarly for the sake of simplicity. But of course, it would change during festivals in two
ways: peak hour time would increase, and the average withdrawal amount would rise too.
Okay. So now how do you arrive at the total cash withdrawn in the whole city?
Started with the total population of the town, and considered 1 ATM per 200 people which gave us the number of
ATMs in the town (did a calculation mistake), and multiplied the amount per ATM
Okay. Thank you.

Table of Contents Consulting Club Casebook 23-24, IIM Indore 151

iPhone Users Retail | High

Estimate the number of iPhone users in India right now.

Thank you. I would like to reiterate the problem statement to ensure that we are on the same page. You want me to
estimate the total number of people using iPhones in India at the moment. Is that correct?
Yes, that is correct.
Okay, let me start by segmenting the population of India based on whether they belong to the rural or urban
sections of the country. To do this I’m taking the population of India to be around 140 crores out of which I’m
assuming that 60% of the total population belongs to the rural sector and the remaining 40% belongs to the urban
sector. Can I go ahead with these numbers?
Your assumptions sound good but we are more interested in understanding your thought process and the way you
are structuring your solution. So, you can always round off the numbers involved which will make your calculations
simpler and your life easier. For instance, you can even consider the population of India to be 100 crores instead of
140 crores as suggested earlier.
Alright. Now I am assuming the population of India to be 100 crores and using the assumptions stated earlier, 60
crores will belong to the rural sector and 40 crores will belong to the urban sector. Further, I would like to segment
the population based on their income criteria into 3 categories: Rich, Middle class, and Poor. Is that okay?
Yes, that sounds okay.
Okay, so now I’m assuming that out of the 60 crores comprising the rural population, 10% belong to the rich
segment, 30% belong to the middle class and the remaining 60% belong to the poor segment of the society.
Similarly, we can break down the 40 crores comprising the urban population into 20% rich, 40% middle class, and
40% poor people. Can I proceed with these assumptions?
Yes, you can carry on with these assumptions.
For the rural segments, we get 6 crores for the rich class, 18 crores for the middle class and the remaining 36 crores
for the poorer sections. If we calculate for the urban segments, we arrive at 8 crores for the rich, 16 crores for the
middle class, and 16 crores for the poor. Among the rural segments, I’m assuming that only the richer sections of
the population will be able to afford an iPhone, and among the urban segments, I’m assuming that the rich and
roughly half of the middle class population will be able to afford an iPhone. Can I go ahead with these assumptions?
Yes, these assumptions are fair enough. Please proceed with your analysis.
This implies that 6 crores (rich) from the rural segment and 8+8 (rich + middle class) that is 16 crores from the
urban segment would be able to buy an iPhone. This sums up to 22 crore potential buyers. But not every member
of this stratum needs to prefer an iPhone. Instead, they may opt for other smartphone brands. Let us assume that
only 40% of the 22 crore population use an iPhone and the rest go for the other brands in the market. This means
that there are about 8.8 crore iPhone users currently in India. Is this estimate good enough?
Yes, this will suffice. Thanks.

Table of Contents Consulting Club Casebook 23-24, IIM Indore 152

Planes in the Sky Airline | Easy

Estimate the number of planes in the sky at any moment

To reiterate the case, I need to estimate the number of planes in the sky at the moment

Yes
Understood, let me start by setting some assumptions. We would have commercial passenger planes, commercial
cargo planes and non-commercial planes, for the sake of this discussion, we can ignore the non- commercial planes.
Moreover, I shall further divide the approach based on domestic and international planes. Does that sound like a
fair approach?
Sounds good so far.
Right. So, in this problem I can use a concept I learnt in operations management, called little’s law. Essentially, I can
consider take-offs and landing frequency as the cycle time, the duration the plane spends in the air as the MLT and
therefore the work in progress will be equal to the number of flights in the air.
Yes, that is the right approach and exactly what I wanted to hear, proceed with your assumptions
Ok, so I shall start with the domestic commercial passenger flights. We can assume that we have 36 domestic
airports in India and for the sake of simplicity, I can assume that the average take-off time is 15 mins. This is only a
simplifying assumption as in reality; the frequency would vary based on the airport and the time of the day. Is this
assumption, ok?
Yes, that works.
Alright, let us further assume that the avg. duration of a domestic flight is 3 hrs. Therefore, in my calculations, the
number of flights in the air = number of airports * (avg. time in the air/Take-off frequency)/2
Why are you dividing it by 2, that is not clear
Oh my bad, I was adjusting for the take-off and landing, however that won’t be necessary given that it is already
factored in.
Yes, exactly.
So essentially, my estimate would be around 432 planes for domestic commercial planes. This seems on the higher
end, should I adjust my assumptions?
That won’t be necessary. Do you want to proceed and do the same calculations for the cargo and international
flights?
Took a MECE approach. Tried to dive into each and every bucket. Gave all possible areas of improvement.

Ok. That’s all

Table of Contents Consulting Club Casebook 23-24, IIM Indore 153

Uber Driver Transportation | Easy

Estimate the monthly earnings of an Uber driver in Mumbai

For the revenues, should I only consider the amount received through Uber?

What other revenues could he have?

Sometimes when I have a flight late at night or early morning, I contact a driver that I personally know instead of
Uber as I am unsure of availability and sometimes due to safety concerns. So, any other rides that he receives from
known customers.
Let's only consider the revenue from Uber.
Should I be assuming that the driver owns the vehicle? (Also mentioned the possibility of renting a vehicle and
whether he has taken a loan for the vehicle)
Yes, he owns the vehicle and there is no loan on him.
I would like to divide this into three parts. First, I’ll look at the revenue generated in a day and the costs involved.
Then I will calculate the monthly earnings based on the number of working days.
Looks good.
I am assuming that the driver works for 12 hrs a day, of which I’ll remove 2 hours for lunch and breaks. A ride on
average would last for 30 minutes, so the rider would have 20 rides in a day.
Ok.
Calculated total revenue in a day by assuming the average price for a 30 min ride based on my experience and
mentioned the same. (I quickly calculated using two different prices for a ride to see which one was giving a realistic
figure for salary and mentioned that while speaking)
Don’t you think that there would be higher demand during some time of the day?
Yes, I will consider that two hours of the day - one hour during the morning office hours and one hour in the
evening when everyone’s getting back after work, would be busy. I would consider surge pricing of 20% for those
two hours.
Ok
Now I would like to look at the costs. Here I listed down and mentioned the costs - commission to uber, fuel based
on average trip length and mileage (wasn’t sure of the mileage so confirmed with the interviewer), the car is owned
so no rent, monthly maintenance, no fines paid assuming that he is a good driver. Calculated the monthly earnings
from the daily earnings by assuming that the person works 25 days a month. Came up with a figure of 70,000 INR.
This seems a bit high. But we can end here.

Table of Contents Consulting Club Casebook 23-24, IIM Indore 154

EV charging stations Automobile | Easy

Estimate the number of EV charging stations in Maharashtra in 2030 Sounds reasonable. Go ahead.

I’ll reiterate the problem statement. (told the problem statement) (Explained my thought process about the filters of accessibility, technological know-how & government
interference, and push to further narrow down the number. Finally reached the number of 7500) Now, as we are
Yeah. That’s correct. calculating for the year 2030 we need to consider the annual growth rate of EVs as well.
I’ll follow a 5-step approach to calculate the answer. Firstly, I’ll discuss the scope of the problem. In the 2nd step, I’ll So, are you saying there are 7500 EVs sold to date or 7500 EVs sold in 1 year?
define the approach that I’m going to follow. Next, I’ll apply some filters or variables which are important to our
(Got confused for a bit. But took a few seconds to think about it) Currently, EV penetration is negligible even in a
problem. In the 4th step, I’ll state my assumptions and assign values to these variables & finally, I’ll do numerical
developed state like Maharashtra. There are multiple reasons like lack of infrastructure and accessibility. But our
calculations to get the final answer.
government is encouraging EV usage through schemes like FAME, and NEMPP2020 which are currently way behind
Okay. That sounds good. Go ahead. their targets. So we can assume the penetration of a total of 7500 Evs in Maharashtra by the year 2025 & then
Okay. So, for this guesstimate, I’m going to consider both individual and commercial EVs which are sold in move ahead with a CAGR of 10 or 20% to calculate no. of EVs by 2030.
Maharashtra. Also, I’ll consider only 4-wheeler cars. And for the approach, I’m going to follow a hybrid approach Okay. I’ll give you some information. Indian Govt. is aiming to have EV sales accounting for 30% of total passenger
wherein I’ll combine both demand and supply-side approaches. vehicles.
Okay. The target is to have 30% of EVs among total annual vehicle sales or 30% of EVs amongst total vehicles on
Exactly how are you planning to go about it?
road?
I’ll calculate the number of EVs in Maharashtra by 2030 and then state the assumption that there will be 1 EV
It’s the 1st one. EV sales contributing to 30% of annual private vehicles sales.
charging station within the span of say 10 or 100 sq. Km (depending on the no. of EVs that need to be charged)
there will be 1 EV charging station, I’ll calculate no. of EV charging stations. I’ll make use of Maharashtra’s Okay. So, shall I assume CAGR of 30% then?
geographical area for this. I agree. So, you are saying that would not affect the second component of the equation. What about the third
That’s a good approach. Go ahead. component?
Okay. To calculate the number of personal EVs I think important factors are the number of households, rural and I have assumed monthly value of customer and multiplying it by 24 months to get customer value in two years.
urban households, affordability, accessibility of both EVs and after-sales services, know-how about the usage of EV
& governmental push. Do you want me to consider any other factor?
No. You have covered the factors.
Okay. As per my understanding, Maharashtra's population is above 100 million. For the sake of simple calculations,
I’ll use 100 million. Also, I’m going to apply a 70/30 rural-urban breakup in our case similar to a nationwide breakup.
Also, I’m assuming avg. household size to be 4 for both rural and urban segments. Is it fine?
Yeah. Sounds good.
Next, I’m dividing the population into 3 segments based on their income levels- Poor and BPL, middle-income
group, and upper-middle & rich group. (Gave percentage breakup for these groups for rural and urban segments.
Calculated the absolute figures as well). Now, let’s assume the average price of an EV to be Rs. 15 lakhs. Based on
this now I’ll calculate how many households can afford an EV. (Gave separate % breakup for income groups in both
rural and urban segments) I think we can consider only the upper-middle and rich segments from both urban and
rural segments. (Explained why to exclude other income groups). Shall I move ahead with these assumptions?

Table of Contents Consulting Club Casebook 23-24, IIM Indore 155

H Guess Estimates

Approaching Guesstimates
The process of solving a guesstimate during an interview entails employing a systematic approach to arrive at the solution, making logical
assumptions, conducting rough calculations, and effectively communicating the entire process.

Solving Guesstimates

These approaches are

Pyramid Approach Supply Demand Approach similar, choosing between
them is based on level of
comfort

Top-Down Bottom-Up Supply Side Demand Side

• Start with the • Start from the • Involves making • Demand side As a rule of thumb, the
broadest possible bottom – some low an estimation of estimations usually bottom up approach can
universe level statistics such the supply of a require a be used when consistent
as ‘Revenue per product or service population side lower level metrics can be
• Apply filters to customer’ analysis established.
narrow down the • Based on available
universe • Build up to the data & reasonable • Constraints to be
universe by assumptions put on factors Do not stick to one
approach particularly,
• Keep applying a applying relevant driving practice all types as they
set of filters to numbers • Constraints to be demand/growth are helpful in case
arrive at the put on factors of interviews as well
solution production

Tip: Segmentation is an important part of guesstimates. Keep these types of segmentation handy in case needed.
Demographics (Age, Sex, Income, Ethnicity etc.)
Psychographics (Attitudes, Behaviours, Values)
Geography (City/Country/Continent, Urban vs. Rural etc.)

© Management Development Institute, Gurgaon | 2023- 2024 116

H Guess Estimates

Let us solve a Basic Guesstimate!

Prompt: "How many tennis balls can fit inside a school bus?"

Step 1 Step 2 Step 3 Step 4 Step 5

Understand Question Make Assumptions Break Problem Down Use Constraints Apply formula

§ The first step is carefully § Guesstimates often § Divide the problem § To limit the § Use basic operations –
read & understand the require you to make into smaller, more guesstimate, use the Addition, subtraction,
guesstimate question educated guesses or manageable parts constraint present in multiplication, division
§ Ask clarifying questions assumptions § Intuitively you can the problem to get to the required
to get a better § Make reasonable think of smaller § It can be volume, number
understanding assumptions and state components Length, mass etc.
them

“Tennis Balls” “Both the School bus “Dimensions” “Volume constraint” “Division Formula”
& Tennis balls are School Bus: School Bus: Vol. School bus=
“School Bus”
standard Size” Volume No. of tennis balls *
Length = 10 m Volume of 1 tennis
= L*B*H
Width = 5 m ball
= 10*5*7
Height = 7 m
= 350m3 350 = N*0.0005
Tennis ball:
Tennis ball: N = 7,00,000
Diameter: 0.1 m Volume
=(4/3)*3.14*r3
=0.0005m3

© Management Development Institute, Gurgaon | 2023- 2024 117

H Guess Estimates

Example for Supply & Demand Side

Prompt: “A car manufacturer is planning to launch a new electric vehicle in a country with a population of around 50 million people. The EV is
targeted at the mid-range segment and offers competitive features at an affordable price. Estimate the potential demand and supply for the new
electric vehicle in the target market.”

Demand Side Supply Side

Total Demand Total Supply

Estimated = Add. Mkt + Add Mkt*GR Estimated =Mon. Prod.*Mon.
Demand = 100k + 100*5% Supply =1280*12
= 105k =15,360
Monthly Production
Addressable Market =(Prod. Hours * PLs)
Monthly No. of
Addressable Growth = Market Size * Share / Cycle time
Production Months
Market Rate = 1 Mn * 10% =(320*8)/2
= 100K =1280
Prod. Hours
Market Size Production Production
Cycle Time =Wkg Days *Shifts *
Market = Population * Penetration Hours Lines Hours/Shift
Market Size
Share = 50 Mn * 2% = 20*2*8
= 1 Mn =320

Working
Shifts Hours/Shift
Penetration Days
Population
Rate

*Assumptions (Give Reasons) *Assumptions (Give Reasons)

Penetration – 2% Penetration – 2%
Market Share – 10% (First Year) Market Share – 10% (First Year)
Growth Rate – 5% (First Year) Growth Rate – 5% (First Year)

© Management Development Institute, Gurgaon | 2023- 2024 118

H Guess Estimates

Fact Sheet

General Statistics Segmentation Geometry Formula

Population (No.) Rural-Urban Gender Marital Status Perimeter

Total 1.5 Bn Rural 70% Male 52% Married 50% Square 4 * Side

Delhi 16 Mn Urban 30% Female 48% Unmarried 50% Rectangle 2* Length + Breadth
Triangle Sum of Sides
Mumbai 18 Mn Age Groups Area Distribution Income Distribution
Circle 2 * 3.14 * Radius
Kolkata 14 Mn 0-14 30% Land 70% High 1%
Bangalore 8.5 Mn Water 10% Area
15-24 20%
Upper Middle 15%
Forest 20% Square Side2
Lucknow 3.3 Mn 25-34 15%
Lower Middle 30% Rectangle Length * Breadth
Chennai 5 Mn 35-44 10% Internet Penetration
Low 25% Triangle (1/2)* Base* Height
45-54 10% Urban 55%
Area (in Km2) Circle 3.14 * Radius2
55+ 10% Rural 45% BPL 29%
Total (‘000) 3,300 Volume
Delhi 1,500 Miscellaneous Statistics Cube Side3
(Overall India)
Mumbai 600 Sphere (4/3) * 3.14 * Radius3
Kolkata 200 Internet Penetration 40% Median Age of Population 25 Cylinder (1/2)* Base* Height

Bangalore 700 Mobile Penetration 48% Population Growth Rate 1.25%

Misc.
Lucknow 600 Smartphone Penetration 22% Life Expectancy 70 Pythagoras a2 + b 2 = c 2
Chennai 425 5G Adoption 40% Population Density 500 Rhombus (A) ½ * D1 * D2

Note: The provided data is only for India.

These data-points are approximate figures for the purpose of guestimates and are valid as of 2023.
© Management Development Institute, Gurgaon | 2023- 2024 119
H Guess Estimates

Guesstimate 1: Fuel Consumption

Prompt: Estimate the quantity of petrol used per day (in litre) for transportation in Delhi

Approach
Demand Side Population of Population of Delhi
Delhi 15 Mn

Methodology (Average Household Size – 5)

1. Calculate the Number of
Household = Population/Avg Size
households Total
= 15 Mn/5
Households = 3 Mn
2. Calculate Vehicles per
household
(Division Basis Income)
3. Calculate the
consumption by each Upper Class – 150K
Upper Class Middle Class Lower Class Poor Middle Class – 750K
vehicle. (5%) (25%) (50%) (20%) Lower Class – 1500K
4. Scale up the consumption Poor – 600K
to the entire population

2W - 1.5 2W - 1 2W - 0.5 2W - 0 (Average No. of Vehicles/household)

Assumptions 4W - 1.5 4W - 0.75 4W - 0 4W - 0
§ Diesel and CNG is majorly
used by public transport
and commercial vehicle
(busses, vans, and trucks) 2W - 225k 2W - 750K 2W - 750K 2W - 0
(Total No. of Vehicles)
4W - 225K 4W - 562.5K 4W - 0 4W - 0
§ Average distance travelled
by each vehicle (per day)
20 km
Distance by 2W = 20 * (750K + 750K + 225k) = 34,500K Total Fuel = 2,265K Litres
§ Mileage (in km/litre)
Litres used by 2W = 34,500K / 50 = 690K (Assuming 20% is Diesel)
2 Wheeler = 50
4 Wheeler = 10 Distance by 4W = 20 * (562.5K + 225k) = 15,750K
Litres used by 4W = 15,750K / 10 = 1,575K Solution: Total Petrol used = 1,812K Litres

© Management Development Institute, Gurgaon | 2023- 2024 120

H Guess Estimates

Guesstimate 2: Maruti Suzuki Cars

Prompt: Estimate the number of Maruti Suzuki Cars in India

Approach
Top Down Population of Delhi
Population of
India 1400 Mn

Methodology (Average Household Size – 5)

1. Calculate the Number of
households Household = Population/Avg Size
Total = 1400 Mn/5
2. Segment the population Households = 280 Mn
on the basis of income
(Division Basis Income)
3. Calculate number of 4-
wheelers per household Upper Class – 14 Mn
Upper Class Middle Class Lower Class Poor
Middle Class – 42 Mn
4. Calculate the total (5%) (15%) (50%) (30%) Lower Class – 140 Mn
number of 4-wheelers. Poor – 84 Mn
5. Calculate the total
number of Maruti cars
4W - 1.5 4W - 0.75 4W - 0 4W - 0 (Average No. of Vehicles/household)

Assumptions
§ Average household size in
the country = 5
4W – 21 Mn 4W – 31.5 Mn 4W - 0 4W - 0 (Total No. of 4W Vehicles)
§ Maruti cars have a 50%
market share over the
years
§ Lower Class and Poor do
4W - 10.5Mn 4W - 15.3 Mn 4W - 0 4W - 0 Maruti Cars – 50% share
not own any cars
§ Only passenger vehicles
are considered Solution: Maruti Cars = 25.8 Mn

© Management Development Institute, Gurgaon | 2023- 2024 121

H Guess Estimates

Guesstimate 3: Number of Schools

Prompt: Estimate the number of Schools in Delhi

Approach
Top Down (Mix) Population of Delhi
Population of 15 Mn
Delhi
Methodology
1. Calculate the number of (Division Basis Age)
people in school attending
age
2. Split into the population 0-3 years 3-18 years 18-35 years 35-60 years 60+ years 3-18 years : 4.5 Mn
that can afford education (3%) (30%) (33%) (25%) (8%) Rest ignored because on non-
school going age
3. Separately calculate the
No. of Student/School
(Division Basis Affordability)
Can Afford
Assumptions Cant Afford
education People attending school
(10%)
§ 90% of students can
(90%) 4 Mn
afford education in Delhi
§ All schools on average No. of Students/School
function from Nursery to No. of = Classes*Sections*Students/sec
12th students/school = 14*10*50
= 7000
§ Average No. of sections
per school – 10
No. of Classes No. of Sections No. of Students/Sections
§ Average No. of (14) (10) (50)
Student/School – 50
No. of Schools
= Students attending school / No. of students per school
= 4,000,000/7,000
= 5,700 Solution: No. of Schools = 5,700

© Management Development Institute, Gurgaon | 2023- 2024 122

H Guess Estimates

Guesstimate 4: Cars on Mumbai Sea Link

Prompt: Estimate the number of cars using Bandra-Worli Sea Link in Bombay in a day

Approach
Supply Side

Total Time
Methodology (24 Hours)
1. Segment day hours by
traffic load and identify a
bottleneck in the system
– toll booth in this case Low Load Medium Load High Load
2. Estimate the cars per unit (6 Hours) (12 Hours) (6 Hours)
time in that bottleneck
3. Bottom-Up calculation to
arrive at total cars that
has passed through the
bottle neck – toll booth 0.5 * Capacity 1* Capacity 2 * Capacity Toll Booth Usage

Assumptions
§ Number of toll lanes
8
0.5 *60 *5 *8 *6 1 *60 *5 *8 *12 2 *60 *5 *8 *6 Total Cars in each traffic load type
§ Bottleneck =7,200 =28,800 =28,800 Hours*Lanes*Capacity/Hour*Multi.
Toll Lanes
§ Average Toll Lane Capacity
per minute: 5 cars
§ 12 hours of medium Total Cars in a day
traffic load = 7,200 + 28,800 + 28,800
= 64,800
§ 6 hours each of high load
and low load in a day Solution: No. of Cars = 64,800

© Management Development Institute, Gurgaon | 2023- 2024 123

H Guess Estimates

Guesstimate 5: Metro Ridership

Prompt: Estimate the number of people travelling in Metro in Delhi daily

Approach
Bottom Up Coaches
Total Coaches =Operating Hours*Freq.*Coaches
= 20*30*3200
Methodology =1.92 Mn
1. Calculate the No. of
operating hours in a
metro Coaches
Frequency of Metro =lanes *Stations *Avg Coaches
2. Segment occupancy of Operating Hours Coaches/Metro
(No. of Metro's/hour) =10*40*8
metro basis hours
(rush/normal)
3. Adjust for occupancy
Metro lanes = 2 per route
No. of Metro Average No. of
Assumptions No. of Stations = 10 lanes
lanes Coaches
§ Number of metro routes
5 (Maximum Capacity of a coach = 200)
§ Number of Stations: 40
Occupancy Weighted Avg. Occupancy
§ Average # of coaches/metro =(1.2 + 4 + 4.2 + 4 + 2.4)/20
8 = 80%
§ Operating Hours: 20
5 - 9PM
§ Frequency: 1 Metro/2 min 5 - 7AM 7 - 11AM 11 - 5PM 9 - 1AM
4 hours
2 hours (60%) 4 hours (100%) 6 hours (70%) 4 hours (60%)
(100%)

Daily Ridership = Avg. occupancy * Max Capacity * Coaches

= 1.92 * 200 * 80%
= 307.2 Mn Solution: Daily Ridership = 307.2 Mn

© Management Development Institute, Gurgaon | 2023- 2024 124

H Guess Estimates

Guesstimate 6: Lakes in the World

Prompt: Estimate the number of fresh water lakes in the world

Approach
Radius of Earth= 6,400 Km = 26 *100 Km
Top down
Surface Area of Earth = 4 * 3.14 * 26 * 100 * 26 * 100 = 510 Mn Km2

Methodology
Surface are
1. Calculate the surface area of Earth
of earth
2. Divide the area into
Water & Land Water Land Water = 357.5 Mn Km2
3. Separate fresh water (70%) (30%)
from salt water
4. Remove glaciers from Fresh Water = 7 Mn Km2
sources of fresh water Salt Water - Oceans Fresh Water
(98%) (2%)
5. Divide by average size of
lake
R, P, L = 0.14 Mn Km2
Assumptions Glaciers Rivers, Ponds, Lake
(98%) (2%)
§ Assuming Earth to be a
perfect sphere
§ Average size of a lake is Rivers Lakes Ponds Lakes= 14,000 Km2
100m * 100m (85%) (10%) (5%)
§ Most fresh water is in
terms of Glaciers
Total No. of lakes = Area of Lakes / Size of 1 lake
= 14,00,000 Lakes

Solution: Total Lakes = 14 Lacs

© Management Development Institute, Gurgaon | 2023- 2024 125

H Guess Estimates

Guesstimate 7: AC working hours

Prompt: Estimate the average number of hours AC are run in Delhi over a year

Approach
Demand Side

Methodology AC Used
1. Divide the months in
terms of seasons Summer Months : March - July
2. Average hours that the Total 153 days
AC is working Autumn Months : Aug – Oct
Summer Autumn/Rainy Winter Total 91 days
3. Consolidate
Winter Months : Nov – Feb
Total 120 days
Assumptions
(Split by Average hours/day)
§ Consideration of a middle
income household Morning - 2 Morning - 0 Morning - 0
Afternoon/Evening -1 Afternoon/Evening - 0 Afternoon/Evening - 0 Total hours per segment of day
§ Division of months done Night - 4 Night - 3 Night - 0
basis judgement

Hours worked = No. of days * Average Hours/day

= 153 * 7 + 91 * 3 + 120 * 0
= 1,344 hours

Solution: AC Working hour = 1344 hours

© Management Development Institute, Gurgaon | 2023- 2024 126

H Guess Estimates

Guesstimate 8: Length of roads in Delhi

Prompt: Estimate the total length of roads in Delhi

Approach
Top Down

Delhi's Region
Methodology
1. Calculate Area of Delhi
2. Divide the area basis its Roads
Residential Commercial Forest/Agri Others
usage Network
(30%) (25%) (10%) (15%)
(20%)
3. Divide the roads then into
usage (Split by Road Maintenance)
Municipal
4. Assume approx. width of State Roads Central Govt
Roads
roads (50%) (15%)
(35%)
5. Divide the two

Assumptions Main Roads Highways Other Interconnected Split based on experience

(30%) (20%) (50%)
§ Area of Delhi –1800 Km2
§ Width of road based on
experience Estimated Avg. Width Total
Municipal – 5m Area of Road Length
Main roads - 15m Municipal 125 Km2 5m 25,000 Km
Highways – 20m
Other Ics – 10m Main 50 Km2 15m 3,000 Km
Central Govt – 25m
Highways 35 Km2 20m 1,750 Km
Other IC 85 Km2 10m 8,500 Km
Central Govt 50 Km2 25m 2,000 Km
Solution: Length of Roads = 40,250 Km

© Management Development Institute, Gurgaon | 2023- 2024 127

H Guess Estimates

Guesstimate 9: Sauces sold in Europe

Prompt: Estimate the total revenue generated from consumption of sauce in Europe

Approach
Top Down Population of Europe – 750 Million
1 Kg of Tomato Sauce = 75 Servings
[1Kg of sauce can be used by a family of 5 for 15 servings (Assumption based on own experience)]
Methodology
1. Divide population basis
eating routine of a day Meals
2. Attach values of Sauce
usage basis experience
3. Calculate servings using 1 Breakfast Lunch Dinner General Snacks
kg of sauce (90%) (90%) (60%) (50%)
4. Consumption * Average (Sauce usage : High) (Sauce usage : Low) (Sauce usage : Low) (Sauce usage : High)
Price

50% consume 10% consume 10% consume 70% consume

Assumptions tomato sauce tomato sauce tomato sauce tomato sauce
§ Considering on tomato
sauce for the guesstimate Based on above analysis –
§ Average price of sauce = Total Population consuming sauce
1.5 Euro/Kg = 750 Mn * (0.5*0.9 + 0.1*0.9 + 0.1*0.6 + 0.7*0.5)

Amount of Ketchup consumed in Kg = (750 Mn * 0.95 Kg)/75 = 95 Mn Kg/Day

Consumption/month = 95 Mn Kg/day * 30 = 285 Mn Kg/Month
Annual Consumption = 3.42 Bn Kg/year

Annual Revenue = 3.42 Bn Kg/year * 1.5 Euro/Kg = 5.43 Bn Euros

Solution: Sauces Revenue = 5.43 Bn Euros

© Management Development Institute, Gurgaon | 2023- 2024 128

H Guess Estimates

Guesstimate 10: T-shirts Sold on Amazon per Day

Prompt: How many t-shirts do you think Amazon is selling in India per day?

