Alle Fragen zum Tag Market sizing - Consulting Q&A - PrepLounge.com

Hello,

You could approach the problem in 2 different ways:

1. Estimating the number of hotels and airbnb apartments in Paris and their occupancy (then you can increase it by a small percentage to include people that are hosted by friends/relatives)
2. Estimating the number of flights and trains that arrive in Paris in a given day. This way is more complex because you have then to estimate the percentage of passengers that are tourists. Furthermore, you are not considering tourists coming by car

I would follow the furst methid and then propose to do a sanity check of the result using the second one.

Does it make sense?

Best,
Luca

Hello,

I would first of all asking if the bar has both tables (how many?) and counter service. I wouldn't worry too much about competition, since in the center of Lond it's fair to assume that you have enough demand. I would then proceed as following:

• Estimate the number of seats of the restaurant and the average time/customer
• Estimate how many customers per hour we can serve with the actual number of cash registers

You do this because the assumption is that the bottleneck of the process is the line that there