Approach
Population of India = 1500 Mn
Top down

Internet Penetration = 0.4*1500

Methodology = 600 Mn
(40%)
1. Segment by gender
considering different
demand = 0.5*500
Ecommerce Shoppers
2. Further break down into = 300 Mn
(50%)
frequent and occasional
shoppers rate of t-shirt
purchase per year = 0.3*300
3. Bottom up estimation for Fashion Shoppers = 90 Mn
total t-shirts sold in a (30%)
year
= 0.5*90
Assumptions Men Women = 45 Mn
§ E-commerce penetration =
(50%) (50%)
50% of Internet
Penetration
Frequent Occasional Frequent Occasional M (FS) : 9 Mn *5 = 45 Mn
§ Fashion Shoppers on
Shoppers Shoppers Shoppers Shoppers M (OS) : 36 Mn *3 = 108 Mn
Ecommerce = 30%
(20%) (80%) (20%) (80%) F (FS) : 9 Mn *7 = 63 Mn
§ Frequent M – 5 T-shirt F (OS) : 36Mn *4 = 144Mn
Frequent F – 7 T-shirt
Total T-shirts = 45 + 108 + 63 + 144 = 360 Mn
§ Occasional M – 3 T-shirt Amazon Market Share = 30%
Occasional F – 4 T-shirt
Total Sales/Year = 108 Mn
Total Sales/Day = 108 Mn / 365 = 295K Solution: Total Sales/Day = 295K

© Management Development Institute, Gurgaon | 2023- 2024 129

H Guess Estimates

Guesstimate 11: Size of Tyre Industry

Prompt: Estimate the size of the tyre industry, particularly SUV in India in 2025.

Approach Population of India = 1500 Mn

Demand Side Average Household Family Size = 5

Methodology = 1500 Mn / 5
Total
1. Divide households based Households = 300 Mn
on income stream
2. Within High income, = 300 Mn * 10%
divide basis the No. of Below Poverty Low Income Middle Income High Income = 30 Mn
SUVs in a household (15%) (40%) (35%) (10%)
3. Divide basis current on
ground + New SUVs
4. Multiple by growth rate One SUV Two SUV Multiple SUVs No SUV
for future years (50%) (30%) (10%) (10%)
5. Divide by average
number of tires per SUV
Based on above analysis –
Total SUVs on ground = 30 Mn * (0.5*1 + 0.3*2 + 0.1*3) = 42 Mn
Assumptions
New SUVs = Total SUVs on ground/ Average Life = 42 Mn/15 = 2.8 Mn
§ No replacement tyre being As of 2023, 42 Mn SUVs & 2.8 Mn new SUVs = 44.8 Mn
considered
As of 2025, Total SUVs = 54.2 Mn ( 44.8 * 1.1 * 1.1)
§ Average life of SUV is 15
years
Each SUV has 5 tyres (4 tyres + extra)
§ Assume growth rate of Total = 54.2 Mn * 5 = 271 Mn Tyres
tyre industry as 10%
§ Each SUV has 5 tyre
Solution: Total No. of tyres = 271 Mn

© Management Development Institute, Gurgaon | 2023- 2024 130

H Guess Estimates

Guesstimate 12: Rakhis sold online

Prompt: Estimate the number of rakhis sold online in India

Approach
Population of India = 1500 Mn
Top Down

Methodology India's Population

1. Divide population of India

basis Internet access
No Internet Access Internet Access
2. Then do a religion split (60%) (40%)
based on the ones
celebrating
(Split by Religion)
3. Incorporate economic Others Muslims Hindu/Punjabi
status split (5%) (15%) (80%)
4. Use the split for
offline/online buying (Split by whether festival is
Non-Celebrating Celebrating
celebrated)
(20%) (80%)
Assumptions
§ Buying pattern for Rakhis Above Poverty Line Below Poverty line
is mostly local and offline (80%) (20%)
§ Most of the ones having
internet access celebrate Buys Offline Buys Online
the festival (90%) (10%)

Total = 1500 * 40% * 80% * 80% * 80%* 10% = 30.7 Mn

Solution: Total Rakhis sold online = 30.7 Mn

© Management Development Institute, Gurgaon | 2023- 2024 131

H Guess Estimates

Guesstimate 13: WhatsApp Messages

Prompt: Estimate the total number of WhatsApp messages sent in India per day

Approach
Total Number of Smartphone users in India Total Number of WA messages/Day
Top Down

Methodology Population
(1500 Mn) No. of SP Users
1. Calculation of total
smartphone users in India
2. Use economic split No. of Households
(1500/5 =300Mn) 0-15 years 15-40 years 40-60 years 60+ years
3. Age split for smartphone (0%) (70%) (20%) (10%)
user
Low Class Middle Class Upper Class
4. Assume usage for each (30%) (60%) (10%) Split basis usage of WhatsApp
customer (90 Mn) (180 Mn) (30 Mn) (Denotes: Average message/day)
5. Weighted average on High Med Low
Average Average Average
number of messages
Smartphone/House Smartphone/House Smartphone/House 15-40 60% (60) 25% (30) 15% (20)
hold : 1 hold : 4 hold : 5
Assumptions 40-60 30% (30) 40% (20) 30% (10)

§ All smartphone users are 90*1 = 90 Mn 180*4 = 720 Mn 30 * 5 = 180 Mn 60+ 10% (25) 30% (15) 60% (5)
using WhatsApp
§ Average Total = 990 Mn *
Smartphone/Household
(0.7 * (0.6*60+0.25*30+0.15*20) +
Total = 90 + 720 + 180 = 990 Mn (0.2 * (0.3*30+0.4*20+0.3*10) +
(0.1 * (0.1*25+0.3*15+0.60*5))
Total = 36,481 Mn

Solution: WA messages/day = 36,481 Mn

© Management Development Institute, Gurgaon | 2023- 2024 132

H Guess Estimates

Guesstimate 14: People flying in/out from Delhi

Prompt: Estimate the number of people flying in/out from Delhi in a day

Approach
Total Number of Flights in Delhi Total No. of People flying in/out
Top Down

Methodology Type of Bookings

Delhi Airport
1. Calculate total number of
flights in Delhi
2. Use efficiency & hour wise Economy Economy
Business
split to get to total Terminal 1 Terminal 2 Terminal 3 Class Plus
number (3 runways) (3 runways) (6 runways)
3. Find average capacity by 7 column*20 rows 4 column*15 row
max capacity * occupancy Average time for take off/landing = 2 mins =25 passengers
= 140 passengers = 60 passengers
Total 30 Planes can land or take off in an hour
4. Divide the two

Assumptions % Efficiency of
Maximum Passengers/plane = 220
Airport
§ Number of runways in Average occupancy of plane = 80%
Delhi airport Total passengers/plane = 180
§ Average time – 2 mins 11PM - 6AM 6AM - 12PM 12PM - 5PM 5PM - 11PM
§ % efficiency split/hour (70%) (90%) (100%) (100%) People flying in/out = Passengers/flight * No. of
Flights = 180 * 7668 = 1.38Mn
§ Occupancy rate of plane =
80%
Total = Hours * % Efficiency * Planes * Runway

= 7*0.7*30*12 + 6*0.9*30*12 + 11*1*30*12

= 7,668 Flights
Solution: People Flying in/out = 1.38 Mn

© Management Development Institute, Gurgaon | 2023- 2024 133

H Guess Estimates

Guesstimate 15: Sports Shoes Market

Prompt: Estimate the annual size of sports shoes in India

Approach
Population of India = 1500 Mn
Demand Side

Methodology India India India

1. Find relevant population
by division in terms of
urban/rural, gender & age Male Female Below 15 15-30 30-45 45-60 60+
Rural Urban
2. Use the weighted average (70%) (30%) (50%) (50%) (25%) (25%) (20%) (20%) (10%)
method
3. Find weighted average Market Market Market Market Market
price of a shoe by market Market Market Market Market
Available Available Available Available Available Available Available
variants Available Available
(40%) (80%) (60%) (40%) (2%)
(70%) (100%) (80%) (30%)
4. Divide the two

Weighted Average: ~80% Weighted Average: ~55% Weighted Average: ~60%

Assumptions
Total Relevant Population = 1500 Mn * 0.8 * 0.55 * 0.6 = 396 Mn
§ Sports shoes are generally
used by athletes and by
Market Spent
citizen for going to a
walk/job or day to day
purpose Mid Level Total Market Value = 396 Mn * 960
Budget (60%) Premium (10%)
(30%) = Rs. 380 Billion
§ Average prices of shoes in
a segment Average Price: Average Price: Average Price:
500 1200 3000

Weighted Average Price: Rs.960

Solution: Total Market Size = Rs. 380 Bn

© Management Development Institute, Gurgaon | 2023- 2024 134

 Water Consumption in Kozhikode in a day Guesstimates

Question
Estimate the water consumption in Kozhikode in 1 day?

Preliminary Questions
Should I consider water intake alone?

- Yes

Should I include Industries?

- Let's ignore that for now. Focus on the daily household

consumption

Can I neglect the effect of COVID-19?

- Yes

Assumptions
± Population of Kozhikode= 3 millio×
± 70% of the population comes under drinking age group of
21-60¨
± Average water consumption is 4L¨
± Average household size is 4¨
± Average alcoholic consumption is 0.75L for which 66% water
is used I.e. 0.5L & average non-alcoholic consumption contain
50% water.Ï
± 50% Males & 33.33% females consume alcohol in a day¨
± Non alcoholic Consum. : < 21 yrs, 0.5 ltr ; 21-60 yrs, 1 ltr ; < 60
yrs 0.25 ltr

Approach
± Water intake happens through direct water consumption, Water Consumption: 0.6 Mn *3L+ 1.8 Mn*4L + 0.6 Mn*5L = 12 Mn Lt.

beverages- alcoholic & Non- alcoholic, and through foo) Alcoholic Beverage: (0.525 Mn +0.315)* 0.5L = 0.42 Mn Lt.

± Number of household = Population / Avg household siz& Non Alcoholic Beverage: (0.9 Mn *0.5 L+ 1.8 Mn *1L + 0.3 Mn * 0.25) *(1/2) =2.325 Mn Lt.

± Consumption of water = Per meal water usage * # of meals Household Water Consumption: 0.75 Mn*4L*3 Meals = 9 Mn Lt.

Total Water intake by Kozhikode per day: 12+0.42+2.325+9 = 23.745 Mn Lt.

80
Indian Institute of Management, Kozhikode
 # of regular size Pizzas Domino’s sells in India per day Guesstimates

Question
Estimate the water consumption in Kozhikode in 1 day?

Assumptions
Timings

Domino’s Opening Hr: 11AM to 12 Midnight (13 Hrs)

Peak Hour Timings Hours
Lunch Time 12:30 - 2:30 2 Hrs
Dinner Time 8:00 - 11:00 3 Hrs
5 Hrs
Stor¤
µ Total Area of India = 3.2 Mn sq. Km. ~ 3 Mn Sq. Km¢
µ 10 % of the area is urbanŽ
µ Each 200 sq. Km. Of urban is served by a single store

# stores = (3,00,000 Sq. Km/200 Sq.Km.) = 1500 Stores

Consumption
µ No of customers visiting onsite in peak hours is 15/ Single store Consumption

Hr. Each customers consume a single regular pizzaò = (Dine In + Home Delivery) = (110+180)

= 290 Pizzas /Store

µ # of customers ordering for home delivery in peak

hours is 20/Hr. Each customers consume a single Total Consumption

regular pizzÏ = (No.of Stores * Consm. Per store) = (1500*290)

= 4,35,000 Pizzas
µ Consumption during Non-Peak hours is 30% of Peak
hours for dine In and 50% for Peak hours for Home
delivery.

81
Indian Institute of Management, Kozhikode
 Guesstimates
Pizzas ordered in a day in India

Question

How many pizzas are ordered in a day in India ?

Assumptions

Population of India = 130 Cr

Population Percentage

Rural 91Cr 70%

Urban 39 Cr 30%

Urban Population Percentage

< 5 Years 10%

5-60 Years 80%

60+ Years 10%

Consumption

¢ People in rural areas don’t consume pizzaŒ

¢ Pizza is luxury so people from the lower income

groups can’t afford itŒ

¢ People above 60 years old don’t consume pizzŽ

¢ People below 5 years, don’t consume pizzŽ

¢ A single pizza is shared by 3 people

Pizza/month ~ 3 Cr

Pizza/day : 3 Cr/30 = 10 lakhs/ day

82
Indian Institute of Management, Kozhikode
 Profit a typical Tea Stall makes in a day in Delhi Guesstimates

Question
How much profit does a typical tea stall make in a day in Delhi?

Assumptions
Timings

Tea Stall Opening Hr = 6 AM to 8 PM (14 Hrs)

Timings Demand
High 6:00 - 10:00 160
Raw Material Rate Used/Cup Cost (450 cups/day)
Low 10:00 - 16:00 120
Milk 40/L 30 ml (450*30*40)/1000 ~ 540
High 16:00 - 20:00 160
Tea Leaves 200/Kg 1 tsp/4g (450*’4*200)/1000 ~ 360
Total Demand ~450
Sugar 20/Kg 1.5 tsp/4g (450*1.5*4*60)/1000 ~ 160

Preparation Masala Fixed Rs 2/Cup 2*450 ~ 900

Gas Cylinder Fixed ---- 100
– In case of high demand, the capacity utilisation is
full. Let us say the tea maker can make around 20 Disposable Cup Fixed Rs 2/Cup 900
cups of tea at a time , that take around 30 minutes to
be consumed by customers. So, the tea cups made Total Demand ~ Rs 3000
in high demand are 40 Cups/hr¤

– In Noon, Let's say only 50% of capacity is utilised. Revenue from selling tea at the rate of Rs 10/Cup = 450*Rs 10 ~ Rs 4,500

So the tea maker makes only 20 cups/hr. Total Cost = Rs 3,000

Profit = Rs 1,500

Consumption
– Consumption of tea across all region of Delhi are
uniformº
– Demand of tea is high in the morning & Evening,
low in the afternoon.

83
Indian Institute of Management, Kozhikode
 Sanitary Napkin use over a month in India Guesstimates

Question
Estimate the number of women who use sanitary napkin in India over a month.

Estimation
Indian population: 130 crores

50% are women = 65 crores

Age group of women who menstruate: 13-50 years

% of women under the age group 13-50 years:

~25% women are <13 years

~15% >50 years

women under menstruation age:

60% of 65 crores = 39 crores

Net market size: ~ 40 crores

û The ideal menstruation cycle: 28 days.

Taking 30 days, there are 12 cycles in a year.

Considering 10% women have 10 cycles/year and 10% women have

14 cycles/year. It averages to 12 cycles/year¿
û Now, to subtract the pregnant women, let’s say there are 2
crore births per year in India

The above assumption can be made based on discussion with the

interviewer¿
Considering the urban-rural divide

û It implies, 2 crore women are pregnant in a year£

30% women in urban areas. Let’s say 75% of them can afford sanitary napkins,

û Let’s assume these women were pregnant for 6 months in a 40*0.3*0.75 = 9 crores

year on average£ 70% women in rural areas. Let’s say 25% of them can afford sanitary napkins,

û So, out of 40 croresø 40*0.7*.25 = 7 crores

Æ 2 crore women have menstruation 6 times/year ,

Total women who can afford = 16 crores

So 1 crore women/ mont¾ Let’s say 10% of them prefer to use options like menstrual cups, tampons,etc = 1.6 crores

Æ 38 crore women have normal menstruation

No. of women using sanitary napkins/month = 16-1.6 crores = 14.4 crores
12 times/year i.e. once per mont¾
û In total, 39 Crores, is the new target market size.

84
Indian Institute of Management, Kozhikode
 How much a typical PVR Plaza make in a day Guesstimates

Question
How much revenue does a typical PVR plaza of India make in a day ?

Assumptions Revenue in Weekdays : Mon-Thu Revenue Assumptions

Timings
Timings Capacity Revenue
PVR Opening Hr = 9 AM to 3 AM (18 Hrs)
09:00 - 12:00 25% 300*0.25*3*1*100 = 22,500
Revenue 12:00 - 18:00 6 0% 300*0.6*3*2*250 = 2,70,000

 Movies and Food outlets are the only source 18:00 - 00:00 7 0% 300*0.7*3*2*400 = 5,04,000
revenue 00:00 - 03:00 50% 300*0.5*3*1*300 = 1,35,000
Capacity Total 9,31,500

¢ PVR has 3 movie halls on averagª Revenue in Weekends : Fri-Sun

¢ Each hall can have 6 movies/day with an average
capacity for 300 people. Timings Capacity Revenue
09:00 - 12:00 35% 300*0.35*3*1*100 = 31,500
Variation Assumption 12:00 - 18:00 7 0% 300*0.7*3*2*250 = 3,15,000
¢ The demand is not varying based on type of moviesÄ 18:00 - 00:00 80% 300*0.8*3*2*400 = 5,76,000
¢ The demand is uniform on all daysÄ
¢ Taking the price of all seats at the same 00:00 - 03:00 6 0% 300*0.6*3*1*300 = 1,62,000
Total 10,84,500

Pricing Assumptions Revenue from the Movie/Day = ( 9,31,500*4 + 1084500*3 ) / 7 = Rs. 9,97,072

Timings Demand Price/Ticket A ssuming 20% of people in each slot buy from PVR food outlet paying Rs. 300 on an average

09:00 - 12:00 1 100 =[( Capacity on a week day *4*300 ) + ( Capacity on a day of a weekend *3*300)] /7

= (475200 +639900) /7

12:00 - 18:00 2 250 = 159300

18:00 - 00:00 2 400

00:00 - 03:00 2 250 Revenue Per Day = Rs 11,56,372 ~ Rs 12 Lakhs

85
Indian Institute of Management, Kozhikode
 Number of deliveries done by swiggy in a day Guesstimates

Question
Number of deliveries done by Swiggy in a day

Approach

Assumptions

p Considering that only the urban population

uses food delivery platforms in IndiŒ

p There is one restaurant/ eateries/ cloud

kitchens etc. per 400 people and 30% of these
restaurants are with aggregators like Dunzo,
Swiggy, Zomato etc. and 20% is with Swiggy
(exclusive and along with other aggregators\

p For estimating number of delivery partners

available on an average day, consider the
average of peak (100%) and non-peak (50%)
days is (1 lacs + 0.5 lacs)/2 = 75,000. On a daily
level, a delivery boy can take 2 deliveries in 60
minutes and work for 10 hours a day.

86
Indian Institute of Management, Kozhikode
 Guesstimates
Number of Ac’s sold in a year in India

Question

To calculate the number of ACs sold in a year, consider only household demand, discount commercial demand for now

Assumptions
Income Wise Distribution Rural (~20 Cr) Urban (~10 Cr)
s Average life of AC = 10 year`
Poor (20%) 4 Cr 2 Cr
s Growth Rate = 5%

Middle Class (40%) 8 Cr 4 Cr

Upper Middle Class (30%) 6 Cr 3 Cr

Approach
Rich (10%) 2 Cr 1 Cr
 There can be two approaches for this guesstimate: Demand

side i.e., population-based and Supply side i.e., number of

manufacturers of air conditioners in Indi£

Penetration Rate Rural Urban

 For this case, I would like to take a demand side approach

Poor Nil Nil
which includes both new demand and replacement demand¬
Middle Class 5%*8 = 0.4 Cr 30%*4 = 1.2 Cr

s New Demand = Growth rate * Market Size Replacement

Upper Middle Class 30%*6 = 1.8 Cr 60%*3 = 1.8 Cr
Demand = Current Market Size/Life
Rich 80%*2 = 1.6 Cr 100% *1 = 1 Cr

Total Market Size for air conditioners = 7.8 cr

New Demand = Growth rate * Market Size = 5%*7.8 = 0.39 crores

Replacement Demand = Current Market Size/Life = 7.8/10 = 0.78 crores

Total Demand = New Demand + Replacement Demand

= 0.39 + 0.78 = 1.17 crores

87
Indian Institute of Management, Kozhikode
GUESSTIMATE

29
2
 GUESSTIMATE FACT SHEET

Stats India City Population (millions)

GDP ₹ 135 lakh crore Delhi 30
GDP per capita ₹ 85,000 Mumbai 22
GDP growth rate (2019) 4% Kolkata 15
Below poverty line 25% Bengaluru 12
People in household 4 Hyderabad 10
HDI Medium Jamshedpur 1.6
Stats India World
Population 1.4 Billion 7.8 Billion
Life expectancy 70 years 73 years
Sex ratio (F:M) 940:1000 990:1000
Vegetarian 30% 10%
Urban 35% 55%
Internet penetration 40% 60%
Literacy rate 70% 85%
English speaking 15% 20%
29
3
Guesstimate Guestimate the No. of E-Rickshaws in Kolkata BCG

Case Statement:
Guesstimate the total number of e-rickshaws in Kolkata

Population
50 lakhs
Background Information
• E – Rickshaw: Typical High Income: Middle Income: (50%) Low Income:
tuk–tuks (totos) seen in (20%) 25 lakhs (30%)
cities these days. 10 Lakhs 15 Lakhs
• New replacement for
cycle rickshaws in the Don’t use public Public Transport Public transport
city. transport (50%) (60%)
• Cycle rickshaws are 2
seater and pedal-driven Using E-
Using E-Rickshaw: (20%) 2.5 Rickshaw: (10%)
or hand pulled. lakhs / 200 km2 = 1,250 / km2 0.9 lakhs / 200 km2
• E rickshaws are 4 = 450 / km2
seater and driver by
Crowded Area Non-Crowded (80 Crowded Area Non-Crowded (80
electric motor. (120 km2) km2) (120 km2) km2)
1.5 Lakhs 1 Lakh 0.54 Lakhs 0.36 Lakhs

3 passengers + 1
4 passengers + 1 3 passengers + 1 4 passengers + 1
driver per ride:
driver per ride: driver: 25,000 driver per ride:
9,000 rides
30,000 rides rides ~11,000 rides

10 rides / 10 rides / 10 rides /

10 rides /rickshaw:
rickshaw: 3,000 rickshaw: 2,500 rickshaw:
rickshaws rickshaws 1,100 rickshaws
900 rickshaws

Case tips
• Include the crowding factor.
• No of rides / demand will be higher in areas with higher population 3,000 + 2,500 + 1,100 + 900 = 7,500 E - Rickshaws
density.
• Include multiple factors and make it comprehensive (not a linear
approach).
• Basic BQs ( tell me about yourself and why is consulting of interest). 294
 Guesstimate Guestimate the No. of E-Rickshaws in Kolkata BCG

C: Sure. By E- Rickshaws we are referring to the typical totos that I: Don’t you think the approach is a bit linear. You can try to
have become popular in some cities right. consider factors like the demand for this vehicle.
I: Yes. You can imagine the usual cycle rickshaw. However, this will C: Sure. In that case, the population density can be factored in. May
not be a 2-seater but a 4-seater and driven by an electric motor. I proceed by taking an area of 200 km2.
C: Alright. I’d like to go by the population approach. Considering the I: Sure.
population of Kolkata to be around 50 lakhs. Will that be a fair start?
C: (Obtained the population density) I’d assume a larger chink to be
I: Okay. Go on. crowded in the city so roughly 60% of the 200 km2 will be crowded
and will have e – rickshaws at full capacity while the non crowded
C: Then I would split the population by the income in the high, places will see roughly 3 passengers on an average. Then I would
medium, low categories with a 20%, 50%, 30% split respectively. In take a horizontal 10 rides per rickshaw per day assumption and
the middle-income category 50% would use public transport and in calculate.
the low income 60% would use public transport. Thereof, 20% and
I: So what is the final number?
10% respectively might go for E- Rickshaws. Next, I’d look into the
number of rides and try to arrive at a figure. I’d request you to let me C: (After a calculation break) I’m getting a figure of roughly 7,500
know if I am missing something. E- rickshaws in the city.
I: Alright. That will be all. We may close now. Thank You.
C: Sure. Thank you so much.

295
 Guesstimate Guestimate the No. of E-Rickshaws in Kolkata BCG
(alternate solution)

C: Before I begin, I'd like to ask a few clarifying questions. I: You may proceed with the demand side approach.

I: Sure. C: Sure, for Delhi, I would estimate using the following levers: First divide
the population into income classes as this would help me analyse the
C: If I understand correctly, e-rickshaws are 3-wheelers that usually commute modes of transportation, then bifurcate by age group. Thereafter I will see
smaller distances, say max 2-3 kms at a stretch. different modes used by people for commute. Here I'd take different time
I: Yes, that’s right. stamps- peak and non-peak hours to determine number of people using say
a metro at a point in time.
C: Do you want me to estimate market size in terms of units or value? Once we have number of people travelling in public transport and given
that an e-rickshaw can include 4 people at a time, I'll get the number of e-
I: Units.
rickshaws in Delhi.
C: Is there a particular timeline we are looking at or the number of units on
I: Makes sense.
road as of today?
C: Lastly, I will extrapolate this to estimate the size in all tier 1 cities in
I: The latter.
India and follow similar process to estimate for the Tier2/3 cities.
C: Sure, lastly, what is the average life of an e-rickshaw?
I: Sure, we can end the case here.
I: Assume it to be 5 years.

C: I would like to elucidate my approach first. To estimate the market size for
India, I'd like to take proxy by estimating for a tier 1 city and a tier2/3 city. I
will extrapolate those numbers for the entire of India using the adequate
weights.

I: Sounds good!

C: I will start with Delhi since I am most acquainted with this place. As per
my knowledge, these vehicles usually are available around metro stations, bus
stops and local busy markets.

I: You can ignore market areas.

C: Sure, there can be two approaches to estimate this. The demand side
wherein I'll see the number of people who commute via these public transports
and use e-rickshaws, and the supply side wherein I'll use data regarding the
number of e-rickshaws on road when they were introduced, the annual average
growth rate coupled with the average life of an e-rickshaw 29
6
 Guesstimate Revenue of Bandra-Worli sea-link BCG

Case Statement:
Estimate the revenue for Bandra-Worli sea-link on any given day.

Background Information Revenue

• Place: A toll-bridge in
Mumbai with 8 tolls: 4
for each side of travel Revenue per type of
Frequency of cars
• Product: Only 4-wheelers No. of Tolls passing through No. of hours
vehicle
each toll gate
Case tips
To (Rush hour Fro (Rush hour Light Motor Heavy Motor Peak Hours (120 Peak Hours (6;
• Supply side approach morning): 4 evening): 4 Vehicle Vehicle cars/hour) To-3 hrs, Fro- 3
optimum for the problem (50/- one way) (75/- one way) hrs)
• One car assumed to take
30s to travel through a Non-Peak Hours
toll gate in peak hours (50 cars/hour) Non-Peak Hours
(15)
(120 cars/hr/toll gate).
This baseline was used to Night Hours (10
estimate frequency of cars cars/hour)
Night Hours (6)
in non-peak & night hours
• For peak hours and non-
Peak Hours Non-Peak Hours Night Hours
peak hours, LMVs were
considered to constitute LMV HMV LMV HMV LMV HMV
90% of vehicle volumes Total Revenue for one side
No. of hours 3 3 15 15 6 6
and HMVs as 10%. The (4 lanes) = 64,800 + 10,800 +
converse was assumed for Revenue/Car 50 75 50 75 50 75 135,000 + 22,500 + 1,200 +
night hours. (Note: The 16,200 = 250,500
108 12 45
assumptions were No. of Vehicles 5 (10%) 1 (10%) 9 (90%)
(90%) (10%) (90%)
validated by interviewer Total Revenue for all both
before proceeding with No. of Lanes 4 4 4 4 4 4 sides (4*2=8 tolls):
the calculations) 250,500*2=501,000
Total Revenue 64,800 10,800 135,000 22,500 1,200 16,200 29
7
 Guesstimate Wheat in India BCG

Case Statement:
Estimate the annual wheat production in India

Background Information India population 75% staple wheat

consumption ~100cr
• Product: Wheat
• Assume consumption is
sourced domestically Urban 30cr Rural 70cr

Low income 12 Middle income Urban average +

High income 3 cr
cr 15 cr 20%

Average = 0.6*10 Low income avg Low income avg

+ 0.4*6 = 8.4 + 2 = 10.4 + 4 = 12.4

Annual consumption calculation

Younger than 15, 15 to 60 yrs –
older than 60 – 40% 60% Total daily rotis: Urban = 294 cr, Rural = 823.2 cr. Total = 11.1 B

Annual rotis = 11.1 * 365 = 4077B rotis

6 rotis daily 10 rotis daily From personal experience: 10kg wheat lasts 15 days with 24 rotis daily
Wheat in 1 roti = 10kg/15/24 = 27g
Annual 11.1B * 0.027 = 113.25M tonne

Case tips
• Walk through the approach before you start multiplying the numbers at every stage. List the factors and weights you would be
considering; in case you miss something, interviewer can easily bring it up. In case you are going into too many details, interviewer can
point that out as well.
• Estimating wheat from "roti" was essential, the approach was appreciated by the Interviewer. Most guestimates have this part where you
29
can estimate something from daily life. 8
 Guesstimate
Guesstimate Fans in India Accenture Strategy

Case Statement:
Can you estimate the number of fans in India currently?

Total fans
Background Information
•Ceiling fans
•Only residential
Residential Commercial

Rural Urban
population population
Account for differences in people/ household

Rural Urban
households Account for differences in fans/ segment households

Low Low
income income

Middle Middle
income income

High High
income income

Total fans
Account for homeless and non-electrified

Final answer

29
9
 Guesstimate Fans in India Accenture Strategy

C: Sure sir, are we looking at any particular type of fan like ceiling or table? I: Yes, you can do that, but tell me how would you define the consumption
Or should I consider all types of fans? pattern of these income groups?

I: Let’s focus only on the ceiling fans for the scope of this problem. C: Sure sir, based on average no. of rooms in the household, I would
roughly estimate the following consumption rates (No. of fans per
C: Understood. Breaking the problem down, the 2 types of establishments that
household) for the respective categories:
we can consider are ‘Residential’ and ‘Commercial’. Since majority of
commercial establishments are centrally airconditioned, for simplicity, can we 1. High Income (Urban) - 6
focus only on the Residential aspect for now? 2. Medium Income (Urban) - 5
I: Sure! You can ignore the ‘Commercial’ aspect for the scope of this 3. Low Income (Urban) - 4
guesstimate. 4. High Income (Rural) - 3
5. Medium Income (Rural) - 2
C: So, I’ll try to arrive at an answer by estimating the number of households in
across demographics and then multiply them with an estimated no. of fans per 6. Low Income (Rural) - 1
household. Can I start with this approach? Now, I would multiply the consumption rates (No. of fans per household)
with the number of households to arrive at the total number of fans. Shall I
I: Yes, please go ahead. proceed with this?
C: Assuming the population of India to be roughly 130 Cr, I would first split I: I understood your approach, no need to do all the computations. Tell me
the population base into 2 major segments ‘Urban’ and ‘Rural’ with 30% & one thing, after arriving at the final number, are there any further
70% split approximately. Is that reasonable? considerations you would like to make?
I: Sure, carry on. C: I can look at the product lifecycle of a fan (how frequently it is replaced)
to estimate a yearly number?
C: That translates to roughly 90 Cr people in rural and 40 Cr in the urban
segment. Next, to arrive at the number of households, is it safe to assume I: No, that is not required. Anything else you might’ve missed on? Take
average family size to be 4 and 6 for urban and rural households your time to think.
respectively?
C: Since we have considered the complete population base, I can probably
I: Sounds reasonable. adjust the final number for homeless people and also areas which are not
electrified within rural segment?
C: Perfect! So that gives us 15 Cr rural and 10 Cr urban households
approximately. Drilling down further, I would like to segregate these I: Perfect! How would you factor that in?
households on the basis of income level - Since it can significantly influence
the house size and therefore the number of fans in the household. C: For the sake of simplicity, I would take a 5% haircut on the final
number to factor in the above considerations. Is that a fair estimate?
I would like to break it into 3 types of income groups – High, Medium and
low. Can I estimate the respective percentages to be 10%, 30% and 60% I: Sure, sounds good! We can close the discussion here. Thank you!
respectively? 30
C: Thank you, sir!
0
 Guesstimate Service Launch For A Hospital Chain Bain

Case Statement:
Client runs a chain of hospitals across India. They want to open specialized maternity care centers. Estimate the market potential and the
number of centers to be launched.

Background Information
Population of
• Client: Current business India
model like Fortis
• Competitor: none
(direct)
Urban Rural
• Service details: End-to-
(30%) (70%)
end (pregnancy to
couple’s training to
initial months nursing)
• Price: 4 lakhs per Tier-1 Tier-1 Tier-3 &
package (10% of (40% of below (50%
30%) 30%) of 30%)

Rich
(10% of 10% of 30%)

Middle Class
(60% of 10% of 30%)

Poor
(30% of 10% of 30%)

Case tips
• Urban/rural and income segmentation are some of the quickest ways to achieve an approximate number in a guesstimate approach.
• Clearly drawing the structure and mentioning the formula makes it easy for the I as well. 30
1
 Guesstimate Service Launch For A Hospital Chain Bain

I: What is the motivation behind launching this service? available.

C: It is part of a long-term business expansion strategy. Competitors are And then target ‘0.3 %’ of child births per year. This will help us identify the
offering the service in parts, but no end-to-end package exists. total market size that we can target.

I: How will you do market sizing? I: We have identified the market potential. How to identify the no. of centers
to be launched?
C: •Start with Indian population. Segment into urban and rural market.
C: No. of centers = Child births per year (from above data)/365 divided by
•This kind of premium service should be targeted at tier-1 cities.
Child births/center/day (from current market understanding)
•Further, rich people in tier-1 cities will be able to afford this kind of service.
Child births/center/day will be dependent on factors such as size of client’s
•This will lead to ‘0.3 %’ of population of India. facility, no. of doctors available, etc.

I: This will help you identify the no. of individuals. How will you proceed No. of centers should also be double checked against the total area of the city
further if official data can be directly accessed? to ensure proximity of center in case of customer need.

C: Client can directly access the data for child births per year across India, if

30
2
 CCTV Cameras in XLRI
Guesstimate Bain

Case Statement:
Guess the number of CCTV cameras that are installed in the campus of XLRI.

Background information
CCTV
• Area of XLRI : Cameras
200000m2
• Area covered/camera:
150m2
Indoors Outdoors

Student Faculty Admin Public

related related related Spaces Roads

Hostels
Offices Eateries

Learning
center
Courts
Residence
Auditorium
& Library

Case tips
• Clearly drawing the structure and mentioning the segmentations makes it easy to visualize.
• This approach can be extrapolated to malls/other campuses/townships etc. 30
3
 CCTV Cameras in XLRI
Guesstimate Bain

I: Let’s consider all CCTVs cameras in the XLRI Jamshedpur campus. C: I would like to estimate the number with the following data and
assumptions. We would thus have around 464 CCTV cameras in the
C: I would like to segment the usage of cameras into indoor and outdoor.
Jamshedpur campus.
Indoor would include all buildings that have cameras and outdoors would
include the roads, eateries, courts etc. .

I: How would you like to segment the indoor usage of cameras?

C: I would like to segment the indoor buildings based on the primary users.
People residing on campus can be segmented into three groups – students,
faculty & admin. The students primarily use hostels, learning centers, library
and the auditorium. The faculty use their offices and their residences. The
administration uses their offices.

I: That seems like a fair approach. What kind of numbers are we looking at?

Place Cameras installed

Hostels, IC & Exceed 2 cameras/floor * 6 floors * 3 buildings + 2 cameras/floor * 4 floors * 3 buildings = 72 cameras

Learning centre 2 buildings * 6 classes/floor * 3 floors * 2 cameras/class = 96 cameras

Faculty offices & Administration offices 240 cabins * 1 camera/3 cabins = 80 cameras

Faculty residence 100 houses * 1 camera/ house = 100 cameras

Auditorium & Library ~ 15 cameras

Eateries & shops 8 places * 2 cameras/place = 16 cameras

Sports facilities 10 courts * 2 cameras/court = 20 cameras

Roads 200000/150 = 45 + 20 (miscellaneous) = 65 cameras

Total Cameras = ~464 cameras

30
4
 Calcium Substitute Market
Guesstimate McKinsey & Co.

Case Statement:
Based on your internship project, Let's do a small guesstimate to evaluate the size and valuation of calcium substitute market in India.

Background information

• Form of Market Size

Substitute: Tablets
• Both off the counter and
prescribed medicines
1-25 years 26-50 years 51-75 years >76 years

Off the Off the

Prescription Prescription
counter counter

Lifestyle Lifestyle Pregnant

Issues Issues Women

Age Age Aging Process

Characteristic Characteristic

Case tips
• It is important to think out loud and bring in as many different aspects to numbers allocated to each bucket.
30
5
 Guesstimate Calcium Substitute Market McKinsey & Co.

I: : Based on your internship project, Let's do a small guesstimate to evaluate C: I believe the largest customer segment for us are the people of the age
the size and valuation of calcium substitute market in India. group 26-50 years. Somewhere around 30% of people might consume it as a
regular substitute. This age group includes majorly majority of pregnant
C: : I would like to reiterate to ensure that we are on the same page. I am to
women who require calcium substitutes in the first trimester of pregnancy.
come with a structure to estimate the market size followed by market valuation
of calcium substitutes in Indian Pharmaceutical market.. I: Are there any other factors that you think should be makes this age bracket
attractive.
I: That is correct.
C:: I have observed that people usually develop bone weakness during this
C:: So, calcium substitutes come in various forms including swallowable
time because of the work-oriented lifestyle. I think here 50% of the people in
tablets, chewable tablets, capsules etc. Do you want me to focus on any
this age bracket consume calcium substitutes because of a medical condition .
specific form.
I: For this scenario consider the tablet forms of the substitute only.
I: For this scenario consider the tablet forms of the substitute only.
C: Proportion of people of age 51-75 years will also require calcium
C: So, I will start off with sizing the market for calcium substitutes in India. I
substitutes as the aging process sets in rapidly and weakening bones are one
would first like to divide the population of India in terms of 4 distinct age
the first symptoms majority of these adults face at the onset. I think 70% of
brackets i.e., 1-25 years, 26-50 years, 51-75 years and above 76 years of age.
this segment might consume calcium substitutes here a majority of which
The percentage of population that I would attribute to each of these age
might be compelled by a medical condition.
brackets is 30%, 30%, 20% and 20%. Does this approach seem appropriate in
your opinion.
I: Okay, go ahead.
I: Yes, go ahead. How will you segment the population further. C:: Lastly people 76 years and above will require calcium substitutes but I
would allocate 55% of people here consume calcium substitutes. The reason
C: I would further divide people in each bracket into a segment who consume
for a lower number to this age bracket is because these people usually are
calcium substitutes because of a specific medical condition (prescription) and
deficient of many other nutrients as well. Hence their consumptions of multi
regular consumption (off the counter). The former is to probably take the
vitamins and externally infused vitamins will be much greater.
calcium substitutes as prescribed tablets by a medical professional while the
latter might use them as off the counter medicines. I: : This seems like a reasonable approach. How would you triangulate the
estimated size you have calculated.
I: That’s fair go ahead.
C:A good way to triangulate this data based on actual data gathered from the
C: To start with, the age bracket of 1-25 years who use for regular condition
sales data of calcium substitutes on various e-commerce platforms that home-
would be 20% as this age bracket is characterized by low levels of nutritious
deliver medicines. Also, on the ground medical representatives and pharmacy
food consumption, decreased outdoor activity combined with rapid body
retailers could provide us concrete data that can be extrapolated to the entire
growth. For the people consuming calcium substitutes because of a medical
population.
condition would be around 5% as this age is associated with good health.
I: : That's a good approach. We can close here. Thank you. 30
I: : Alright, go ahead
6
 Guesstimate Pet Food Market in India McKinsey & Co.

Case Statement:
Estimate the market size of pet food market in India.

Market Size

Rural (70%) Urban (30%)

Low Income Medium High Income

(30%) Income (60%) (10%)

Non- Non- Non-

Current Prospective Current Prospective Current Prospective
Interested Interested Interested
Owners (5%) Owners (5%) Owners (30%) Owners (30%) Owners (40%) Owners (40%)
Owners (90%) Owners (30%) Owners (20%)

Dog-Cat Dog-Cat Dog-Cat Dog-Cat Dog-Cat Dog-Cat

owners owners owners owners owners owners

Other Other Other Other Other Other

Domestic Pet Domestic Pet Domestic Pet Domestic Pet Domestic Pet Domestic Pet
Owners Owners Owners Owners Owners Owners

Case tips
• The number of different aspects one can bring in would help bring out a more comprehensive structure.
• Analyzing both the need and want aspect of owning a pet (dog or cat) will help get a perspective on the numbers.
• It is important to think out loud and bring in as many different aspects to numbers allocated to each bucket. 30
7
 Guesstimate Pet Food Market in India McKinsey & Co.

I: Estimate the market size of pet food market in India. C: In the medium income segment, I think the current and prospective owners
will be more, with the later taking precedence because of increased inclination
C: Are we looking at a particular category of pets here? towards having pets and awareness of plights of street animals. This segment
I: You can consider only dogs and cats here. too will have a greater number of people in the dog-cat owner segment as their
disposable income will allow them to invest in the training, upkeep,
C: The pet food market consists of both affordable and premium brands. vaccinations, and affordable food for the pets.
Would you like me to focus on a particular brand segment.
I: Okay, how will you triangulate the numbers you are allotting to the
I: You can consider pet food brand that operates at an average price range of segments based on this rationale.
Rs. 50 – Rs. 60 per kg.
C: To triangulate this data with real hard facts, we can approach local pet
C: Okay so I will be considering only the affordable brand segment. stores and analyse their sales data to cross check our estimations. We can also
look at records with animal welfare organizations like PETA, on their animal
I:Okay go ahead.
adoption records.
C: So first I would like to consider the population of India and divide them into
I: That's great, go ahead.
urban and rural dwellers. Majority of the people in the rural areas mostly do
not own cats and dogs or even if they do usually left-over household food is C: Lastly the high-income segment might have least number of people in the
served to them. Hence, they do not present a substantial market for us. Would non-interested segments and greater proportion in the other two branches.
it be fair assumption to ignore that portion. However, with greater disposable income these people might be inclined
towards buying more exotic pets. Also, the dog and cat owners of this segment
I: That's a fair assumption, go ahead.
might go for premium pet food rather than the affordable one.
C: In the urban segment I would like to divide the population into 3 segments:
I: Great! We can close now. Thank you.
Low income, medium income, and high-income segment. I would further
divide each of these segments into the Current Owners, Prospective Owners,
and Non-Interested Owners. The former two segments can be divided into
Dog-Cat owners and other domestic pet owners.

I: Okay, go ahead and give me the rationale on how you will allocate
population percentages to each segment.

C: In the low-income segment, because of less disposable income majority of

the population will be in the non-interested owner segment. In the current and
prospective owner segment, the percentage of people will be higher in dog and
cat pet owners as most other pets in market will require more resources from
the owner's end.

I: Fair, go ahead. 30
8
 Guesstimate Petrol Pumps in India BCG

Case Statement:
Estimate the number of petrol pumps in India. Also tell me if their number should be increased.

Background Information
• Place: Food wastage in Mumbai
India is assumed using a Area = 500 sq km
demand side approach Population = 20 million
• Product: Food wastage in
domestic household
Case tips
• Guesstimates like this one Medium Traffic area = 200 sq
High Traffic area = 200 sq km Low Traffic are = 100 sq km
km
are a combination of a 1.5 pumps per sq km = 300 0.5 pumps per sq km = 50
0.75 pumps per sq km = 150
small guesstimate and a pump pumps
pumps
small case.
• The interviewer will nudge
you to spend less time on
the numbers and focus on 500 pumps per 20 million
people
the second part of the
25 pumps per million people
problem statement, which
is essentially a case.
• Country level numbers can
be hard to calculate
sometimes. In such cases, Urban population in India Rural Population in India
finding the prevalence per 70%*1400 million*25
million population for a 30%*1400 million*25
pumps pumps
smaller unit like a city or
state can be a good =10500 pumps =24500 pumps
approach.
• For the second part of the Urban (10500) + Rural (24500) = 35,000 petrol pumps in India
case, consider using
proxies to estimate growth 30
9
 Guesstimate Petrol Pumps in India BCG

I: Please calculate the number of petrol pumps in India. Also tell me if their C: Number of petrol pumps should be increased if either the current demand is
number should be increased. not met or if the future demand is growing faster than the growth in petrol
C: Interesting. I can estimate the number of petrol pumps either by taking the pump capacities.
number of vehicles or the population density as a proxy. Which one should I I: How do you know the current demand is not met?
go with?
C: We can find that using metrics like the average wait time for vehicles and
I: Whichever you like. compare with historic and seasonal baselines to see if areas around particular
C: I'll take the population route. Can I estimate the petrol pump for one city petrol pumps are underserved.
and then try to scale it up to India level? I: If the wait time is increasing, would you make a new petrol pump?
I: Sure, go ahead. C: Not always. Increased wait time would mean a demand-capacity mismatch.
C: I have lived in Mumbai, so I’ll start there. Mumbai is about 500 sq km. We Depending on other factors, this can be addressed by increasing more pumps
can divide Mumbai into 3 zones, high traffic, medium and low traffic. We can or lanes in existing petrol pumps or building new ones depending on traffic
assign them 200, 200 and 100 km sq respectively. Does this figure sound etc.
okay? I: Okay. Tell me about how would you predict future petrol demand?
I: Yes. C: We can start with first finding the types of customers and their cumulative
C: If high traffic zone would have 1.5 petrol pumps per sq km, 0.75 petrol volume consumption of different petroleum products.
pump in medium and 0.5 petrol pumps in low traffic zone, the Mumbai would I: For now, let's just consider private cars.
have around 500 petrol pumps.
C: Understood. We can use the sales of new and used cars to predict the
I: Please keep going. demand of petrol since petrol consumption would lag with car sales. We can
C: Mumbai's number could be good representative of petrol pump density for also use construction of new roads and highways.
urban populations which is 500 pumps per 20 million people = 25 per million. I: What if we don't have the data for car sales. How would you predict car
For rural population, given the lower familial income, the number of private future car sales?
cars would be less, but they would have high energy consuming farming
vehicles like tractors. Would it be a good ideas to assume rural demand to be C: We can use the data on input materials in car production as a proxy. Maybe
75% of urban demand? make a composite metric of steel, rubber and other inputs and see how closely
the change their production mimicked car sales. This can give us some insights
I: You can take them to be equal. on future car sales.
C: Understood. This makes the calculation easy, 1400 million*25 petrol I: Alright. We can end the case here. Thank you for the discussion
pumps/million = 35,000 petrol pumps in India, approximately.

I: Sounds reasonable. Should this number increase? If yes, what metrics would 31
you use? 0
 Guesstimate Food Wastage in India Accenture

Case Statement:
Estimate the food wastage in India in a year – target only household wastage

Background Information
• Place: Food wastage in Households
India is assumed using a
demand side approach
• Product: Food wastage in
domestic household Medium Income
Low Income households High Income households
households (20%)
Case tips (30%)
• The important question to
ask before beginning this
case is whether we need
to estimate the food Low Income households
(10%) Food wastage percentage for
wastage at a household different income levels:
level, restaurants level or Data provided by the interviewer
at a production level in a when nudged for it.
field to fork model
• The food wastage is Below Poverty Line (5%)
Food Consumed per person =
directly proportional to Mean of 1.0 kg/day is assumed
the income level of the from the person’s prior work ex
in food tech Industry
household since a high- Household categories
income household would
HIH MIH LIH BPL Total wastage for a year
most likely waste a higher (only households) =
proportion of the food as No. of households 20%*23cr 40%*23 cr 30%*23 cr 10%*23 cr
2417.7+3358.0+1359.9+201.4 =
compared to a low- No. of 7337.08 cr kg
4 5 6 6
income household. people/household
• Supply side approach Food consumed per
would be a better 1.2*365 1.0*365 0.9*365 0.8*365
person (in kgs)
approach if we are going
Food Wastage (%) 30% 20% 10% 5%
for the production level
Total wastage 2417.7 cr kg 3358.0 cr kg 1359.9 cr kg 201.48 cr kg 311
wastage.
 Guesstimate Used Commercial Vehicle market in USA Bain

Statement:
Estimate the number of used car market in the US – specifically used cars in ride sharing apps

Background Information
• Place: Used car market in Cars
US forms a major chunk
of the secondary car
market. Commercial
vehicles Private Vehicles
• Product: Secondhand car
sales in USA

Case tips Light vehicles Heavy Vehicles

• The important question to

ask before beginning this
guesstimate is whether we Ride Sharing Trucks
need to estimate used cars
in the private sector or in
the commercial sector.
• We need to figure out the Government
Buses
Vehicles
further divisions in the
commercial sector since it
doesn’t conform to the Location
standard divisions Life of a car = 3 years in
commonly used in the Rural Urban
rural and 5 years in urban
other examples Total population 300 miliion setting
• Demand side approach %age of population 30% 70% Total number of used cars in
would not work since a population : - 140k + 580k
commercial markets need Ride sharing adoption %age 5% 20% = 720k
to be dealt with at a
individual use case level. Life of cars (in years) 5 3
Total Used cars 140k 580k 312
 Guesstimate Guesstimate the no. of students completing XIIth McKinsey & Co.

Case Statement:
Guesstimate the total number of students completing XIIth Standard in Delhi every year.
.
Population
2 Crores
Background information
0-5 years 6-18 years 19-40 years 19-40 years 60+ years
• Students from both 10% 30% 30% 30% 10%
Public and Private 20L 60L 60L 60L 20L
schools would be
included BPL APL
25% 75%
• The number of students 15L 45L
passing need to be
calculated and not Attend School Don’t Attend School Attend School Don’t Attend School
50% 50% 100% 0%
attending 7.5L 7.5L 45L 0

Total Attending
School
52.5L

Ist-XIth Standard XIIth Standard

92% 8%
~48L ~4.5L

Pass Fail
90% 10%
~4L ~0.5L
Alternate Approach (Supply side)
No. of students = # Metro X # Localities per X # Schools in X # Students in X Proportion of X Pass %
completing XIIth Standard stations metro station 1 locality 1 school XII Std. students

Case tips
• % of students attending school would be different across income categories
• Include multiple filters in-order-to make it more comprehensive
31
• •Basic BQs ( tell me about yourself, why consulting and why McKinsey & Co.)
3
 Total Plates in Delhi
Guesstimate Bain

Case Statement:
How many plates are there in Delhi?

Dinner plates in Delhi

Background Information
(Population= 20 Mn)
• Type: Dinner plates
made of glass, steel,
china etc. Households Shops/
factories Offices Restaurants
• Used in: Homes, (20/5=5 Mn)
restaurants, shops,
factories, and BPL
Supply side
office/school cafeterias (20%, 1 Mn)
No of restaurants* No
1*1=1 Mn
of tables*No of
Low Income plates/ table
(30%, 1.5 Mn)
1.6*1.5=9 Mn
Demand side
Mid Income Total no. of customers
(40%, 2 Mn) at peak hour* No. of
2*6*2=24 Mn plates/ customer
High Income
(10%, 0.5 Mn)
3*6*0.5=9 Mn

Case tips
• Approach is more important than the final answer
• Calculations under pressure are tested 31
• Give logic behind the assumptions taken 4
 Total Plates in Delhi
Guesstimate Bain

I: Lets only consider houses and restaurants for now. I: But what metric would a restaurant use to decide the number of plates it
would need? If you look at the weekly demand of the restaurant, is it the same
C: I’ll look at houses first. I would like to look at the income segments because
across the whole week?
people with more purchasing power will buy different dinner sets for different
purposes-for example, steel or plastic for everyday use and glass or ceramic C: The demand is concentrated at the end of the week because most people
for special occasions. Also, I will take the size of a dinner set as 6. Based on like to go out on weekends. So, the number of plates required to service
these assumptions, total dinner plates used by households is 43 Mn. customers on weekends would help the restaurant decide the number of plates.

For restaurants, we can look at the number of restaurants in Delhi and the I: Ok, good. Assume 100 people visit a restaurant on the weekend.
number of plates used in each restaurant, assuming all restaurants are uniform.
C: We will observe a variation across the day, more people will go to a
I: What would the number of plates used in a restaurant depend upon? restaurant in the evening. Let’s say 60% of the people go in the evening from 7
pm to 11 pm for dinner and assuming that a person spends 1 hour in the
C: The capacity of the restaurant, i.e., the number of tables and then number of
restaurant and a uniform distribution, there are 60/4= 12 people in a restaurant
plates required for each table. So, assume that the average restaurant has 20
at the peak hour. Therefore, the restaurant needs 12*(1+1+2) = 48 plates. They
tables and they usually provide one plate for the main course and one smaller
must also keep extras in case of breakage and damage, taking a 20% buffer
plate for the starters. Apart from this there are plates for the dishes served too,
that comes to approximately 58 plates.
say two on average.

31
5
Table of Contents
The “START BIG” approach & The “START SMALL” approach ....................................................... 2
Estimate the market for bicycles India. ................................................................................................... 3
Estimate the litres of cola sold in Delhi per day ..................................................................................... 3
Estimate the number of denims sold in India in a year. .......................................................................... 4
Estimate the amount of petrol used in Delhi per day .............................................................................. 4
Estimate the market size of paint industry in India ................................................................................. 5
Estimate the amount of water that can fill an aircraft ............................................................................. 6
Estimate the number of red lights in Delhi ............................................................................................. 6
Estimate the number of leather balls sold in India in a year ................................................................... 7
Estimate the number of light bulbs used in St. Stephens in a year. ........................................................ 8
Estimate the number of one rupee coins that would fit in a normal size room ....................................... 9
Estimate the number of vehicles crossing the road in front of FMS. ...................................................... 9
Estimate the number of number of subscribers of Economist in India ................................................. 10
Estimate the number of cars sold by Maruti in India. ........................................................................... 10
Estimate the number of beers sold in Goa every day. ........................................................................... 11
Estimate the market for air conditioners in Bangalore in 1 year........................................................... 11
Additional Examples............................................................................................................................. 12
Useful Data Sources.............................................................................................................................. 13

1
The “START BIG” approach & The “START SMALL” approach

1. How many gallons of white house paint are sold in the U.S. every year?

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

2
Estimate the market for bicycles India.

 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 buy 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

Answer: Market size of cycles bought in India is ~Rs. 20M

Estimate the litres of cola sold in Delhi per day

 Population of Delhi is approximately 15M people.

 Assume 30% of population is below poverty line, does not consume cola.
 Population above poverty line is approximately 10M people. Cola drinking population
is typically in age group 5-40 years of age. Approximately 60% of Delhi’s population
is in this range, i.e ~6M people.
 Dividing this population into High consumers (20%), Moderate consumers (40%) and
Low consumers (40%). Assume high consumers consume approximately 1L per day,
moderate consumers consume 0.5L per day and low consumers consume 0.1L per
day.
 Thus total daily consumption = 1.2*1 + 2.4*0.5 + 2.4*0.1 = 1.2 + 1.2 + 0.24 = ~2.6M
litres per day.

Answer: ~2.6M litres of cola is consumed in a day in Delhi

3
Estimate the number of denims sold in India in a year.

 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 atleast single denim in a year and atleast 50%
women buy single denim.
 Let’s divide the population further into high income, middle income and low income
groups. Assue 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 buy 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

Estimate the amount of petrol used in Delhi per day

 Population of Delhi is approximately 15M people.

 Assume family size of 5, there are approximately 3M households in Delhi.
Approximately top 40% households would own a car. Thus there are approximately
1.2M cars in Delhi.
 We can divide these car users into Heavy, Moderate and Light users. Let’s assume
heavy users are 20% of total car owners and travel approximately 80km a day, to and
fro work in Delhi-NCR etc. Assume Moderates are 50% of total car owners and travel
~30 km per day. Assume Light users are 30% of population and travel 10km a day
(including low usage multiple cars owned by families etc.)
 Thus total distance traversed by Delhites in a day = 1.2*(0.3*80 + 0.5*30 + 0.3*10) =
1.2*70 = 84M km.
 Assume average mileage of a petrol car is 12km/l. Thus, amount of fuel consumption
in a day = 7M litres

Answer: ~7M litres of petrol is consumed in a day in Delhi

4
Estimate the market size of paint industry in India

Working assumption – The household sector contributes to approximately 70% of the

paint industry in terms of volume of paint used. The industrial sector accounts for the
remaining 30%. We will first calculate the contribution of the HH sector and use this as a
proxy for industrial segment.

 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 a 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

Answer: Market size of paint in India is approximately Rupees 1600 crores or 1.6B

5
Estimate the amount of water that can fill an aircraft

Working assumption – This estimation is done taking into consideration Boeing 747, the
most common airplane in India. Also, the plane is considered to be empty from inside, i.e.
there are no seats, baggage cabins installed.

 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

Answer: ~1000cu. m of water can fit in a plane

Estimate the number of red lights in Delhi

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

6
Estimate the number of leather balls sold in India in a year

Working assumption – Number of leather balls primarily implies leather cricket balls.

 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 are bought by this segment.
o 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 years.
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:
o Adding 5% of this number to account for balls used in national/international
cricket matches = 1,10,000
o Also, adding 2% for balls bought by those with income below $2 per day and
for girl cricket enthusiasts = 44,000
o Total = 22,00,000 + 1,10,000 + 44,000 = ~ 23,50,000

Answer: ~23,50,000 balls are bought every year in India

7
Estimate the number of light bulbs used in St. Stephens in a year.

Working assumption – Required to calculate the number of incandescent/CFL bulbs

replaced in a year but not the number of tube lights replaced in a year.
 St. Stephens has approximately 20 classrooms in the main wing of the building and
approximately 10 classrooms in the science block. Since each classroom has 4 tube
lights and 2 incandescent bulbs, we can calculate the total number of bulbs fixed in
classrooms.
o Number of incandescent light bulbs in classrooms = 2* 30 = 60
 The mess has 8 long tables and each table has 4 bulbs hanging above it. Tube lights
line the walls of the mess. Further, the kitchen area inside the mess has approximately
5 light bulbs.
o Number of incandescent bulbs in the mess = 4 * 8 + 5 = 37
 The Stephens café has 2 sections; each section has approximately 6 incandescent light
bulbs. Further, the cop store inside the café has 2 bulbs and the café kitchen has 3
incandescent light bulbs.
o Number of incandescent bulbs in the café = 6*2 + 2 + 3 = 17
 The administration block has close to 10 small rooms with 2 bulbs each and a main
reception area with 4 bulbs.
o Number of incandescent bulbs in the admin block = 10*2 + 4 = 24
 The main seminar hall and stage are largely lit by tube lights. There are approximately
4 bulbs backstage.
 The academic block has 10 washrooms (men + women) and each has 3 incandescent
light bulbs.
o Number of bulbs in washrooms = 3*10 = 30
 The corridors, pathways and parking area of Stephens are lit by incandescent bulbs.
There are approximately 30 bulbs that light these corridors and pathways.
 Stephens has 6 residence blocks, with 30 rooms in each block. Each room has one
light bulb and one tube light. Further, the washrooms in the residence blocks have
only tube lights.
o Students in residence prefer to buy CFL bulbs. I have seen that one in every
two students has a CFL bulb instead of an incandescent bulb.
o Number of total bulbs in residence = 6*30 = 180
o Number of incandescent bulbs = 90
o Number of CFL bulbs = 90
 Total number of incandescent bulbs = 60 + 37 + 17 + 24 + 30 + 30 + 90 =288 or ~300
 Total number of CFL bulbs in use in Stephens = 90
 Let’s assume the life of an incandescent bulb at an average is 2 months and the life of
a CFL bulb is 6 months
o The total number of incandescent bulbs used in a year = 12/2 * 300 = 1800
o The total number of CFL bulbs = 12/6 * 90 = 180
o Total number of bulbs used in a year = 1980
Answer: 1980 bulbs are used in St Stephens per year

8
Estimate the number of one rupee coins that would fit in a normal size room

Working assumption – The room is empty, i.e bare, without any furniture/person/object.

 Let’s assume the dimensions of a normal size room to be as follows,

o Height = 3m; Length = 4m, Breadth = 5m
o 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

Estimate the number of vehicles crossing the road in front of FMS.

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 FMS in the day: 1200 approx.

9
Estimate the number of number of subscribers of Economist in India

 Population of India: 120 crores

 Economist target market is urban upper middle class, educated
 Percentage of population in target: 20% (24 crores)
 Percentage of people interested in economics, intelligent articles etc.: 10% (2.4
crores)
 Percentage of people who prefer reading hard copy magazines: 20% (50 lakhs
approx.)
 People choosing economist among similar magazines (time, Newsweek
etc): 40% (20 lakh)
 Hence estimated subscriptions will be less as there may be more than 1 person
using a subscription
 Total number of subscriptions: 15 lakhs
 Add subscriptions by libraries, colleges, corporates etc: 1 lakh appox

Answer: Total subscriptions approx. 16 lakhs

Estimate the number of cars sold by Maruti in India.

 Population of India= 120 crores

 % of population willing to purchase cars= 30% of 120 crores= 36 crores
 % of population looking for cars in higher range( above 7 lakhs)= 10% of 36 crores=
3.6 crores
 Popular brands in this category =5
 Maruti cars sold in higher range (assuming equal sale of all cars= 3.6/5=720000 cars
 Life of a car= 10 years
 Cars in higher range sold annually = 72000 cars
 % of population looking for cars in lower range( below 7 lakhs)= 20% of 36 crores=
7.2 crores
 Popular brands in this category = 8
 Maruti cars sold in higher range (assuming equal sale of all cars= 7.2/8=900000 cars
 Life of a car= 15 years
 Cars in higher range sold annually = 60,000 cars

Answer: Total Maruti cars sold in India in a year= 60,000 + 72,000= 1,32,000

10
Estimate the number of beers sold in Goa every day.

 Population of Goa: 15 lakh (approx.)

 Percentage of population drinking per day: 20% ( 3 lakhs)
 Average amount of beer consumed per person: 3 pints
 Total beer consumed by local: 9 lakh pints
 Tourist population on an annual basis:20 lakh (approx.)
 Tourist population on a day (assuming tourist season is on, December first
week):50,000
 Percentage of tourist population drinking: 90% (45000)
 Per capita consumption: 5 pints
 Total consumption by tourists: 2.25 lakhs pints

Answer: Total in a day during season = 23 lakh pints approx.

Estimate the market for air conditioners in Bangalore in 1 year.

 Population of Bangalore: 80 lakhs approx.

 Two types of ACs sold: Domestic use and Commercial use.
 For the purpose of this guesstimate we will assume only domestic ac sales.
 ACs can only be afforded by middle class and above
 Percentage of population falling in above category: 40% (approx.) = 32lakhs
 Average size of household: 4 Number of households: 8 lakhs
 Bangalore has pleasant weather throughout the year hence people who can avoid
spending the money will do so.
 Also fresh grads and students form a very large percentage of the population.
They will also avoid buying ACs even though they fall in the target demographic.
 Number of households thinking of buying AC = 60% of target = 5 lakhs approx.
 Average age of an AC = 5 years
 Average number of ACs in a household of 4 = 3
 Number of target ACs = 15 Lakh

Answer: The number of sales of air conditioners in Bangalore in a year = 15/5 = 3 lakh.

11
Additional Examples

1. Estimate the number of working ATMs in NCR?

2. How many Maggi packets are used in the FMS canteen in a week?
3. Estimate weekly revenues of the FMS canteen (Assume normal week and no seasonal
variations)
4. Estimate the number of males who get a haircut in Delhi in 1 day?
5. How many people in India will be in flight at any given point of the day?
6. What is the number of flights leaving the Mumbai terminal each day?
7. What is the market size of disposable diapers in India?
8. Estimate the total number of chairs in the Delhi University North Campus
9. What is the total number of cars sold in Delhi per year?
10. Estimate total rail-line network in India
11. Estimate the number of vehicles crossing the road in front of FMS.
12. Estimate the number of popcorns sold in PVR Saket everyday.
13. Estimate the market for Pepsi in India.
14. Estimate the number of number of subscribers of Economist in India.
15. Estimate the market for Iphone4 in India
16. Estimate the number of Macbooks sold in India per day.

Find a lot more interesting guesstimate Sexamples at:

http://www.consultingcase101.com/tag/estimate-or-guesstimate/page/5/

12
Useful Data Sources

http://www.census2011.co.in/states.php.

http://www.indiademographics.com/default.aspx

13
 3. Cases: Guesstimates

Vashi’s Note: Guesstimates teach you the basics of structuring mathematically, finding
the correct bottlenecks, and are the stepping stone to solving cases. Sometimes, cases also
include guesstimates that test your quant ability with general knowledge, so mastering
them is essential.
Mohika’s Note: It is not necessary that the interviewee reaches the exact same number
that is derived here, or uses the same approach. It should, however, be of the same order.

3.1 Case 1 (Cricket Countdown, A&M)

Question: Can you tell me the number of cricket balls used in the Inter IIT sports
meet?

Preliminary Questions

1. Do we consider only balls used during the matches or even the balls used in practice?
Just the match balls.

2. What kind of matches are being played? Are these one-day matches, T-20, or test?
T-20 matches.

3. What is the format of the tournament?

Each team will play against each other, and the top 4 teams will go into the semifinals,
which will lead to the final.

4. How many IITs are participating? How many teams are participating from each IIT?
Take ’N’ IITs to be participating, with each having one team.
Note: In case the interviewee is stuck, or the case has to be made simpler, here N
can be replaced with 23.
 18 Chapter 3. Cases: Guesstimates

5. How often are the balls changed? Are they changed after every few overs?
No, one ball is used in each innings of the match.

Case Solving Begins!

Okay, let’s break the problem down mathematically. Do we also consider lost balls
during the matches?(illustrates structure)

Totals No. of Balls

Used

Total Number of No. of Balls used

Matches per match

N
C2 3 (2 Semi + 1 Final) 2 (1 per Innings) Extra Balls

Look at Historic Data

Quality of Balls Used
Dimensions of the Ground
Surroundings of the Ground

Yes, consider them.

Can you list down a few ways you can find the number of balls lost?

Sure. (lists them down) We can look at the historical data if it is available with the
client. Apart from that, we can look at the quality of the ball and determine its dura-
bility in a given match. We can look at the stadium’s dimensions and surroundings
to determine the probability of finding the ball if it goes out. All these factors will
determine the average number of extra balls used in a match.

Note: Since the structure here is depicted for N, there is no final number that
can be given. The interviewer should ensure that the interviewee can present a
formula for it.

3.2 Case 2 (Pharmacy Potential, Bain)

Question: Can you calculate the market size of the E-Pharmacy Market?

Preliminary Questions

1. What do we mean by market size here? Are we talking in terms of the number of
people, revenue...?
Total revenue per year in Rupees.
3.2 Case 2 (Pharmacy Potential, Bain) 19

2. And we are considering only medicines? Are we not considering, say, other medical
supplies they might be selling?
Yes, consider only medicines.
Case Solving Begins!

Okay, let’s break the problem down. First estimating the Market Size in terms of
people and then further how chronic and non-chronic users might spend. (illustrates
structure)

Market Size

Total Yearly Revenue

Chronic Users Non-Chronic Users

(35%) (65%)

Tablets/day (3) 4 times yearly

Cost of Strip (150) Sick for 4 days

No of tablets in a strip (10) Requiring 4 tablets/day

No of Users Cost of Strip (100)

No of tablets in a strip (10)

No of Users

Population

Urban Rural
(30%) (70%)

Low Income Middle Income High Income Low Income Middle Income High Income
(40%) (50%) (10%) (40%) (50%) (10%)

Internet Internet Internet Internet Internet Internet

(80%) (90%) (100%) (70%) (80%) (100%)

Probability Probability Probability Probability Probability Probability

(0%) (10%) (30%) (0%) (0%) (10%)

The final amount seems to be around 6000 Cr Rupees

Great! That seems right.
Note: The probability here is the joint probability of logistics and availability and is
estimated from the interviewee’s general knowledge.
 20 Chapter 3. Cases: Guesstimates

3.3 Case 3 (Soda Summation, ICG*)

Question: Estimate the amount of Soda consumed in IITK on a normal day.

Preliminary Questions

1. Do we consider only the soda consumed on that day or the stored soda as well?
Consider only the amount consumed.

2. Are we considering the entire campus population, including the guards and the
workers?
No, consider only the students.

3. What soda products are we talking about? Are we considering club soda and baking
soda too?
No, just the soft drinks.

4. What kind of a day and month is it?

It is a regular academic day in the month of April.

5. In what unit do we want to estimate the consumption? Is it in litres, rupees, or the

number of soft drink bottles?
Calculate in liters.

Case Solving Begins!

Okay, let’s break the problem down mathematically.(illustrates structure)

Students

Undergraduate Master's Students Ph.D Students

(5000) (2000) (2000)

Low Volume Medium Volume High Volume Low Volume High Volume
(0.125L) (0.25L) (0.5L) (0.125L) (0.25L)

55% 30% 15% 60% 40%

Low Volume Medium Volume High Volume

(0.125L) (0.25L) (0.5L)

50% 40% 10%

 3.4 Case 4 (Tea Tally, ICG*) 21

The total soda consumption comes out to be around 1900L.

Good job! We can end the case here.

3.4 Case 4 (Tea Tally, ICG*)

Question: How many tea cups are sold at a railway station?

Preliminary Questions

1. Tea cup sounds a lot like porcelain tea sets. We mean the tea sold, right?
Yes, the tea.

2. When we say railway station, what do we include? Tea can be sold on the platform,
outside the railway station, and inside the train.
Consider only the platform.

3. Do we consider selling tea to everyone on the platform?

Yes.

4. How many platforms are at the railway station?

There are seven platforms. Platform 1 has an entry exit. The other six are between
the railway lines.

5. So platforms 2-3, 4-5, and 6-7 are joined?

Yes, they are.

6. What is the length of the platform?

We are talking about CNB here. The minimum length for a passenger platform, as
per Indian Standards, is 180 meters. You can take some ideas from that.
Okay. So, I am taking the effective length as 200 meters for platform 1 and 150m for
platforms 2-3, 4-5, and 6-7. Here, by the effective length, I mean where the crowd is
present for a station like CNB. That sounds good. Go ahead.

7. Is there any difference between the frequency of trains arriving during the day and
those arriving at night?
Why don’t you try to analyze that?
I would assume that we have a higher frequency of passenger trains during the day.
You can go ahead with that assumption.

8. What is the size of the cup?

Take it to be 50ml.

9. Are there any seasonal considerations?

Say it is the fall season, and the month is September.

Case Solving Begins!

Okay, let’s break the problem down mathematically. (illustrates structure)

 22 Chapter 3. Cases: Guesstimates

Cups Of Tea

Effective Length is 200m Other Platforms Effective Length is 150m

1 tea shop every 20m
Platform 1 1 tea shop every 30m
(3)

10 PM - 8 AM 8 AM - 5 PM 5 PM - 10 PM 8 AM - 10 PM 10 PM - 8 AM
(0.25 cups/min) (1 cup/min) (0.5 cups/min) (0.5 cups/min) (0.1 cups/min)

8400 Cups/Day 7200 Cups/Day

My estimate is 15,600 cups daily.

Excellent, that’s the end.

3.5 Case 5 (Energy Estimate, Dalberg)

Question: Estimate the energy and resource consumption at an African Refugee
Camp.

Preliminary Questions

1. What type of energy are we referring to?

Why don’t you list some down for me?
There can be electricity, water, gas, food, etc.
Let’s consider electricity, water, and gas.

2. What does the refugee camp look like?

So the refugee camp has ten blocks. Each block can be considered to be a five-room
house with one kitchen. Each room can accommodate two people and has one tube
light and one fan. The kitchen has one refrigerator and one microwave. Consider the
washrooms to be outside the camp and out of consideration here.

3. What is the current occupancy of the camp? Also, are there any other common
facilities in the camp that we need to consider?
80% of the camp is occupied currently. You can ignore energy consumption for the
common facilities.

4. Where is the refugee camp located? What time of the year is it?
It is located in Spain. It is the summer season.

5. What does an average day look like for a refugee?

The refugees currently do not have a work permit or access to leave the compound.
So, they stay in the camp all day and cook three meals a day.

6. In what unit do we want to estimate the consumption? For what duration do we want
to estimate?
 3.6 Case 6 (Newspaper Network, ICG*) 23

Consider the electricity consumption in kWh and the water and gas consumption in
liters. Estimate the consumption in a day.

Case Solving Begins!

Okay, let’s break the problem down mathematically.(illustrates structure)

Energy Consumption in
a single block

Electricity Water Gas

Drinking (2L/day x 10 People)

Cooking (10 L/day)

5 Rooms Kitchen Washing (10 L/meal x 3 Meals/day)

Cleaning (20 L/day)

Gas used
No. of Meals
Light Fan Refrigerator Microwave per Meal
(20W) (30W) (250W) (1000W)
A gas cylinder having 30L gas
3 Meals/day x
lasts 540 meals, so for a single
10 People
10 Times x meal, 1/18L of gas is used
6 Hours/Day 20 Hours/Day 24 Hours/Day
30 Secs/Day

Note: The structure shows the energy consumption for just one block. This should
be multiplied by the number of blocks and the percentage occupancy of the camp to
reach to the final number.

The daily electricity consumption comes out to be around 78 KWh. The water
consumption comes out to be 640L, and that for gas is 13.33L/day .
I think we can end the case here.

3.6 Case 6 (Newspaper Network, ICG*)

Question: Estimate the revenue of the newspaper company Business Times.

Interviewer’s Note: Not all information in this case is revealed directly. Specific questions
are asked even if the information seems obvious to be given already. This is a practice that
can be followed in a few other cases as well.

Preliminary Questions

1. What are the different revenue streams?

There are two major streams, the offline and the online division.
24 Chapter 3. Cases: Guesstimates

2. What is the time frame of our estimate?

Let’s consider a day.

3. Are we a national paper? For what geography do we want to estimate the revenue?
We are a national English newspaper, but we are only interested in the revenues from
the Kanpur division.

4. So what audience are we talking about in the online and offline divisions? What is
our revenue model?
For the offline mode, consider households, and our revenue source for this is through
advertisements and newspaper charges. Our online revenue model is a subscription-
based model with INR 100 per month pricing.

5. What is the price of our paper? Do we have any info on the competitors and their
pricing?
Our price is INR 10 per newspaper. There are two major competitors, A and B. Their
prices are INR 9 and 9.5, respectively. You can assume that they are similar to us.

6. What is our advertisement revenue model?

So our newspaper has 20 pages, a maximum of 15% of the space is occupied by
national advertisements and another 15% is occupied by local advertisements at
maximum.
Since we are focusing on the revenue from Kanpur, can we ignore the revenue from
the national ads?
Yes, you can.
So, can you detail the local ads revenue model in more detail?
We earn 20,000 per one full page of ads. There can be any number of ads on that
page. On Mondays and Tuesdays, 60% of the space dedicated to local ads is covered.
This number is 80% on Wednesdays and Thursdays, and 100% of it is covered on
the remaining days.

Case Solving Begins!

Okay, let’s break the problem down mathematically.(illustrates structure)

Do we have a break-up of the Kanpur customers for the online and offline segments?
30% of the population prefers reading the online paper, and the rest prefers the offline
mode.

The daily revenue from the Kanpur region comes out to be 16.4L INR.
Good job! This is the end.

Note: The advertisement model is not shown in the structure here. If the inter-
viewee mentions the model in their structure, then ensure that the information is
mentioned accurately, as this can be a little confusing.
 3.7 Case 7 (Cash Capital, McKinsey) 25

Sources of Revenue

Advertisements Offline (21L) Online (9L)

Households (4.2L)
0-18 Years 18-30 Years 30+ Years
(20%) (20%) (60%)

Low Income Middle Income High Income

(40%) (50%) (10%) Probability of Probability of
buying (30%) buying (15%)

English English English

(20%) (40%) (80%)

Affordability Affordability Affordability

(40%) (80%) (100%)

3.7 Case 7 (Cash Capital, McKinsey)

Question: What is the average ATM cashflow in IITK?

Preliminary Questions

1. Do we want to calculate it for all the ATMs?

No, consider only one ATM.
So, what we can do is estimate the total cashflow on campus and divide it by the
number of ATMs.
Good, you can go ahead with that.

2. In what units and for what time frame to want to estimate?

Calculate in INR and for a time frame of a month.

3. What time of the year are we talking about?

Assume a normal working semester.

Case Solving Begins!

Okay, let’s break the problem down mathematically.(illustrates structure)

So, the number of ATMs comes out to be 8.

That seems fair. Go ahead.

The cash flow for an ATM in a month comes out to be roughly 9.45L INR.
Great work! We can end the case here.
26 Chapter 3. Cases: Guesstimates

Population

Professors Workers Students

(1000) (2000) (9000)

Frequency Amount
50% use ATM
(5 times/month) (3000) UG Masters Ph.D
(5000) (2000) (2000)

Frequency Amount
(4 times/month) (1000)

Frequency Amount
(1 time/month) (1000)

Frequency
(2 times/month)

Amount
(1000)
Frequency Amount
(2 times/month) (1500)

Campus Population Average Frequency

Number of ATMS 8 hours/day 5 mins/withdrawal 8 ATMs
(12000) (2 times/month)
 Part H – Practice Guesstimates

Main Index

© The Consulting Club, FMS Delhi 2023-24 83

 Main Index Practice Guesstimates: Table of Contents
S.No. Item Page # S.No. Item Page #
1. Gurgaon Delhi Toll plaza 85 15. Daily Revenue of an Airport 103
2. Revenues for TOI 86 16. Automobile Tire Market Size 104
3. Smart watches in India 87 17. Wine Consumption in India 105
4. Number of Smokers 88 18. Users of Bisleri Water Bottles in Delhi 106-107
5. Number of Cheese Burst Pizzas 89 19. Sanitizer Demand in Delhi 108
6. Petrol Pumps in India 90-92 20. BCom Admissions in Delhi University 109
7. TT balls in Delhi 93-94 21. Credit Cards Issued 110
8. White Shirts in Delhi 95 22. Revenue of Dream11 111
9. Schools in Delhi 96 23. Revenue of Pickpocket 112
10. Daily Departing Flights 97 24. Number of Movie Screens 113
11. Tractors in India 98-99 25. Petrol Consumption (in litres) 114
12. Market Size of EV in India 100 26. Swiggy Drivers 115
13. Revenue of Flat Screen TVs 101 27. Delhi Metro Passengers 116
14. Daily Amazon Orders 102 28. Number of Toothbrushes 117

© The Consulting Club, FMS Delhi 2023-24 84

 Guesstimate Index | Main Index Toll Plaza
Estimate daily revenues of Delhi-Gurgaon toll plaza
Assumptions Methodology
• Toll plaza operates 24x7 • Daily revenue = (Daily revenue per toll booth) x (# toll booths)
• 2/3 wheelers do not have to pay toll fare • For one toll booth, the # vehicles crossing it will vary with time of the day (peak / non-peak hours)
• Toll fare is the only revenue source • Average time for a car to pass the booth = 15 secs; Hence, 4 cars can cross the booth in a minute
• Traffic across booths is uniform • Therefore, capacity = max. vehicles that can cross a booth in an hour = 240 vehicles / hour

Traffic Distribution Revenue per toll booth We are treating the toll booth as a bottleneck
to estimate traffic throughout the day
1
Capacity

# vehicle per booth Fare per vehicle

0.5 Noon

Midnight
0
Time
High traffic hours Medium traffic hours Low traffic hours 4 wheelers > 4 wheelers (trucks)

‘High traffic hours’ include

8 hours /day 8 hours /day 8 hours /day 70% of daily traffic 30% of daily traffic
morning and evening office Full capacity 60 % capacity 30 % capacity Fare – INR 30 Fare – INR 70
peak hours. ‘Medium traffic
hours’ include afternoon Total # vehicles per booth = (8)x(240)x(1+0.6+0.3) ~ 3600 Weighted Fare = (0.7)x(30) + (0.3)x(70) = INR 42
and late night hours. ‘Low
traffic hours’ include hours
from ~11 PM to 7 AM.
Daily revenue = (3600)x(42)x(20) ~ INR 30 lacs

© The Consulting Club, FMS Delhi 2023-24 85

 Guesstimate Index | Main Index TOI Revenues
Estimate daily revenues of Times of India
Assumptions Methodology

• Households order 1 copy of ToI each • Revenue Streams = Revenue from newspaper circulation + Advertising Revenue
• No online subscription model for ToI • Newspapers can be bought by individuals/households or organizations (libraries, offices, schools)
• Print ads are priced only on area basis • Advertising Revenue Streams = Printed Ads on the paper + Ads on website/mobile apps
• Digital ads are priced on CTR basis • Digital revenue is calculated based on the number of online users who click the advertisement

Daily Revenue We have assumed that the online ad revenue

model is based on #clicks. It can also be based
on #impressions or #conversions.

Subscriptions Advertisements

Newspaper penetration
can be assumed based on #Buyers Price Print Digital
literacy rate and income Print Revenue
levels (#Pages)*(Area per page)*(% area for
ads) *(Price per unit area)
Households Organizations = 25*(60*40)*25%*5000
Ad Area Price per unit Area = Rs. 75 million
Digital Revenue
#Household Buyers = (#Households)*(English Newspaper penetration)*(ToI’s Mkt. Share) (#Users)*(Number of ads)*(Click Through Rate)*(Cost per click) =
#Households Buyers = 300 mil * (50%*20%)*25% = 7.5 million 25 mil *10*0.5%*10= Rs. 12.5 million
#Organizational Buyers = 50% of Household buyers = 3.75 million
#Revenue = 7.5 + 3.75 = 11.25 * 5 = Rs. 56.25 million Total Revenue = 56.25 + 75 + 12.5 = Rs. 144 million

© The Consulting Club, FMS Delhi 2023-24 86

 Guesstimate Index | Main Index
Smart Watches
Estimate the market size of Smart Watches in India

Assumptions Methodology

• Average price of Smart Watch is Rs. 10k (considering • Population of India = 130 Cr; Urban = 30%; Rural = 70%
watches range from 3k-30k) • Age Group classification: 0-25 = 50%; 25-50= 30%; 50+= 20%
• Penetration in rural area will be much lower than in • Income Spread: BPL = 20%; Low Income= 40%; Middle Income= 30%; High Income= 10%
urban area • In Urban India 1% of 0-25 will own one, 10% of 25-50 will own one and roughly 0.5% of 50+ will own one in the middle-income population while in
• Market exists only in middle & high income groups the higher income population, the penetration would be roughly double. In Rural India, 0.5% will own one in age group 0-25, 1% will own one in 25-50
• Income Spread is uniform in Rural & Urban and 0.1% will own one in 50+

Market Size of Smart Watches

Rural 0.70*130cr = 91cr Urban 0.30*130cr = 39cr

Income Spread Middle Income Middle

0.30*91cr = 27.3cr High Income 0.10*91cr = 9.1cr 0.30*39cr = 11.7cr High Income 0.10*39cr = 3.9cr
Income

Age Spread 0-25 25-50 50+ 0-25 25-50 50+ 0-25 25-50 50+ 0-25 25-50 50+
0.50*27.3cr = 0.30*27.3cr 0.20*27.3cr 0.50*9.1cr = 0.30*9.1cr 0.20*9.1cr 0.50*11.7cr 0.30*11.7cr 0.20*11.7cr 0.50*3.9cr 0.30*3.9cr 0.20*3.9cr
13.65cr = 8.19cr = 5.46cr 4.55cr = 2.73cr = 1.82cr = 5.85cr = 3.51cr = 2.34cr = 2.95cr = 1.17cr = 0.78cr

0.005*13.65cr 0.01*8.19cr 0.001*5.46cr 0.005*4.55cr 0.01*2.73cr 0.001*1.82cr 0.01*5.85cr 0.1*3.51cr = 0.005*2.34cr = 0.02*2.85cr = 0.2*1.17cr = 0.01*0.78cr =
Penetration
= .0683cr = 0.0819cr = 0.0055cr = 0.0228cr = 0.0273cr = 0.0018cr = 0.0585cr 0.351cr 0.0117cr 0.057cr 0.234cr 0.078cr

Market Size of smart watches = 0.998 cr * Rs 10000 = 9980 cr ~ 10,000cr

© The Consulting Club, FMS Delhi 2023-24 87

 Guesstimate Index | Main Index Smokers in India
Estimate the number of Smokers in India
Assumptions Methodology
• Smokers can be divided into 4 categories- Chain smoker,
• Total smokers= Smokers in urban area + Smokers in rural area
Regular smoker, smoking with alcohol and occasional smoker
• Urban and Rural areas to be divided into 3 classes each- High income , Middle income & low income
• Cigarettes can be filtered and non-filtered
• Every class will have different proportion of male & female smoking cigarettes
• Population having filtered cigarettes to be estimated
• Probability of smoking will depend on the respective age group
• Sex Ratio – 1:1
• Low income group will not consume filtered cigarettes
Total Population
Total Smokers=
Rural (60%) Urban (40%) = 68.35mn

High Income (15%) Mid Income (35%) Low Income (50%) High Income (30%) Mid Income (50%) Low Income (20%)

(Non-filtered) (Non-filtered) (Non-filtered)

Rural High Urban High Urban Mid
Income Group Male Female Income Group Male Female Income Group Male Female
Age Group % Prob Total % Prob Total Age Group % Prob Total % Prob Total Age Group % Prob Total % Prob Total

0-18 10% - - 10% - - 0-18 10% 0.05 0.5% 10% - - 0-18 10% - - 10% - -

18-30 15% 0.3 4.5% 15% 0.15 2.25% 18-30 15% 0.5 7.5% 15% 0.25 3.75% 18-30 15% 0.3 4.5% 15% 0.15 2.25%

2.25 30-50 15% 0.2 3% 15% 0.05 0.75%

30-50 15% 0.15 15% 0.05 0.75% 30-50 15% 0.3 4.5% 15% 0.15 2.25%
%

>50 10% - - 10% - - >50 10% 0.05 0.5% 10% - - >50 10% - - 10% - -

© The Consulting Club, FMS Delhi 2023-24 88

 Guesstimate Index | Main Index Cheese Burst Pizzas
Estimate the number of cheese bursts pizzas sold by Dominos daily
Assumptions Methodology
• Working hours from 11AM to 11 PM • Daily orders from one outlet= 0.4*3*[(Peak hrs/20mins)*0.8+(Non peak hrs/20)*0.5]+50% of calculated figure for online and
• Peak hrs - 2-4PM & 7-10PM(occupancy rate-80%), Non home delivery orders]
Peak- 7 hrs(occupancy rate-50%) • No. of Cheese bursts can be calculated for one city i.e. Delhi and then extrapolated to arrive at a national figure
• An outlet is able to serve in a 4 km radius on average. • Area of Delhi= 1600 sq.km. and one outlet serves 4 km radius
• Services available- Dine in or Home Delivery • Therefore no. of outlets in Delhi= 1600/3.14*4*4= 32
• 40% of Pizza orders are Cheese burst orders. • Consumption of pizzas in urban and suburban areas = 50% of that of metropolitan areas
• 1 Order has approximately 1.5 pizzas.
• On avg. it takes 5min to receive & 15 to process = 20mins Number of Cheese Bursts
• 3 orders can be processed in parallel. per outlet
• Delhi constitutes 10% of total metropolitan area
• No Dominos outlets in rural areas.
We’ll divide country in 3 categories: Non Peak (7 hours) Peak (5 hours)
• Metropolitans(20%)
• Urban and Suburban areas(30%)
• Rural Areas(50%)
In Store Purchase Online & Home Delivery In Store Purchase Online & Home Delivery

Orders = Online orders= Orders = Online orders=

3*(420/20)*0.5 = ~ 32 0.5*32= 16 3*(300/20)*0.8= 36 0.5*36= 18

Cheese bursts orders = Pizzas per orders = All over Delhi = stores *
Total Cheese 0.4*102 = ~ 41 41*1.5 = 61 per store= 32*61= 1952
bursts sold daily
= 34160 Since Delhi’s area = 10% of total metropolitan area, no. of Cheese bursts sold daily in urban and suburban areas=
cheese bursts sold in metropolitans= 10* 1952 = 19520 0.5*(19520)*1.5 = 14640

© The Consulting Club, FMS Delhi 2023-24 89

 Guesstimate Index | Main Index Petrol Pumps
Estimate the number of Petrol Pumps in Delhi (Approach 1)
Assumptions Methodology
• High Traffic Zone = 30%; Medium Traffic Zone = 40%; • Area of Delhi = 1600 sq. km
Low Traffic Zone = 30% • Area served by 1 petrol pump in:
• Distance between petrol pumps in: • High traffic Zone = 2x2 sq. km = 4 sq. km
• High traffic zone = 2km; Medium Traffic Zone = 3km; • Medium traffic Zone = 3x3 sq. km = 9 sq. km
Low Traffic Zone = 4km • Low traffic Zone = 4x4 sq. km = 16 sq. km

No. of petrol pumps

in Delhi

High Traffic Zone Medium Traffic Zone Low Traffic Zone

Area # of petrol pumps Area # of petrol pumps Area # of petrol pumps

0.30*1600 = 480 sq km / 4 0.40*1600 = 640 sq km / 9 0.30*1600 = 480 sq km / 16

480 sq km sq km = 120 640 sq km sq km = 71 480 sq km sq km = 30

Total No. of petrol pumps = 120 + 71 + 30 = 221

© The Consulting Club, FMS Delhi 2023-24 90

 Guesstimate Index | Main Index Petrol Pumps
Estimate the number of Petrol Pumps in India (Approach 2)
Assumptions Methodology
• Avg. distance travelled by user does not vary across • We use a demand side approach for this
vehicles and = 30 km/day • Number of Petrol Pumps = Demand of Petrol / Petrol supplied by a pump (avg)
• Non-peak time hours of a petrol pump can be • Demand of Petrol = Distance travelled by a vehicle user (avg) * ( Number of bike users /Bike mileage + Number of
surmised in a single number = 50% in this case Car users / Car mileage)
• Pooling usage internalized in distance assumption • Petrol Supplied by a pump = Overall Petrol Suppling Capacity * Weighted Average of Usage ratio

Number of Petrol Pumps

Demand of Petrol Petrol Pump usage ratio Petrol Pump Capacity

2 wheeler 4 wheeler Peak Time Non-Peak Time Avg time to fill 15 l petrol = 3 min
Avg booths = 4
Empirical 25 crore users 15 crore users 4 hours 10 hours Capacity = 15*4*(4*60/3 = 16800 l
Assumptions 70 km/l mileage 10 km/l mileage 100% 50%

Demand of Petrol = [Avg distance = 30 km] * Petrol Supplied per pump = [Avg Usage Ratio = (1*4+0.5*10)/14 = 0.64]*[Capacity =
[25cr/70 km/l + 15cr/10 km/l] = 55.7 crore litre 16,800 l] = 10,800 litre

Number of Petrol Pumps = [Demand = 55.7 Cr ltr] / [Per Pump Supply = 10,800 ltr] = 51,587 = 51,500 (approx.)

© The Consulting Club, FMS Delhi 2023-24 91

 Guesstimate Index | Main Index Petrol Pumps
Estimate the number of Petrol Pumps in India (Approach 3)
Assumptions Methodology

• Petrol pump operated 12 hours a day • # of pumps in India = (# of cars in India) / (number of cars serviced by a petrol pump) * (frequency of
• Uniform petrol consumption by a car visits to a petrol pump by a car)
• Arrival rate uniform across peak hours • For a petrol pump , peak hours ( 7 am to 11 am ) and ( 4 pm to 8 pm) = 8 hours
• Arrival rate different across non peak hours • Car arrival rate in peak hours = 4 cars in 5 minutes
• Considering only 4 wheelers in the analysis • Car arrival rate in non-peak hours = 4 cars in 10 minutes
• Assuming average size of a household = 5 members. • This implies total cars serviced daily = 8*(4/5)*60 + 4*(4/10)*60= 384+96= 480 cars serviced a day.
• Total population of India = 1.25 Billion • Assuming an average car needs petrol twice in a month

• 50% households = 0 vehicles 0.7 vehicles Assuming commercial vehicles form 20% of
Total number of cars in India = +
• 30% households = 1 vehicle per the total private vehicles = 35 million
250*0.7 = 175 million private vehicles.
• 20% households = 2 vehicles household commercial vehicles

# of cars in India Total 210 million vehicles in India . Assuming each vehicle needs fuel twice in a
month we get = (2/30)*210= 14 million vehicles using fuel daily
Total households in
Average number of
India = ( 1250/5) = × cars in each household Number of petrol pumps = (# of cars 14million/480= 30000 petrol stations in
250 million visiting a petrol pump) / (# of cars India . Asuming each station has 4 pumps
serviced by a petrol pump = 120000 pumps

© The Consulting Club, FMS Delhi 2023-24 92

 Guesstimate Index | Main Index TT Balls (Part 1)
Estimate the number of TT balls used in a day in Delhi
Assumptions Methodology
• Considering only matches in commercial spaces for calculation • Solved from demand side
• Using a factor of 20% for private spaces • #Matches per hour = 1/(Time per match + Idle time)
• Considering all matches to be similar (for avg. life of a ball) • Time per match = 14 minutes (for simplicity; assuming 11-point match)
• Avg. Life of a ball = 5 matches

# of TT balls
G1 – Guesstimate the number of TT balls
used in a day in Delhi 14208
G2 – Estimate the number of TT matches
played everyday
G3 – Estimate the number of TT racquets # of matches/day # of balls/match
used in a lifetime by an international player
& many more 71040 1/5

Commercial Private Average life of ball

14800 x 4 = 59200 59200 x 0.2 = 11840 5 matches
Basic Thoughts:-
Approaches – Demand side and Supply
# of table hours Matches/ hour
Side (try both)
Be ready with enough information about 14800* 60/15 = 4
the manufacturing aspects of the sport or
make assumptions
Time/match Idle time

*Refer to the next Slide for calculations 14 mins 1 min Cont. on next slide:-

© The Consulting Club, FMS Delhi 2023-24 93

 Guesstimate Index | Main Index TT Balls (Part 2)
Estimate the number of TT balls used in a day in Delhi
# of Table Hours Available
14800

Large Spaces Medium Spaces Small Spaces

100*60 = 6000 4800 4000

# of Spaces Hours per Space # of Spaces Hours per Space # of Spaces Hours per Space
200*1/2 = 100 60 200*1 = 200 24 200*2 = 400 10

# Regions - 200 # Tables - 5 # Regions - 200 # Tables - 3 # Regions - 200 # Tables - 2

# of Spaces – 1/2 # Hours Open - 12 # of Spaces – 1 # Hours Open - 8 # of Spaces – 2 # Hours Open - 5
Methodology:- Methodology:- Methodology:-
Need proxy for #regions – Use metro stations # Tables per space=> Total number of table hours available = Hours available
Total metro stations = 200 For Large spaces = 5 tables in :-
# commercial spaces per region=> For Medium spaces = 3 tables Large Spaces + Medium Spaces + Small Spaces
For Large spaces = 1 space per 2 regions For small spaces = 2 tables
For Medium spaces = 1 space per 1 region For any space:-
For small spaces = 2 spaces per 1 regions Hours available = Number of spaces * Hours per space
Number of spaces = (Number of regions * number of
# Hours available per space => spaces per region)
For Large spaces = 12 hours Hours per space = (Number of tables * number of hours
For Medium spaces = 8 hours open)
For small spaces = 5 hours
(Assuming full capacity utilization) Total number of TT balls used in a day = 14800 * 4 * 1.2 * (1/5) = 14208 ~14200 TT Balls
© The Consulting Club, FMS Delhi 2023-24 94
 Guesstimate Index | Main Index White Shirts in Delhi
Estimate the number of people wearing a white shirt (WS) in Delhi on any particular day.
Assumptions Methodology
• Calculations are done for a working weekday i.e. assuming • Population segmentation has been done across various age groups.
offices, schools and various institutions are open • Calculation for each age group is done on the basis of observed/experienced characteristics.
• One person is wearing one shirt a day. • Various percentages are used based either on memorised or experienced statistics.
• No considering School going children and their shirts.
• Population of Delhi = 20 million Population (Age
Segmentation)

25-59: 60+ :
30% = 8 million 20% = 2 million

Student 5% = 0.4 M Employed (Formal): Employed (Informal): Unemployed: Females : Males :

20%= 1.6 M 70%= 5.6 M 5% = 0.4 M 50% = 1M 50% = 1 M
Non Professional
Assuming don’t wear
Courses 95%
20% wear 10% wear WS: 0
Males 60% Female 40%
10% wear WS. ~ white shirt white shirt
0.038 M Employed/Self Retired:
40% wear WS. ~ 20% wear WS ~ Employed:20%= 0.2 M 80%= 0.8 M
1.12 million 0.04 million
Professional 0.384 M 0.128 M
Courses 5% 50% wear white shirts 5% wear white shirts
~ 0.1 M ~ 0.04 M
30% wear WS
~ 0.006 M
Total number of people wearing white shirts in Delhi =5.76 million= 6 million (approximately)

© The Consulting Club, FMS Delhi 2023-24 95

 Guesstimate Index | Main Index Delhi Schools
Estimate the number of Schools in Delhi
Total Number of Schools = School going children in Delhi / Avg. children in one school

Part 1: Estimating no. of school going children in Delhi Part 2: Avg. children in one school

We know that the population of Delhi is around 20 million. Since India is a young country, we can assume 50%
Schools
is under 25. So population under 25 would be 10 million. Let’s equally divide the population under 25 across
every year, which would be equal to 400,000.
Small schools Medium Schools Large Schools
Population (40%) (40%) (20%)

Middle & Small Medium Large

Kindergarten & Senior Secondary Age 18+ (out
Age 0-3 Secondary School Class Size/Section 20/3 30/4 40/6
Primary School School (Class 11- of school)
(Class 6-10) -
(Class 1-5) - 2.8m 12) - 0.8m Standards Up till 10 Up till 12 Up till 12
2.0m
Avg Student in one school = (20 * 3 * 10) * 0.4 + (30 * 4 *12) * 0.4 +
% Going to School (40 * 6 * 12) * 0.2
= 600 * 0.4 + 1440 * 0.4 + 2880 * 0.2
Middle Class & Above (40%) 100% 100% 100%
= 240 + 576 + 576 = 1392
Lower Class (40%) 90% 75% 50%
BPL (20%) 50% 25% 0% Total Number of Schools in Delhi = 43,88,000 / 1392
~ 4,400,000 / 1400 = 3142 ~3150
Total: (1.0 * 2.8 + 1.0 * 2.0 + 1.0 * 0.8) * 0.4 + (0.9 * 2.8 + 0.75 * 2.0 + 0.5 * 0.8) * 0.4 + (0.5 * 2.8 + 0.25 * 2.0 + 0) *
0.2 = 4.388 To triangulate this number, you can take a 5km sq. area, estimate
Hence estimated school going children = 4.388 million the number of schools and extend it for the entire city

© The Consulting Club, FMS Delhi 2023-24 96

 Guesstimate Index | Main Index Daily Departing Flights
Estimate number of flights departing from Delhi Airport in a Day
Assumptions Methodology
• Airport doesn’t operate at max. utilisation • Total departing flights (total out-bound flights) in a day = Total In-Bound Flights in a day = Total Flights Operating from an airport
• Turnaround time –
• No of flights operating from one hangar = (24hrs/avg. turnaround time) [Turnaround time is the time a flight would stay in a hangar]
• Domestic flight – 1.5hr
• No of flights operating from one terminal = 2 x (No of flights operating from one hangar)
• International Flight – 3hrs
• There is a possibility that a flight landing in Delhi, might not depart that day itself. But there would also be flight which depart on a day that did
• Terminal at IGI –30 (Domestic), 40 (Intl.)
not land on that particular day. Hence it should have a balancing effect.
• Every terminal has 2 hangars

No. of domestic terminals 30 Max. Domestic

Domestic Flights Flights =
No. of domestic flights operating from one 32 x 30 = 960
terminal 2 x (24/1.5) = 32
Maximum Operating Flights
No. of international terminals 40 Max. Intl. Flights
International Flights =
No. of international flights operating from one 16 x 40 = 640
terminal 2 x (24/3) = 16
Utilisation rates have been
considered to be different for To triangulate the number, we can use the alternate approach from the
different hours for domestic Hours of the day demand side, by figuring out how many people would be departing from
Delhi on one day divided by the avg. number of people in one flight
flights only. For international
flights, uniform rusk of 50%
has been assumed
6:00am – 9:00 am 9:00am – 6:00 pm 6:00pm – 9:00 pm 9:00pm – 12:00 am 12:00am – 6:00 am
Domestic Flight Rush 100% 60% 100% 60% 20%
Avg. Utilisation for Domestic Flights = (6 x 100 + 12 x 60 + 6 x 20) = 60%

Total flight departures = Operating Domestic Flights + Operating Intl. Flights = (0.6 x 960) + (0.5 x 640) = 896 flights

© The Consulting Club, FMS Delhi 2023-24 97

 Guesstimate Index | Main Index Tractors in India (Part 1)
Estimate the number of tractors in India
Assumptions Methodology
• Tractors are only being used for agricultural purposes in the primary sector • The population of India = 130cr.; Rural= 70% & Urban= 30%; Avg.
• There is uniform supply of tractors around the country household size= 5
• All the tractors in consideration are functional and others have been disposed off • Avg. life of tractor is 5 years (approx.)
effectively. • Methodology = [# of tractors in rural areas + # of tractors in urban areas
• Each farm-owner has only one tractor for her land/ one tractor per farming (10% of figure from rural areas)] / (avg. life)
household • Applying 80/20 rule to estimate figures that are not commonly known

Population of India Rural (70%)

Urban (30%) Rural (70%) Secondary Sector (20%) Primary Sector (80%)

10% of rural area figure Other primary occupations (20%) Farming (80%)

Farm Owners (20%) Wage Earners (80%)

To filter down to target
population, See the Financially capable of owning tractor & tractable Poor Finances and/or non-
need, accessibility, farming area (20%) tractable farm area (80%)
affordability, preference
for tractors
Limited/No access to markets Access to markets
(20%) (80%)
Cont. on next slide:-

© The Consulting Club, FMS Delhi 2023-24 98

 Guesstimate Index | Main Index Tractors in India (Part 2)
Estimate the number of tractors in India

Limited/No access to Access to markets (80%)

markets (20%)

Willing to buy (80%) Unwilling to buy (20%)

Final Calculation
Number of tractors = [# of tractors in rural areas + # of tractors in urban areas (10% of figure from rural areas)] / (avg. life)
= [(# of tractors in rural area) * 1.1] / 5
= [(26 cr. *0.7*0.8*0.8*0.2*0.2*0.8*0.8 ) * 1.1] / 5
= 6.56 lac.

Area of India = 32.9 lac sq. km Triangulation & Sanity Check (Area based approach) Alternate Approach
= 32.9 cr. hectares
Number of = [Area of India * %Arable land * %Tractable area]/[Tractor usage per day * days of harvest]
Arable land ~ 50% tractors
Tractable area ~ 40% of arable area = [32.9 * 107 * 0.5 * 0.4] / [5 * 21]

Average usage of tractor ~ 5 hectares

= 6.27 lac.
daily
Average 3 weeks of harvest per season

© The Consulting Club, FMS Delhi 2023-24 99

 Guesstimate Index | Main Index EV Market Size
Estimate the market size of EV in India
Assumptions Methodology
• Considering market for only 4 wheeler passenger vehicles. • Population of 5 major tier 1 cities in India – Approx. 90 million.
• Only available in tier 1 cities. • Average household size – 4.5 member/family.
• Premium Car Segment – Price > Rs. 10,00,000. • Poor wont choose a costly EV.
• Average years a car is used by rich people – 5 years. • In middle class, 50% families will have a car. Among them, 80% will buy normal cars, and 20% premium cars.
• Average years a car is used by middle class – 7 years. • In elite class 100% population will have a car, all will be buying premium cars.
• EV is currently in introduction stage in India, so not • Innovators/Environmentalist among rich - 10%, who will adopt to EV. Innovators/Environmentalist among middle class- 5%,
considering second hand EV. who will adopt to EV.
• All families in rich sector own a car, 50% middle class
families own a car.
Population – 90 million
• Demographic Divide: Uber Rich = 2%, Middle Class = Tier 1 Cities #families ~ 20 million
38%, Poor = 60%.
Population = 2%
Income Spread #families =
Population = 60%
400,000
and product #families = 12 million Poor Middle Class Uber Rich
lifecycle factor Car buyers ~ 100%
Car buyers = 0
All Premium
Population = 38%
Lifecycle ~ 5 years
#families = 7.6 million
#families ~ 440,000 Car buyers ~ 50%
Average Price = Rs. 6,00,000 Normal Cars Lifecycle ~ 7 years Normal Cars
Penetration Penetration - 5%
Factors/Acceptability #families ~ 80,000
#families ~ 110,000
Factor Average Price =
Average Price = Rs. 22,00,000 Premium Cars Premium Cars
Rs. 22,00,000
Penetration – 5%
Penetration – 10%

Market Size of electric vehicles = # of normal EV*Avg. cost of non-premium EV + # of premium EV*avg. cost of premium EV
= 22,000*5,00,000 + 13,500*22,00,000 ~ Rs. 40 billion
© The Consulting Club, FMS Delhi 2023-24 100
 Guesstimate Index | Main Index Flat Screen Televisions
Estimate the revenue of flat screen televisions sold in Australia in the past 12 months
Assumptions Methodology
• Population: 25 million people • No of households = Total Population/Average size of household
• Size of average household : 3 people • No of TVs per household is considered by taking an average of all the households (Across all categories)
• Average life of TVs = 4 years • Calculating the average price of 1 TV, assuming price under each category
• Average number of TVs / household = 1 • Here we are not considering reused TVs; only fresh purchases

Revenues

Total No. of
Price of 1 TV
TVs

Premium Medium Low End No. of Households No. of TV/Household

25 million
Share and 20% of Total 60% of Total 20% of Total
3 member/family
1 TV per
Cost Cost = $1000 Cost = $600 Cost = $200 household/4 years
~ 8 million

Average Price of 1 TV = (0.2*1000+0.6*600+0.2*200) Total no of TVs sold = 8 / 4= 2 million

= $600

Total Revenues = Average Price of 1 TV * Total no of TVs sold = $1.2 Billion

© The Consulting Club, FMS Delhi 2023-24 101

 Guesstimate Index | Main Index Amazon India
Guesstimate the number of daily orders of Amazon India
Assumptions Methodology
• No of mobile users in India – 800 million. • Population of India – 1.3 billion. No of mobile users – 800 million.
• No. of internet users – 70% of mobile users. • Frequency of buying varies across age groups and income.
• Urban Internet Users – 60% • Considering just the sales of Amazon India, not its grocery or any other chain.
• Rural Internet Users – 40% • Amazon has around 35% market share in India e-commerce industry. Flipkart – 45%, rest others.
• Users belong to Age group 10-55. • Considering 50% of the population is aged below 25 and uniform distribution of population across ages.
• Rural low income group’s buying pattern is negligible. • Considering life expectancy of 65 years.

# of internet users – 560 million

Location Basis 40% Rural Urban 60%

60% 30% 10% 40% 40% 20%

Low Income Medium Income High Income Low Income Medium Income High Income

Income Group Rural – Medium Income Rural – High Income Urban – Low Income Urban – Medium Income Urban – High Income
Age 10-25 25-40 40-55 10-25 25-40 40-55 10-25 25-40 40-55 10-25 25-40 40-55 10-25 25-40 40-55
Orders/Yr 2 2 0 4 6 3 2 4 0 4 6 3 9 12 6
Total Orders/day 1.1L 69K 0 73K 69K 35K 2.2L 2.7L 35K 4.4L 4.2L 2L 5L 4.1L 2L

Summing all order values gives the total orders delivered in a day around 30 lacs. Out of this around 35% belongs to Amazon,
that would give a figure of around 10 lakh packages per day.
© The Consulting Club, FMS Delhi 2023-24 102
 Guesstimate Index | Main Index Daily revenue of Airport
Estimate the daily revenue of an airport
Assumptions Methodology
• 1. Considering only the most substantial revenue streams. • Advertisement Revenue= revenue from billboards & experience areas.
2. Assuming that shops pay only a fixed rent. • Area of billboards * price/m2. Assuming Price/m2 is 200 for a day. Assuming 6 experience areas and 75k/day as charge.
3. Number of flights per hour is a function of the kind of rush. • Airline charges: Fixed charges (domestic) : 50k, Fixed charges (international) : 100k.
4. Assume that the primary revenue source from advertisements are • Average shop charges = 4000/day; parking charges = 100.
billboards. • No of airstrips – 3, Low rush – 1 flight per strip ; Medium – 2 flights per strip High – 4 flights per strip
5. No. of Flights per hour is a function of the no. of airstrips.
Revenues of an airport

Area of
Airlines Shops Parking Advertisement
billboards

Fixed Fees Per Landing revenues # of shops x Rent/shop # of vehicles x Fees/vehicle Experience Areas Billboards
250*4000 = 10 lacs Taking 6 experience areas

Domestic International
# of people / people/vehicle 3
Domestic International Type Small Medium Large

= # airlines* fees/airline Rush Hour Low Medium High Number 500 250 50
# of flights x people/flight
Hours 6 6 12 Size (sq. ft) 2*2 3*5 8*10
= 10 * 50000 = 15*100000 = 396*200/3*100
Flights/hr 6 12 24 Total 2000 3750 4000
= 5 lacs = 15 lacs = 2.64 lacs (Assuming 10%
Flights/da 36 72 288 vehicles are parked) =9750*200 = 19.5 lacs
Flights Domestic Int. y
= 4.5 + 19.5 =24 lacs
# flights 297 99

Flights/hr 2500 7500 Summing all values gives the total revenue as 20 + 14.85 + 2.64 + 24
Total 7.425 7.425
= Rs. 61.5 lacs/day.

© The Consulting Club, FMS Delhi 2023-24 103

 Guesstimate Index | Main Index Automobile Tire Market
What is the size of automobile tire market in India in 2020?
Assumptions Methodology
• By size we mean the number of automobile tires • Total number of 4 wheelers = Number of cars owned by rural + urban households
• Consider only 4 wheeler passenger vehicles • Urban and Rural areas to be divided into 3 classes each- High income , Middle income & low income
• Average household size is 6 for rural and 4 for urban • Every class will have different proportion of families owning the car
• Lower income segment don’t own cars and Urban high incomes • Total Number of 4-wheelers(N) owned can be divided in New (N/12) and Old (11N/12)
own 2 cars per family, • New cars will have 5 new tires and old cars will have 5 old tires replaced over 5 years
• Average life of 12 years for cars and 5 years for tires
• Each new car has 5 tires (4 operational + 1 spare)
Total Population (1.4 bn)

Rural (70%) Urban (30%)

High Income (15%) Mid Income (35%) Low Income (50%) High Income (30%) Mid Income (50%) Low Income (20%)

#Households​ =1.4*0.7*0.15/6 = 24mn​ =1.4*0.7*0.35/6 = 57mn​ =1.4*0.7*0.5/6 = 81mn​ =1.4*0.3*0.3/4 = 31mn​ =1.4*0.3*0.5/4 = 52mn​ =1.4*0.3*0.5/4 = 52mn​

Penetration of 4-
80%​ 10%​ 0%​ 95%​ 60%​ 0%​
wheelers​
# 4-wheelers
1​ 1​ 0​ 2​ 1​ 0​
per household​

New 4-wheelers ~ 10 mn Number of tires per year = 10*5 = 50

Total # Tires
Total #4-wheelers = 115 mn ~ 155mn
Old 4-wheelers ~ 105 mn Number of tires per year = 105*5/5 = 105

© The Consulting Club, FMS Delhi 2023-24 104

 Guesstimate Index | Main Index Wine Consumption
Estimate the number of bottles of wine consumed in India in a week.
Assumptions Methodology
• Rural, older and urban-lower income group people don’t consume
• Demand side approach is used
wine. • Number of bottles of wine consumed (drinking) = Number of glass of wine consumed per week/ Number of
• Wine in India is consumed for both drinking and cooking purposes. glasses in an average wine bottle
• Penetration of wine drinkers in India is 40% of the total alcohol • Number of glass of wine consumer (per week) = Total wine drinkers * #glasses of wine consumed by an average
drinkers. drinker per week
• Wine consumed for cooking purposes would be 5% of the total wine • Number of bottles of wine consumed = Total bottles of wine consumed for drinking + cooking purposes
consumed for drinking purpose.
Total Population (1.4 bn)
Rural people
won’t drink wine Rural (70%) Urban (30%)
Children (0-19 yrs) Older population
#People who drink =1.4*0.3*0.7*(0.4*0.7+0.2*0.9) (20%) (65yrs+) (10%)
alcohol (Urban area) =13mn
Working Population (19-
65 yrs) (70%)
% of wine drinkers in India 40%
Lower Income Middle Income High Income
Total wine drinkers in India 13mn * 0.4 = ~5mn (40%) (40%) (20%)
#Wine Glasses consumed by
2 glasses/week #Bottles of wine 70% drink alcohol 90% drink alcohol
an average wine drinker #Bottles of wine
consumed for
consumed for
#Glass of wine consumed per cooking purposes =
~10mn drinking purpose = Total #Bottle of wine consumed
week in India 5%*1.25mn =
1.25mn in India in a week
#Glasses in an average wine
~0.05mn
8 glasses ~1.3mn
bottle

© The Consulting Club, FMS Delhi 2023-24 105

 Guesstimate Index | Main Index Bisleri Water Bottle (Part 1)
Predict the user base of 1 litre Bisleri water bottle in Delhi
Assumptions Methodology
• Yearly tourists will be our potential customer base and • User Base = Tourists in a year + restaurants in Delhi
restaurants will be potential business that buys Bisleri Bottle • Tourists that Delhi experience in a year = Domestic Tourists + Foreign Tourists
• Tourists and Restaurants are Direct Customers • Domestic Tourists in Delhi = # People who go on trip somewhere in India/# Tourists Spots
• Foreign tourists are 1% of domestic tourists in Delhi • Foreign Tourists = 20% of Domestic Tourists
• High Income Indians prefer to travel internationally

Income Group % Penetration Total Tourists in Delhi in a year

High 10% 40% 0.04 Percentage of tourists buying
Medium 30% 80% 0.24 mineral water bottle = 50% of
Domestic Tourists Foreign Tourists Domestic Tourists + 100% of
Low 60% 10% 0.06
Foreign Tourists
= 0.5*20 million + 200K
Foreign Tourists = = 10.2 million
Total Tourists in India # Tourist Spots Domestic * 0.01
= 200K Percentage of tourists buying
Age Group: 18-50 yrs (65% of 1.4bn) 15 Bisleri = 50%

# Potential Customer base in a

Domestic Tourist = Total Tourist in India/ Tourists Spots year = 5.1 million
= (0.04+0.24+0.06)*0.65*1.4bn / 15
~ 20 million

© The Consulting Club, FMS Delhi 2023-24 106

 Guesstimate Index | Main Index Bisleri Water Bottle (Part 2)
Predict the user base of 1 litre Bisleri water bottle in Delhi
Assumptions Methodology
• 3,4,5 Star and Mall restaurants = 50% 1 & 2 star • # Restaurants in Delhi = # People who dine out per day/Average daily capacity of a restaurant
• Bachelors prefer to order and dine out only during special
occasion
• Working Hour of a restaurant = 8
# Restaurants in Delhi
Category % Penetration Frequency
Office Goer Once in 2 1 & 2 Star 3, 4 & 5 Star & Mall restaurant
30% 50%
s Parties weeks
Bachelors 30% 50% Once a week
Total # 3,4, 5 star & Mall
Once # People who dine out daily Avg. Daily Capacity
Family 20/4 50%
a month restaurant = 0.5*10.2K
= 5.1K
Total = 0.3*0.5*20million/14 + 0.3*0.5*20 Seating capacity * occupancy rate
Population of Delhi = 20 million * working hours
million/7 + 5million*0.5*4/30 ~ 980K
= 40*0.4*8 = 96

Total # 1 & 2 star restaurant =

Total restaurants in Delhi = 15.3k
980k/96 = 10,200K

User base of Bisleri Water Bottle =

Tourists + 50% of all restaurants =
5.1 million tourists & 7,650 restaurants

© The Consulting Club, FMS Delhi 2023-24 107

 Guesstimate Index | Main Index Sanitizer Demand
Estimate the amount of sanitizers used in Delhi in a month
Assumptions Methodology
• Only personal use sanitizers estimated. •Amount of sanitizers used = Consumption/ family * number of families using sanitizers
•If a family uses sanitizers, the usage remains same across all •To estimate number of families using sanitizers, we start with the population of Delhi and divide it into income
income segments. (usage = 100ml/month/family segments
•Family size remains the same across all income segments (=4) • Number of families in each segment is calculated
• Penetration of sanitizers is 0% in below poverty line section of •Based on observation, penetration in each segment is decided and then number of families using sanitizers in
economy as increases as income increases each segment is estimated
•Number of families is multiplied by average consumption to get the sanitizer demand in a month

Population (19m) Let avg monthly usage of

household = 100ml

Below Poverty Line Low Income Mid Income High Income

Monthly Demand = No. of
households * avg monthly
Proportion of population 10% (1.9m) 30% (5.7m) 40% (7.6m) 20% (3.8m) usage/ household

No. of families (if size =4) 0.47 m 1.4 m 1.9 m 0.95 m

Monthly Sanitizer
Demand = 25.1m
Penetration 0% (0m) 30% (0.42 m) 60% (1.14 m) ~100% (0.95 m) L/month

2.51m

© The Consulting Club, FMS Delhi 2023-24 108

 Guesstimate Index | Main Index BCom(H) Admissions in DU
Estimate the total
numbernumber of BCom(H)
of BCom admissions
(H) admissions in Delhi University in a year.
in DU
Assumptions Methodology
• Equal number of Tier 1, 2 & 3 colleges offering BCom • DU colleges are to be divided into 3 categories – Tier 1, Tier 2 and Tier 3 colleges
• 80 colleges under DU, 70% offers BCom • Total number of BCom (H) admissions = Number of students in Tier 1 + Tier 2 + Tier 3 colleges
•Tier 1, 2 & 3 colleges reserves 80%, 60% and 40% of their BCom •Number of admissions in a college = Total #seats * % seats reserved for specific course * {1-% seats that
seats respectively for BCom (H) admissions will remain vacant}
• Less number of seats will remain vacant in top tier colleges at the end • There are two different courses offered under BCom – BCom (Honors) & BCom (Program)
of admission cycle • Top tier colleges will have more BCom (H) seats as compared to the other tier colleges

Colleges offering BCom(H) (80*0.6)

Tier 1 (16) Tier 2 (16) Tier 3 (16)

#BCom Seats =16*500 = 8000 =16*500 = 8000 =16*500 = 8000

% of BCom (H) seats 80% 60% 40%
% of vacant BCom (H) seats at
the end of admission cycle 5% 15% 20%

BCom (H) seats (Tier 1) ~ 6400 Number of students = 6400*0.95 = ~6080

Total #BCom
Total #BCom seats = 24000 BCom (H) seats (Tier 2) ~4800 Number of students = 4800*0.85 = ~4080 (H) students
~ 12,720
BCom (H) seats (Tier 3) ~3200 Number of students = 3200*0.8 = ~2560
© The Consulting Club, FMS Delhi 2023-24 109
 Guesstimate Index | Main Index Credit Cards Issued
Estimate the number of credit cards issued in Delhi per annum
Assumptions Methodology
• Population of Delhi : 20 million • No of households = Total Population/Average size of household
• Size of average household : 4 people • No of cards per household is considered by taking an average of all the households (Across all categories)
• Average life of card = 3 years • Considering that all households of high income group of rural population own a card would compensate for multiple
• Average number of cards / household = 1 cards in households of high income group of urban population.
• High income and Middle income groups of Urban population • Considering only upper middle income group of urban population owning cards, we take 50% of the middle income
and High income group of Rural population own credit cards segment for simplicity.

Population of Delhi

Urban (70%) Rural (30%)

High Income Middle Income Low Income High Income Middle Income Low Income
(25%) (55%) (20%) (15%) (30%) (55%)

Population owning credit cards 4 member/family 1 card works for 3 years

= 20 million [ 0.70 ( 0.25 + 0.55 * 0.50 ) + 0.30 ( 0.15 ) ] So, 8.25/4 So, 2.0625/3
= 8.25 million =2.0625 =0.6875

Total Credit cards issued in Delhi per annum = 687500

© The Consulting Club, FMS Delhi 2023-24 110

 Guesstimate Index | Main Index Revenue of Dream 11
Assuming Dream11 is starting its business in India (2018), Estimate its total possible revenue in the first year
Assumptions Methodology
• 70% of total revenue is generated by cricket • We start with calculating the revenue from cricket and will adjust for other sports in the end
• No other competitor are present in the online fantasy • Filters for geography (urban & rural), gender, sports viewership, cricket viewership, digital payments and frequency
gaming sector of usage are utilized to arrive at the final numbers
• Focus is more on the urban market as it's the first year • The calculated value for urban (cricket) is extrapolated to include all sports and demography
of business

Revenue of Dream11
# population 400 M
# Dream11 players 6 M Dream11 Players
Cricket (70%) Other
Sports(30%)
Revenue from
Frequency High (10%) Medium (30%) Low (60%)
Urban: 80%
Urban Revenue from
Rural (70%)
(30%) Rural: 20%
0.5M 2M 3.5M
Age 18-25 (15%) 26-40 (20%) 40+ (%0%)
Male Female Male Female Male Female # bets placed 60/year 48/year 18/year
30M 30M 40M 40M 100M 100M Avg. value per bet 500 250 100
% watch sports 60% 18M 40% 12M 50% 20M 30% 12M 40% 40M 20% 20M
% watch cricket (70%) 13M 8M 14M 8M 30M 14M
Total Revenue from cricket (urban) = 4530 Cr
% Use digital payments 70% 70% 80% 80% 50% 50%
9M 6M 11M 6M 15M 7M
Total Revenue accounting all demography and sports = 8090 Cr
% play fantasy sports 20 % 2M 15% 1M 15% 1.5M 10% 0.5M 5% 0.8M -

© The Consulting Club, FMS Delhi 2023-24 111

 Guesstimate Index | Main Index
Revenue of Pickpocket
Estimate the earnings of a pickpocket working in the kamla
marketmarket
area of a metropolitan city
Assumptions
Assumptions Methodology
Methodology
• The pickpocket operates only in one market • We start with calculating the total number of people present in the market at any time based on shop timings and
• 70% of people visiting the market visit at least one customer count.
shop • Further we segregate the customers based on the number of people who carry wallets and pickpockets success
• Customer count per shop is considered constant for frequency.
ease of calculation but could be segregated further. • Finally an average wallet value is used to calculate the total revenue of the pickpocket.

#people in the market who visit shops = 150*200/3 = 10000

People who
visit shops # people visit Kamla Accounting for 30% of people who are passerby = 10/7 * 10000 = 14000
70% market

Visit shops Passerby # people visit Kamla market

Average no of
shops visited by # people not
4-9 PM # people carrying
one person = 3 carrying wallets
wallet (60%)
9-4 PM # viable pickpocket (40%)
targets = 2700
20 customers/hr Careless Careful Average amount
10 customers/hr
(30%) (70%) per wallet = 500

Total number of people = (No of shops)*(customers per Revenue (day) = (# viable targets)*(% attempted (3%))*(% successful (10%))*(Avg. amount per wallet)
shops)/(Avg. no of shops visited by one person) Revenue = 2500

© The Consulting Club, FMS Delhi 2023-24 112

 Guesstimate Index | Main Index Number of Movie Screens
Estimate the number of movie screens in Delhi
Assumptions
Assumptions Methodology
Methodology
• No of seats in a screen is 300
• We start by splitting the population into income groups
• Operational hours in a day = 15 hours
• Further we segregate the people into movie buffs and casual watchers
• 1 show during morning, afternoon, and evening
• We calculate the number of people watching movies on a single screen
• 2 shows during night time
• Using the total number of movie watchers in a month we can identify the number of movie screens required to
• Frequency of a movie buff is 4 and casual watchers
cater to the demand.
is 1 movie per month

1 week Total audience catered by a screen =

(Occupancy)*(No of shows)*(No of days) Population of Delhi
(2 cr)

Weekdays Weekends Middle Income High Income

Low Income (25%)
(45%) (30%)

Day time Night time Morning & Evening &

Casual Casual
shows shows Afternoon Night Total number of seats booked = Movie Buffs
Watchers Movie Buffs Watchers
(10%) (10%) (25%)
(25%)
(No of people either movie buff or
Occupancy = Occupancy = Occupancy = Occupancy = casual watchers)*(Frequency)
60% 60% 60% 60%
No of
Total no of seats booked (Middle income + High
No of No of No of
shows = 3 shows = 2 shows = 2 shows = 3 income) = (90L*10%*4 + 90L*25%*1) +
(60L*10%*4+60L*25%*1) = 97.5L
For a month one screen can cater to 32000 seats
No of Screens = 97.5L/32000 = 300 screens

© The Consulting Club, FMS Delhi 2023-24 113

 Guesstimate Index | Main Index Petrol Consumption
Estimate the quantity of petrol used per day (in litre ) for transportation in NCR
Assumptions Methodology
• Diesel and CNG is used by public transport and commercial • Calculate the Number of households
• vehicle (busses, vans, and trucks) • Calculate Vehicles per household.
• Average distance traveled by each vehicle (per day) = 20 km • Calculate the consumption by each vehicle.
• Mileage (in km/liter) : 2-Wheeler = 50, 4-Wheeler = 10 • Scale up the consumption to the entire population

Population of NCR
Average no. of vehicles possessed by households
(2cr) Average members
Low Middle High per household = 5
Vehicle BPL
Income Income Income Total # of
2 Households = 60L
0 0.5 1 1.5
wheelers
4 Low Income (50%) Middle Income (25%) Upper Income (5%)
0 0 1 2 BPL (20%) = 2L
wheelers
= 30L = 15L = 3L
Total number of vehicles Litres of Fuel Used
Vehicle Distance Travelled (km)
Low Middle High (Distance/Mileage)
Vehicle BPL
Income Income Income 2
20*(15,00,000 + 15,00,000+ 4,50,000) = 6,90,00,000 13,80,000
wheelers
2 Total Fuel Used =
0 15 lakh 15 lakh 4.5 lakh 4
wheelers 20*(15,00,000 + 6,00,000) = 4,00,00,000 40,00,000 53,80,000 Litres
wheelers
4
0 0 15 lakh 6 lakh Assuming the 20% consumption is of diesel, Petrol Used in NCR (by NCR People) is 0.8 * 53,80,000 = 43,04,000
wheelers
Assuming 10% of the consumption is by non-residents of Delhi, the quantity of petrol used per day (in liter) for transportation in Delhi
= 1.1 * (43,04,000) ≈ 48,00,000 Litres
© The Consulting Club, FMS Delhi 2023-24 114
 Guesstimate Index | Main Index
Swiggy Drivers
Estimate the number of Swiggy drivers in Mumbai
Assumptions Methodology
• Lower income strata not considered for analysis • No. of Swiggy drivers would be the number required to deliver orders during the
• Market share of Swiggy = 45% peak hours
• 1 order is delivered by a driver in 30 minutes during peak hours ( from restaurant to delivery • Let no. of orders = Population x Smartphone penetration x Internet penetration x
location ) Market share of Swiggy x Frequency of orders
• The order requirement during peak hours is double the normal hour
Population of Mumbai Total No. of people using Swiggy for ordering food = 34L * 45% = 15.3L ~ 15L
2 crore
Swiggy Users (15L)

Upper Income Middle Income Lower Income

Income
(10%) (40%) (50%) High Order Medium Order Low Order
20L 80L 1 crore Frequency (2%) Frequency (20%) Frequency (78%)
0.3L 3L 11.7L
Income Smartphone Internet Online food
Total
Segment Penetration Penetration delivery
10 orders per 5 orders per 1 order per
Upper week week week
100% 100% 70% 14L
Income
Total # of orders per week = (0.3x 10 + 3 x 5 + 11.7 x 1.5) = 35.55L
Middle
80% 80% 40% 20.4L
Income # of orders/ hr = 51/18*7 = 28.2K
~20L
Order during peak hours = ~56 K
Lower
40% 10% 1% 4,000 # of drivers required to meet this demand = Peak orders/ Orders delivered by a driver
Income
in 1hr
Total # of people preferring online food delivery = 34L Total No. of Swiggy drivers = 28K
© The Consulting Club, FMS Delhi 2023-24 115
 Guesstimate Index | Main Index Delhi Metro Passengers
Estimate the number of passengers in Delhi Metro per day for yellow line
Assumptions Methodology
• Hours of operations of Delhi metro is from 5:30 AM to • We will use a bottom up approach to find the number of passengers/coach and extrapolate it to find no. of
11:30 PM passengers/day.
Capacity per coach is 50 and no. of coaches is 6. • No. of passengers per day = #. of passengers/hr * Hours of operations/ day
• If time interval between two trains is 2 min for one • No. of passengers per hour= # of metro trains/ hr * no. of coaches * capacity/coach* % occupancy
direction, for both directions, it would 1 min

No. of passengers

Weekday Weekend

High occupancy Med occupancy Low occupancy High occupancy Med occupancy Low occupancy
hours (6hrs) hours(7hrs) hours(5hrs) hours(3hrs) hours (7hrs) hours (8hrs)

Occupancy % 150 % 100 % 40 % 120 % 90 % 30 %

Frequency of trains / min 1 2/3 1/2 1 2/3 1/2

60*50*6*1.5 40*50*7*1 30*50*5*0.4 60*50*3*1.2 40*50*7*0.9 30*50*8*0.3

# of passengers
=27K =14K =3K =10.8K =12.6K =3.6K

# of Passengers/day = (44*5+27*2)/7 = 39.1K ~39K

Total No. of Delhi Metro passengers per day in yellow line = 39K

© The Consulting Club, FMS Delhi 2023-24 116

 Guesstimate Index | Main Index Toothbrushes
If a UFO sucked all the toothbrushes in India, how many would it have?
Assumptions Methodology

• Toothbrush penetration-90% urban and 80% • Total no of toothbrushes = No of toothbrushes in Households + No of toothbrushes in supply
rural chain(retailers/distributors)
• Replacement frequency- 2 months Urban; 4 • No of toothbrushes in the supply chain can be assumed to be equal to 1 replacement cycle.
Months Rural • Therefore just double the no of toothbrushes in households to account for the total number.

Total No of toothbrushes

Urban Rural

#Population Penetration #Population Penetration

400 million 0.9 900 million 0.8

No of toothbrushes in households = 360+720=1.08 billion

Total Number of toothbrushes = 2.16 billion

© The Consulting Club, FMS Delhi 2023-24 117

 FOCUS 3

Guesstimates

© Consulting & Analytics Club, IIT Guwahati

 What’s a Guesstimate?
The term is a portmanteau of guess and estimate and was rst used by American
Statisticians in 1934-35. As such, a guesstimate aims to test a candidate’s qualitative and
quantitative skills, as well as capacity for thinking on-their-feet.

A guesstimate combines conjecture and calculation. Therefore, it is an estimate based on

rudimentary information and perception.

By the end of this section, you'll understand:

1. Using guesstimates in a practical context,
2. Advantage and scope of guesstimates,
3. How to approach a guesstimate question.

Additionally, we've added solved problems and curated a collection of guesstimates asked by
MBBs, The Big Four, software giants and other Wall Street rms.
© Consulting & Analytics Club, IIT Guwahati
fi
fi
Advantage and Scope
1. Testing the candidate
It tests the analytical power as well as the numeric skills of the candidate. Moreover, it gives an
idea of their ground-level understanding and ability to approach a problem.

2. Rough Validation of Data

It gives a vague estimate of values to check the practicality of numbers; hence it works as an
excellent tool to discard wrong statistics.

3. Feasibility of Executing a Plan

It checks for the chances of successful execution of an event, preventing wastage of resources.

4. Quick and Economical

Surveys are generally too expensive and time taking. So, guesstimates get the upper hand when
the exactness of data doesn't matter.

5. No prerequisites
There is no need for preplanning and thus provides exibility. This could be performed at any
moment with the information we usually possess.
© Consulting & Analytics Club, IIT Guwahati
fl
 Tips and Tricks
1. Do not take guesstimates lightly, as they are now essential for several interviews. Prepare
various guesstimates beforehand so you are well equipped to tackle them during the
interview.

2. Remember, it is not about numbers but the approach. However, take assumptions with
justi cations and keep them as sane as possible (you can ask the interviewer if your belief
makes sense).

3. Think out loud – walk the interviewer through your thought process.

4. Work with simpler numbers, round off intelligently, and use percentages.

© Consulting & Analytics Club, IIT Guwahati

fi
General Method (1/2)

Step 1 Step 2

Developing a Strategy
Clarifying the Scope
Think of an overall strategy that exhaustively covers
Anything that is/may seem ambiguous must
the issue at hand. Try to think from different
be cleared right away. Ask questions! Smart
perspectives as well. Generally, three methods are
questions show that you have an eye for
used (with tailored variations) — Top-Down
detail. You need to identify the constraints you
Method, Bottom-Up Method, and Employing a
are working under to answer clearly and move
Proxy. Base your strategy on general 2-3 element
through the case faster.
formulae (such as averages, demand, and supply).

© Consulting & Analytics Club, IIT Guwahati

General Method (2/2)

Step 3 Step 4

Break down the problem

Perform Math & Sanity Checks
Once you identify the constraints you must
Estimate each piece separately and multiply them
work under, break down the problem into
to get the overall estimate. Perform sanity checks
smaller pieces (a MECE approach of-course).
mentally wherever there is “sensitive data” and
Do keep in mind the following factors while
suggest sanity checks on your own to the
breaking it down into pieces — Population,
interviewer. Making sure your answer is realistic
Income split, Age split, Gender, and Rural-
shows that you are methodical in your approach.
Urban split.

© Consulting & Analytics Club, IIT Guwahati

Solved Examples

Problem Statement 1

Estimate the number of routers sold in India in a year.

Clarifying Questions

1. Should I consider the impact of the COVID-19 pandemic, or is scenario one during the pre-
COVID times? (Consider post-pandemic times)
2. By router, do you mean the traditional router where I should also include a modem that
comes with it or the modern one which does both the work of a router and a modem?
(Consider both)

Notes

1. Always proceed with caution and be mentally aware of what you are doing.

© Consulting & Analytics Club, IIT Guwahati

 2. Think of an overall strategy (You could try thinking of supply/demand/geography
demographics, but these approaches involve too many variables). So, we will proceed by
considering the end user demographic, i.e. residential and commercial users.
3. Since we have talked about COVID-19 and its impact, we must necessarily include the
growth of the Internet over the years (be it post-COVID or pre-COVID), bringing in new
users.
4. We must now consider approaches to solve both the pieces (Residential and Commercial
components) by breaking them down further.
5. We must also factor in the lifetime of a router and the replenishment rate.
6. Keeping in mind all the above points, we now base our guesstimate on the following two
formulae:
1. =

2. = +

= ( × )+
𝖠
𝗏
𝖾
𝗋
𝖺
𝗀
𝖾
𝗇
𝗎
𝗆
𝖻
𝖾
𝗋
𝗈
𝖿
𝗉
𝖾
𝗈
𝗉
𝗅
𝖾
𝗎
𝗌
𝗂
𝗇
𝖫
𝗀
𝗂
𝖿
𝖺
𝖾
𝗍
𝗋
𝗂
𝗈
𝗆
𝗎
𝖾
𝗍
𝖾
𝗋
© Consulting & Analytics Club, IIT Guwahati
𝖭
𝗎
𝗆
𝖭
𝖻
𝖾
𝗎
𝗋
𝗆
𝗈
𝖻
𝖿
𝖾
𝖾
𝗋
𝗑
𝗈
𝗂
𝗌
𝖿
𝗍
𝗂
𝖾
𝗇
𝗑
𝗀
𝗂
𝗌
𝗋
𝗍
𝗈
𝗂
𝗇
𝗎
𝗀
𝗍
𝖾
𝗋
𝗋
𝗈
𝗌
𝗎
𝗍
𝖾
𝗋
𝗌
𝖦
𝗋
𝗈
𝗐
𝗍
𝗁
𝗋
𝖺
𝗍
𝖾
𝖭
𝗎
𝗆
𝖻
𝖾
𝗋
𝗈
𝖿
𝗇
𝖾
𝗐
𝗋
𝗈
𝗎
𝗍
𝖾
𝗋
𝗌
𝖺
𝖽
𝖽
𝖾
𝖽
𝖬
𝖺
𝗋
𝗄
𝖾
𝗍
𝖦
𝗋
𝗈
𝗐
𝗍
𝗁
𝖱
𝗈
𝗎
𝗍
𝖾
𝗋
𝗌
𝖱
𝖾
𝗉
𝗅
𝖺
𝖼
𝖾
𝖽
𝖭
𝗎
𝗆
𝖻
𝖾
𝗋
𝖭
𝗈
𝖿
𝗎
𝗆
𝗎
𝗌
𝖻
𝖾
𝖾
𝗋
𝗋
𝗌
𝗈
𝖿
𝖾
𝗑
𝗂
𝗌
𝗍
𝗂
𝗇
𝗀
𝗋
𝗈
𝗎
𝗍
𝖾
𝗋
𝗌
Considerations

1. Internet growth during the pre-COVID times was around 4%, and that during post-COVID
times was about 10%.
2. The percentage of users who have access to the internet in India pre-COVID was 40%, and
post-COVID was around 50%.
3. The workforce of India is around 35% of its population, and the percentage of people
working in unorganised and other blue-collar jobs is about 80%. India's population can be
considered to be 1.40 billion.
4. The average lifetime of a router is ve years.

© Consulting & Analytics Club, IIT Guwahati

fi
Structure

% using routers

Residential Component
Total Population

+
Routers replaced
Existing Routers for Internet

÷ No. of people using

routers for Internet
Average lifespan ÷
No of users per
Annual Sales

household
+
+

% using routers

Commercial Component
Working Population

+
for Internet
Growth Rate
Market Growth

No. of people using

routers for Internet
Existing Routers ÷
No of users per
router

© Consulting & Analytics Club, IIT Guwahati

Calculations

Residential Component Commercial Component Total number of routers

sold in India every year
Population having access to internet Working population of India = 35%
= 50% = 700 million = 500 million = Residential Component +
Commercial Component
20% of 700 million rely on routers People working in White and Grey
= 140 million Collar Jobs = 20% = 100 million
= 8.4 + 1.9
Apply Formula 1 → say if 5 people No. of people with access to router = = 10.3 million
share a router → 70% of 100 million = 70 million ~ 10 million
Number of existing routers
= 140/5 = 28 million Apply Formula 1 → say if 10 people
share a router →
Number of existing routers
Apply Formula 2 →
= 70/10 = 7 million
Number of New Routers
= (28 x Growth rate) + (28 / Lifetime)
= (28 x 0.1) + (28 / 5) = 8.4 million Apply Formula 2 →
Number of New Routers
= (7 x Growth rate) + (7 / Lifetime) =
(7 x 0.1) + (7 / 5) = 1.9 million

© Consulting & Analytics Club, IIT Guwahati

Problem Statement 2

What is the market size of sofas in India?

Clarifying Questions

1. Does market size refer to volume or value? (Consider the value of sofas sold in a year)
2. Are single-seated armchairs included here? (By definition, no.)
3. Sofas sold or produced? (Consider only the sofas sold in India.)

Considerations

1. Sofas are, of course, needed in homes. Other possible uses are – in of ces, airports, lounges,
cafeterias, and hotels. Since India has many households, the lion's share of sofas is for
residential purposes. (An adjustment may be applied to the final answer to account for the
rest of the possible users – say 30%).

2. Since the problem requires the value of sofas sold in a year, it is helpful to assume that there
is no bottleneck in supply. Thus, an approach from the demand side would be the most
useful.
© Consulting & Analytics Club, IIT Guwahati

fi
3. Consider the population of India as 1.4 billion and an average of 5 people per household.
4. Higher-income households may have more than one sofa. On average, assume two sofas per
household.
5. Penetration Rate: Consider the 30:70 split of urban and rural populations in India. Let's say
this applies to families too. Assume 30% of urban and 1% of rural households have sofas.
This amounts to approximately 10% penetration.
6. Assume sofas get replaced every ten years on average. Assume the sofa market grows
parallel to population growth. Thus, 1% market growth.
7. Standard sofas may cost INR 20k, with a 75% market volume. Premium sofas may cost ve
times as much.

© Consulting & Analytics Club, IIT Guwahati

fi
Structure

Total no. of sofas Number of households Population

Sofas replaced ÷ ÷
Annual Sales

+
Average lifespan No. of people
Sofas per household
+ per household

+
Total no. of sofas
Market Growth Penetration rate

+
Sofa Market Size

Growth Rate
+

Average unit cost

Average unit price

Standard Sofas +
% market volume
+
Average unit cost
Premium Sofas
+

% market volume

© Consulting & Analytics Club, IIT Guwahati

Calculations

56 million 280 million households 1.4 billion

6.16 million/year 5.6 million/year ÷ ÷

+
10 years 5 people/
2 sofas/household
+ household

+
56 million
0.56 million/year 10%

+
INR 246.4 billion

1%/year
+

Sanity Check
INR 20,000 Dividing the total market size by the number
INR 15,000 of households yields a spend of INR 880 per
+
75%
INR 40,000

household per year (~70/month).

+
INR 100,000 This is reasonable, as each household would
INR 25,000 on average only spend a little on sofas (a fairly
+

25% long-term purchase).

© Consulting & Analytics Club, IIT Guwahati

 Main Index Introduction to Guesstimates
What is evaluated through What is necessary to solve a good
Guesstimates? Guesstimate?

Ability to think on Approach & Quantitative Communication &

Logical Thinking
your feet Structure Skills Presentation

Guesstimates: Short, number–intensive estimation Cases

Ideal Time Limit: 15-20 minutes

Top Down and Bottom Up Supply Side and Demand Side

Approach Approach

© The Consulting Club, FMS Delhi 2023-24 33

 Main Index Guesstimates Do’s & Don’ts
Ideal Flow Don’ts
Do’s
Confirm Objective
• Take about a minute to • Ask too many clarifying
gather your thoughts and Think logically and come up with possible set of questions
decide approach approaches
• Questions about approach
Explain the best approach & confirm if you should go
• Use tree diagrams, normal ahead with it
diagrams, anything that • Start solving without
explains your thoughts State your assumptions first hand discussing the approach
clearly in a visual way
Lay down structure neatly on paper and solve it step by step • Start with a population set
• Relate your assumptions to Make logical assumptions and always confirm them with
every time
facts, experiences and Interviewer
sellable logic • Be text heavy on your sheet
Keep communicating & asking the interviewer for buy-ins
• Keep talking as you write, • Unreadable writing
Calculate your answer. Be ready for a conversation around
engage the interviewer
error estimate, other approaches
• Guessing the numbers
If possible, reconfirm & triangulate your answer with a
ballpark estimate from another approach • Solving without explaining
what you are doing

© The Consulting Club, FMS Delhi 2023-24 34

 Main Index Top Down & Bottom Up Approach
Top down approach Bottom up approach

• Start with an entire population (in other words, the top level) • Start from the bottom—some low-level statistic, such as
and then breaking it down until you arrive at an answer. Revenue per store, which does not change across your universe
and build your way up to the answer.
Identify a Starting
Universe
Identify relevant conditions/filters and

Identify the smallest replicable block

segments

Segment A Segment B
Estimate for a single identified block

Segment Segment Segment Segment Segment Scale up!

A1 A2 A3 A2 A3

Guesstimate = A1 + A2 + B1 + B2 +B3 • Bottom up approach is much more subjective than top down
approach.
• Segments: • Especially replicable blocks depend on the case in hand, it can be
• Demographics (age, sex, income) one single store, one family to a single person. Be careful while
• Psychographics(attitudes, behaviors, values) picking your block and while scaling up.
• Geography (city/country, urban vs. rural) • Bottom up approach though gives accurate results provided you
• And many more depending on the case! scale up properly.

© The Consulting Club, FMS Delhi 2023-24 35

 Main Index Guesstimate Cheat Sheet
Total Population of India ~143 Cr Age-wise Distribution of Population Largest cities (by Population)

Age Bracket Percentage Cities Pop. (Mn) Cities Pop.

(Mn)
0-14 Years 26% Delhi 32.94 Chennai 11.78
15-24 Years 18% Mumbai 21.3 Hyderabad 10.8
Gender Split Urban to Rural Average
Women/Men: 35:65 Household Size Kolkata 15.33 Ahmedabad 8.65
25-34 Years 17%
48:52 4-5 people
Bangalore 13.61 Surat 8.06
35-44 Years 14%
Birth Rate Population Avg. Life
1.69% Density Expectancy GDP Distribution
Death Rate 480/km^2 70.4
45-54 Years 10%
0.74%
55+ Years 15%
Agriculture 20% Total GDP
(~3.4 Trillion
Industry 54% USD)
Geographical Distribution Internet Connectivity Distribution 26%
Service
Length of India 3200 kms (North to South)
Internet Penetration
Breadth of India 3000 kms (West to East) 50% Income wise distribution of population
Number of States 28 Smart Phone Penetration Income Segment Percentage
Number of UTs 8 71%
Lower Income 10%
Area of Delhi 1500 square kms Lower Middle Income 38%
Religion split
Population of Delhi 3.3 Cr Religion Hindu Islam Christianity Others Middle Income 40%
Area of Mumbai 600 square kms Upper Middle income 10%
Population 80% 13% 2.3% 4.7%
Population of Mumbai 2.2 Cr Upper Income 2%

© The Consulting Club, FMS Delhi 2023-24 36

 GUESSTIMATES

© Consulting Club, SPJIMR

GUESSTIMATES
Coffin Maker’s Profit – Transcript
Problem: Estimate the profit made by a coffin maker in The Maldives.

Sure. So I have to estimate the profits of a coffin maker in the Maldives in a year. Right?
I: Yes. Go ahead.
I will need some information for the analysis. May I know what is the population of Maldives?
I: You can assume it to be 40 million.
Alright, may I now know the age distribution of the population?
I: People across all age groups till 100 years live in the country. Assume that the population is
divided uniformly.
What is the average life expectancy in the country?
I: Assume it to be 75 years.
Would it be fair to assume only Christians use a coffin to bury the dead?
I: Good point. But for now, assume that 10% of the dead people are buried in a coffin.
Sure. What is the selling price of a coffin and costs associated in making it?
I: Take Selling price = 100 USD, Fixed cost = 100000 USD, Variable cost = 60 USD/coffin.
Thank you! Next, what is the market share of our client?
I: That would be 10 percent.
Okay. Let me take a minute. The population is uniformly distributed across ages (0,1,…100).
Therefore 4,00,00,000/ 100 = 4,00,000 would be the population for each age. Furthermore, as
average life span is 75 years, shall I assume people who cross age of 75 per year i.e. 4,00,000 is the
death toll of the country per year.
Yes, you can assume that. What next?
So, No. of coffins sold by our client = 4,00,000 * % of people buried in a coffin * market share of our
client = 4,00,000 x 0.1 x 0.1 = 4000. Thus Profit = [No. of units sold x (selling price– variable cost) –
Total fixed cost] = 4000 x (100-60) - 100000 = 60,000 USD per year.
Thank you for your analysis.

© Consulting Club, SPJIMR

GUESSTIMATES
Coffin Maker’s Profit – Case Structure
Problem: Estimate the profit made by a coffin maker in The Maldives.

Problem Scoping
Segment People with 75+ years of age Location Maldives
Social Segments Lower Income, Middle Income and High Income

Notes
Profit = No. of units sold X (Selling price – Variable cost) – Total fixed cost

Total Coffins
Sold

Population of % of people % of people

X X X Market Share
Maldives above 75 Yr. buried in Coffin

Key Insights

• The candidate exhaustively broke down the problem into its underlying levers
• The candidate maintained a conversational format and enquired the interviewer on all the levers. This also reduced the need for many assumptions

© Consulting Club, SPJIMR

GUESSTIMATES
Toothpaste Consumption – Transcript
Problem: Calculate the total number of toothpaste tubes consumed in India annually.

I am required to calculate the total number of toothpaste tubes consumed in India annually. Is that Lower Income Group = 40% of population = ~140 million households. Most of the lower income
right? groups will purchase a small size toothpaste pack.
Use once a day = 100% = 140 million households. One small toothpaste would last for 3 weeks. Total
I: Correct!
purchase = 140 Mn x 16 = 2240 million small toothpastes.
I would like to divide the population based on income. Then for each income group, for simplicity, I
I: Your household numbers don’t add up. You have excluded 20% of households. Care to explain?
would assume that only one kind of SKU is consumed. In toothpastes, SKUs are typically large,
medium, small. Then the income groups will be further divided into sub-groups based on 20 percent of households fall below the poverty line. They have an income of less than Rs 80 per
consumption patterns (usage once a day or twice a day). day. They wouldn’t use toothpaste. Rather they prefer salt or toothpowder because of lower prices.
I: Okay, that seems like a good approach. I: That’s fair.
I would start with the higher income group. So, the total number of toothpastes consumed in India are 3.4 billion, across various SKUs.
I: Okay. I: Thank you for the analysis.
Higher Income Group = 5% of population = 17.5 Mn household. Most of the higher income groups
will purchase a large toothpaste pack.
Use once a day = 70% = 12 million households. One large toothpaste would last for 2 months. Total
purchase = 12 Mn x 6 = 72 million large toothpastes.
Use twice a day = 30% = 5.5 million households. One large toothpaste would last for 1 months. Total
purchase = 5.5 Mn x 12 = 66 million large toothpastes.
I: Alright, proceed.
Middle Income Group = 35% of population = ~120 million households. Most of the middle income
groups will purchase a medium size toothpaste pack.
Use once a day = 80% = 96 million households. One medium toothpaste would last for 1.5 months.
Total purchase = 96 Mn x 8 = 768 million medium toothpastes.
Use twice a day = 20% = 14 million households. One medium toothpaste would last for 3 weeks.
Total purchase = 14 Mn x 16 = 224 million medium toothpastes.
I: Okay, what next?
I would now compute the usage for lower income groups.
I: Yes.

© Consulting Club, SPJIMR

GUESSTIMATES
Toothpaste Consumption – Case Structure
Problem: Calculate the total number of toothpaste tubes consumed in India annually.

Households In
India (350Mn)

Middle Income Lower Income Below Poverty

Rich (5%)
Group (35%) Group (40%) Line (20%)

Large tube Medium tube Small tube

Brush twice a Brush once a day Brush twice a Brush once a day Brush once a day
day (30%) (70%) day (20%) (80%) (100%)

Key Insights
• It is advisable to down the problem in sub-buckets as done by the candidate in this case – it makes taking assumptions easier and presents the solution more comprehensively
• Identifying the different SKU sizes and frequency of usage introduced depth in the analysis

© Consulting Club, SPJIMR

GUESSTIMATES
McDonald’s Burgers – Transcript
Problem: Calculate the total number of McDonald’s burgers sold in Mumbai in a month.

Jut to confirm, I have to calculate the total number of McDonald’s burgers sold in Mumbai in a month
. Is that right?
I: Correct.
I would like to divide the area of Mumbai, which is ~600 sq. km, into urban and semi-urban. I would
assume there is a McDonald’s outlet every 5 sq. km & 10 sq. km in urban & semi-urban areas
respectively.
I: Okay, that seems like a good start!
Based on the computation, there are ~100 outlets in Mumbai.
I: Alright.
Next, I would like to focus on sales per store per day. Assuming the store remains open 15 hours a
day. For weekdays, I assume 10 hours of non-peak demand and 5 hours of peak demand. While, for
weekends, I have assumed 7 hours of non-peak demand and 8 hours of peak demand.
I: Good. What next?
For peak hours, there would be supply side constraint and supply would be at maximum production
capacity, which we can safely assume to be 2 burgers a minute.
For non-peak hours, we can assume the burgers sold would be half that in peak hours i.e. 1 burger a
minute.
I: Okay!
For weekdays (peak): 100 stores x 5 hours x 60 min x 2 = 60,000
For weekdays (non-peak): 100 stores x 10 hours x 60 min x 1 = 60,000
For weekend (peak): 100 stores x 8 hours x 60 min x 2 = 96,000
For weekend(non-peak): 100 stores x 7 hours x 60 min x 1= 42,000
Demand for a month = Demand for a Week x 4 Weeks = [(1,20,000 x 5) + (1,38,000 x 2)] x 4 = ~35
lac burgers per month.
I: Great, Thank you!

© Consulting Club, SPJIMR

GUESSTIMATES
McDonald’s Burgers – Case Structure
– CaseCalculate
Problem: Structure
the total number of McDonald’s burgers sold in Mumbai in a month.

Mumbai (600 Sq.km)

Urban Area Semi-Urban

(60%) Area (40%)

1 store every 1 store every

~100 McD Stores
5 sq. km 10 sq. km

Weekday sales Weekend Sales

Non-peak Non-peak
Peak demand Peak demand
demand demand
(5 hrs) (8 hours)
(10 hrs) (7 hours)

100 x 5 x 60 x 2 = 60,000 100 x 10 x 60 x 1 = 60,000 100 x 8 x 60 x 2 = 96,000 100 x 7 x 60 x 1 = 42,000

Monthly Demand = Weekly Demand x 4 Weeks = (1,20,000 + 1,38,000) x 4 = Approx. 35 lac burgers

Key Insights
• Focus more on the structure rather than arriving at a particular number
• Discuss the structure & assumptions with the interviewer prior to proceeding with the calculations
• Keep the numbers simple, as close to round figures as possible to ease out the calculations

© Consulting Club, SPJIMR

GUESSTIMATES
Coffee Consumption in Airlines – Transcript
Problem: Estimate the amount of coffee consumed in Jet Airways domestic flights in a year.

So, I need to estimate the amount of coffee consumed in Jet Airways in a year. Is that correct? Assuming an average occupancy at around ~60%. Number of passenger seats per day works out to
be = 540 x 150 x 60% ~48,000.
I: Correct!
I: Makes sense.
The total coffee consumption in a year for Jet domestic flights would be a function of multiple
factors: (Total no. of domestic planes flying per day) x (Jet Airways share) x (Capacity per plane) x Assuming that an average of 30% passengers in the flight drink coffee.
(Occupancy depending on route) x (Proportion of people ordering coffee) x (Amount per serving) x
I: Okay.
360 days. I would like to evaluate each parameter individually.
Typically in a cup of coffee has 100 ml served. So total coffee consumption in litres = 93,000 x 30% x
I: Good approach. Go ahead.
100 ml/1000 ~1500 litres.
For the total numbers of flights, I would like to assume 6 major airports & 20 minor airports.
I: That’s for a day.
Depending on the time of the day and runways.
Right, for a year the consumption would be = 1500 x 360 ~5.5 lakh litres.
I: Okay.
For ever major airport I would like to assume on an average 3 runways and 2 runways for a minor I: Right. Now Jet Airways needs to reduce its spending on coffee. How would you help them ?
airport. Assuming there are on an average 3 flights on every runway for minor airport & 6 flights per I would suggest the following options:
hour for a major airport. For a minor airport total flights per day = 3 flights x 24 hours x 2 runways • Renegotiate Prices with suppliers
x 20 airports. Similarly for major airports, total flights per day = 6 flights x 24 hours x 3 runways x 6 • Bulk Purchases
airports. Total number works out to be ~ 5,400 flights. • Hedge the prices
I: Are you sure? • Reduce the number of types of coffee offered
• Look at alternative suppliers
(After few seconds) This number has to be divided by 2 since a flight say from Mumbai to Delhi is
counted twice (at Mumbai as well as Delhi airport). The number of domestic flights per day is ~2700 I: That would be all, thank you.
flights.
I: That’s right.
Jet Airways has a market share of 20%. So assuming 15% of the above flights are operated by Jet
Airways ~540 flights.
I: Okay.
Typically an aircraft has 150 seats.
I: Yes.

© Consulting Club, SPJIMR

GUESSTIMATES
Coffee Consumption in Airlines – Case Structure
– CaseEstimate
Problem: Structure
the amount of coffee consumed in Jet Airways domestic flights in a year.

Approach
Domestic Flights
The total coffee consumption in a year for Jet domestic flights = per day
(total no. of domestic planes flying per day) x (Jet Airways share) x
(capacity per plane) x (occupancy depending on route) x (proportion of
people ordering coffee) x (amount per serving).

Major airports Minor Airports

Number of flights Number of flights

Total flights per per hour x 24 hrs
per hour x 24 hrs
day X # Runways x 20
X # Runways x 6

Total seats % of coffee (Millilitres served per cup

Jet flights per day X X X drinkers in the X /1000) x 365 days
Avg. Occupancy flight

Key Insights
• Focus more on the structure and flow of logic than calculations
• Discuss the structure & assumptions with the interviewer at each step
• Try to make an exhaustive approach from the beginning to cover all the parameters that might affect the calculation.

© Consulting Club, SPJIMR

GUESSTIMATES
Dinner Plates – Transcript
Problem: Calculate the total number of dinner plates in Mumbai.

I will first find out the total number of plates used in domestic households. Do you think that’s a Okay. Let me quickly lay down the structure (refer to the next page). Number of plates used in
good way to go ahead? Mumbai households = 26.5 Mn plates and Number of plates for restaurants = 0.2 Mn.
Thus, total dinner plates in Mumbai is ~26.7 Mn .
I: Okay that seems like a good starting point.
I: Thank you.
The population of Mumbai is 20 million. Assuming average household size as 4 in the urban areas,
we are looking at 5 million households. They can be divided into income classes – low income,
middle income, high income. Assuming total number of plates in a household for each income class,
total number of plates can be computed.
I: Okay, where else can you find plates?
All sorts of eating places such as restaurants, dining halls, roadside stalls etc.
I: How will you find number of plates there?
We can calculate total number of people who go out in Mumbai on a single day. Population –>
income classes –> average number of times people would go out in a month ( this can further be
narrowed down to weekdays and weekends ) –> divide the total monthly demand into per hour
basis (assuming restaurant remains operational 8hrs/day) i.e. for a single hour, this will help us
calculate the number of people eating together across Mumbai. These will be the number of plates in
restaurants.
I: Is that all?
All restaurants will have buffer plates for the next set of customers that come at the restaurant as
these plates will need washing.
I: Good point. Where else do you think you can find dinner plates?
Canteens, Cafeterias, Hostel Messes?
I: Where else? What about shops that sell dinner plates?
That is a good point. I did not consider this.
I: For the time being you can exclude those. But proceed with your calculation for the remaining 2
categories.

© Consulting Club, SPJIMR

GUESSTIMATES
Dinner Plates – Case Structure
– CaseCalculate
Problem: Structure
the total number of dinner plates in Mumbai.

Total dinner
plates in
Mumbai

Mumbai
Restaurants &
Households
Dinning Halls
(5 Mn)

Middle Income Lower Income Middle Income Lower Income

Rich (15%) Rich (15%)
Group (35%) (50%) Group (35%) (50%)

20 Mn 20 Mn 20 Mn
5 Mn*15%*8 5 Mn*35%*6 5 Mn*50%*4
people*15%*4 people*35%*2 people*50%*1
plates per plates per plates per
times a month = times a month = time a month =
household = household = household =
12 Mn per month 14 Mn per month 10 Mn per month
6 Mn plates 10.5 Mn plates 10 Mn plates
~50,000 per hr ~ 60,000 per hr ~40,000 per hr

Total plates = 150,000 plates. Assuming restaurants will keep over 30%
Total plates = 26.5 Mn plates.
buffer ~ 200,000 plates

Total dinner plates in Mumbai = 26.7 million

Key Insights
• Focus on covering all the parameters. It’s recommended to take a minute to think back, lay down the exhaustive structure and then discuss
• Always pay attention to the exact question being asked. In this case, the candidate had missed the dinner plates for sale – which is a significant part of dinner plates in Mumbai.
This could have been avoided. It’s a good practice to reiterate the question and confirm your understating

© Consulting Club, SPJIMR

GUESSTIMATES
Denim Fabric Consumption – Transcript
Problem: Estimate the amount of denim fabric consumed in Mumbai annually.

Sure. Denim is used to make jeans, bags, shirts etc. Since it is primarily used to manufacture jeans, Factor Upper Class Middle Class Lower Class
should I focus on the jeans market for the purpose of the calculation?
Segment Population
I: Yes. Go ahead. 10% 30% 60%
Share
I would consider the demand side approach; I will first divide the population into different income Segment Population
segments and look at the frequency of purchase of each segment. 2.0 6.0 12.0
(Mn)
I: This sounds like a good approach, You can assume the average denim consumption for jeans is 1.5
Males - 50% (Mn) 1.0 3.0 6.0
meters and average life of jeans to be 2 years. Consider male and female purchase frequency to be
equal.
Females - 50% (Mn) 1.0 3.0 6.0
Thank you, that is helpful! The formula for denim fabric consumption will be as follows:
Total fabric consumed = Addressable segment x frequency of purchase x fabric used per pair % of Males Wearing
100% 100% 50%
Jeans
I: You may proceed with your analysis.
% of Females Wearing
The population of Mumbai is 20 Million. Of this, 60% are lower class, 20% middles class and 10% 75% 50% 20%
Jeans
upper class. For the lower class, I will assume 50% of men and 20% of women wear jeans as fewer
Total Target Population
women wear jeans due to a more traditional dressing style. Also, we can assume that the lower-class 1.75 4.5 4.2
(Mn)
segment will buy 1 pair of jeans every 2 years.
Pairs Purchased
I: Okay, what about the other income groups? 4.0 2.0 0.5
Annually
For the middle and upper class, I will assume that 100% of men wear jeans since it is a common Total Jeans Purchased
7.0 9.0 2.1
attire and affordability is not an issue. For women, I will assume that 50% middle class and 75% (Mn)
upper class women wear jeans as western wear is more common amongst the upper class. I assume Average Consumption
that the middle class purchases 2 pairs a year while the upper class purchases 4 pairs. 1.5 1.5 1.5
(Meters/Jeans)
I: Go ahead with your calculations. Total Consumption
10.5 13.5 3.2
(Meters)
Let me draw a table to work out the number (Refer to the right half of the page)
I: We may stop the case here. It was nice interacting with you. Total Denim Fabric Consumed Annually = 27.2 Mn Meters

© Consulting Club, SPJIMR

GUESSTIMATES
Denim Fabric Consumption – Case Structure
Problem: Estimate the amount of denim fabric consumed in Mumbai annually.

Problem Scoping
End-Use Jeans Location Mumbai
Social Segments Lower Class, Middle Class and Upper Class

Notes Population of
Mumbai
Total Denim Fabric Consumed = Addressable segment x frequency of
20 Mn
purchase x fabric used per pair

Upper Class Middle Class Lower Class

(10%) (30%) (60%)

Frequency of Frequency of Frequency of

Purchase Purchase Purchase

Average Life of Average Life of Average Life of

Denim Denim Denim

Denim Denim Denim

Consumption Consumption Consumption

Key Insights
• The candidate clarified the assumptions initially, to simplify calculations
• The candidate could have considered age as a factor while determining the target population
• Calculation of the inventory of jeans in stores would have made the case stronger

© Consulting Club, SPJIMR

GUESSTIMATES
Dogs in Mumbai – Transcript
Problem: Estimate the number of dogs in Mumbai.

I will divide dogs into three categories – domestic pets, stray dogs and police dogs. I: That’s good. We can leave the calculations for now. For the police dogs, what would your
approach? Don’t give me the numbers, just tell me how you would do it.
I: Makes sense.
The area of Mumbai is 600 sq. km. I will assume there would be a police station every 5 sq. km. and
I will analyze the domestic pets first by segmenting the population of Mumbai by income groups.
then estimate the police dogs per station.
Shall I proceed?
I: Great. We can stop the case here.
I: You may go ahead.

The population of Mumbai is 20 million people. Of this, 20% of the population is high income, 50%
Factor
is middle class and 30% is low income. In the low income group, I assume 10% of families will keep Upper Class Middle Class Lower Class
(Domestic Pets)
a pet since expenses such as food, grooming etc. may not be affordable and I will assume 4 people in
one family. Since families could own dogs, cats, fish etc., with equitable distribution, 25% of pets will
be dogs. This gives me a total of: Segment Population
20% 50% 30%
5 Mn families x 30% of the pop. x 10% owning pets x 25% dogs = 0.04 Mn pets for the lower class. Share (20 Mn)

I: Okay and what about the other income groups?

Segment Population (Mn) 4 10 6
For the middle income segment, I will assume 50% are pet lovers who will own cats, dogs, fish and
other pets. Assuming an equal distribution, 25% will be dog owners. This gives me 5 Mn families X
50% X 50% X 25% = 0.31 Mn pets for the middles class. Similarly for upper class, assuming 60% Segment Families (Mn) 1 2.5 1.5
pet lovers, there will be 5 Mn families x 20% x 60 % x 25% = 0.15 Mn pets for the upper class. The
total domestic pets is 0.15 + 0.31 + 0.04 = 0.5 Mn domestic pets.
I: Great analysis. What about stray dogs? % Having Pets 60% 50% 10%

Stray dogs generally live in groups of 2-3 and are territorial about their space. I will assume they
occupy 10m x 10m or 0.1 sq. km. of space.
% Dogs 25% 25% 25%
I: I see where you are going with this. Would stray dogs be equally distributed across the city?
Stray dogs will be higher in areas with a higher population density such as slums due to easier Total Pet Dogs 0.15 0.31 0.04
access to food. I can break up Mumbai into 3 parts – areas with high dog population (30%) such as
near garbage disposal areas and slums which will have dogs in packs of 5. Residential societies with
medium population density (50%) which will have dogs in packs of 3 and low density areas such as
Total Pet Dogs = 0.5 Mn
Tech parks (20%) which will have packs of 2 dogs.

© Consulting Club, SPJIMR

GUESSTIMATES
Dogs in Mumbai – Case Structure
Problem: Estimate the number of dogs in Mumbai.

Notes
Number of pet dogs = No. of families x % who own pets x % of pets that are
dogs x dogs per family
Number of dogs
in Mumbai

Domestic Pets Stray Dogs Police Dogs

Low Income Low Density Area

Medium Income Medium Density Area

High Income High Density Area

Key Insights
• The candidate’s method of segmentation of dogs was good. Adding dogs in shelters would have made the segmentation more exhaustive.
• The candidate missed out on estimating the dogs per family in the equation – there was an implicit assumption of 1 dog per family.
• It’s advisable to run your assumptions by the interviewer as and when they are made

© Consulting Club, SPJIMR

GUESSTIMATES
App-based Two Wheeler Ride Sharing Service – Transcript
Problem: What do you think would be the demand for a two-wheeler only app-based ride sharing service in the Bangalore Market?

If I understand the question correctly, you would like me to estimate the size of the two-wheeler Only about 40% of people, all belonging to the age group of 18-35 are the primary target audience,
ride sharing market, in terms of the number of users in the Bangalore market? and a small percentage of people belonging to the 35-65 age group, would end up using this product.
I assume this age group to be 15% of the target market. This gives us about 2 million people.
I: Yes. And since you are not from Bangalore, assume the population.
Further, we should divide on the basis of usage of these 2 million people, because a single person
Essentially, we have to look at the market wherein single riders will be using the application. So, I ride sharing service would only be used in a few cases.
will be starting from the overall population, try to arrive at the number of people who would be
I: Which are?
using the app, and then further delve into the usage in multiple situations to determine the demand
for such an app. So, this could be, commute to the office as the maximum usage, low probability of being used when
going out for leisure activities - because a lot of people end up being in groups, and finally, running
I: That is correct. Please proceed. errands which don’t require space.
Since I don’t have an idea about the Bangalore population, I am assuming the population to be close I: I think that is good enough. We can stop here. One question though. How would your answer
to 15 million, and for the purposes of the calculation, would be close to 20 million. I know this change if it was a two-wheeler ride sharing service, where you get the bike at your doorstep and two
number is on the higher side. people can ride.
I: Seems about right, for this purpose. This now requires the rider or at least one of the riders to know how to ride a bike. Hence, the usage
Now, we can divide this population basis the income, as app-based aggregator usage, requires as per income group would still be the same. But amongst the 2 million people who would have
smartphones to operate. The population can be divided into 30% BPL, 30% lower Income, about otherwise used such a product, we would have to further see the number of riders, which might be
30% in middle income and 10% in the upper income. Now, since its a two-wheeler based ride close to 40%.
sharing service, we will have to discount the upper income population where the usage would be I: Fair enough. Thank you for the analysis.
minimal. Further, we will be discounting the BPL category, since they would typically not have
access to smartphones to use such an app. Even within the lower income category, I assume only
about 25% of them would actually be using smartphones. Hence, only about 6 million people
belonging to the Middle Income, and 1.5 million people belonging to the lower income could be the
target income groups for the app.
I: So far so good.
Going further, we would have to divide these 7.5 million people, as per the age demographic. About
65% of India, is below 35, and close to 20% are above 65. Assuming people above 65 would have a
problem riding a two-wheeler, especially using a single person transport, and their problems with
smartphone usage, we will have to discount this demographic. About 25% are children below 18
and would rarely be using such a service, hence, we would have to discount those. This leaves us
with two age groups – 18 to 35 years, and 35 to 65 years.

© Consulting Club, SPJIMR

GUESSTIMATES
App-based Two Wheeler Ride Sharing Service – Case Structure
Problem: What do you think would be the demand for a two-wheeler only app-based ride sharing service in the Bangalore Market?

Problem Scoping
Segment People in the age of 18-65, Smart Phones Users Location Bangalore
Social Segments Lower Income, Middle Income, High Income & BPL

Population of Bangalore
Notes (20 Mn)
Market Size = Population x % of population in social bracket x % of
population in age bracket x % of population using smartphones x % using High Income Middle Income Lower Income
BPL (30%)
two-wheelers (10%) (30%) (30%)
6 Mn
2 Mn 6 Mn 6 Mn

Smart Phone Smart Phone Non

Users (25%) Users (75%)
1.5 Mn 4.5 Mn

7.5 Mn

Age <18 (25%) 18-35 (40%) 35-65 (15%) 65+ (20%)

Groups 1.875 Mn 3 Mn 1.125 Mn 1.5 Mn

40% 70%

2 Mn

Key Insights

• The candidate followed a multi-level structured approach

• The candidate accounted for quantitative and qualitative factors

© Consulting Club, SPJIMR

GUESSTIMATES
Replenishment of Currency – Transcript
Problem: The Government of India expected it would be able to meet the demand of new notes within 4-5 months post demonetization. It hasn’t happened as planned. How much
time would it take to replenish currency (demand of notes) in India?

The ask is to estimate the time required to replenish currency notes after demonetization. Is that I: Okay.
correct?
Let me draw a table for computation (refer next page). Assuming 20% people below poverty line,
I: Correct! we can ignore them for this calculation. Next, I will assume 5% in the rich income bracket, 15% in
First I would like to divide the population based on income – BPL, lower middle income, upper upper middle, 20% in middle & 40% in lower income group. Total households in India are 350 Mn.
middle income & high income. I: Okay, go ahead.
I: That sounds fine. There is demand for 9 bn Rs. 500 notes and 1.3 bn Rs. 2000 notes. RBI can print 16 billion notes
For every cluster I would then assess the average monetary assets owned and then determine the annually in two shifts. Assuming 70% capacity will be used for Rs. 500 & Rs. 2000 notes, ~11 bn
percentage of monetary assets in bank versus monetary assets in physical form. notes can be printed in a year.

I: Fantastic! I: Great.

The per capita average wealth in India is approximately $14,500. Based on this figure, I would So the currency demand is approximately 11 billion, and 11 billion is the annual printing capacity
determine the wealth of each class. Then assess the percentage held in each asset class in monetary for the Rs. 500 and Rs. 2000 notes. The currency can be replenished in a year, that is, 7 more
and real assets. Post that, I’ll further evaluate the percentage of monetary assets held in cash. months would be required.

I: Go ahead with the calculations. I: That’d be all. Thank you for the comprehensive analysis.

Total cash assets = Average wealth x % in monetary assets x % held in cash (All numbers in INR)
• Rich: 50,00,000 x 30% x 15% = 2,25,000
• Upper Middle Class: 20,00,000 x 15% x 25% = 75,000
• Middle Class: 10,00,000 x 10% x 35% = 35,000
• Lower Class: 5,00,000 x 5% x 45% = 11,250
I: Why do you think lower income group will have more percentage of assets held in cash than other
classes?
The banking facilities & infrastructure in rural areas are not that developed. So I have assumed they
will hold a greater chunk of their assets in cash.
I: Great!
Now assuming that 40% of the monetary assets will be held as Rs. 500 notes and 20% as Rs 1,000 Note: This case makes assumptions on facts and figures at each step. In the interview, a candidate is
notes. Then based on the number of households in each category, total demand of Rs. 500 and Rs. not typically required to know these many figures; most of the times an approach suffices. We have
2000 notes can be computed. worked the calculations at our end to demonstrate the complete working of the case.

© Consulting Club, SPJIMR

GUESSTIMATES
Replenishment of Currency – Case Structure
Problem: The Government of India expected it would be able to meet the demand of new notes within 4-5 months post demonetization. It hasn’t happened as planned. So, how much
time would it take to replenish currency (demand of notes) in India?

Category Rich UML ML LC

20% are assumed to be below poverty line and
A % Households 5% 15% 20% 40% are excluded
B Households Mn 17.5 52.5 70 140
Wealth per household bifurcated for each class:
C Total Assets 50,00,000 20,00,000 10,00,000 5,00,000 Assumed that Rich will have 5X, UML 2X, ML 1X
D Monetary Assets 30% 15% 10% 5% and LC 0.5X of average wealth ($14,500 x 70 ~
INR 10L)
E Cash Assets 15% 25% 35% 45%
F Total Cash Assets 2,25,000 75,000 35,000 11,250 Total cash assets per household = C x D x E
G Rs 500 (40%) 90,000 30,000 14,000 4,500
H Rs 1000 (20%) 45,000 15,000 7,000 2,250
I Total Demand - 500 Note 3,150 3,150 1,960 1,260 Total Rs 500 notes demand = [G/500] x B
J Total Demand - 2000 Note 394 394 245 158 Total Rs 2000 notes demand = [H/2000] x B
Total Currency demand for
K 10,710 Mn Sum total I & J
500 & 2000
RBI Annual Currency capacity for
L 11,200 Mn
500 & 2000
M Time required ~12 months [K/L] x 12 months

All numbers in the above table are in INR

Key Insights
• Clarify assumptions with the interviewer; at times you may be asked to clarify what you are assuming in which you must provide a logical backing
• In a number-heavy case, round-off the numbers in such a way so as to simplify the calculations.

© Consulting Club, SPJIMR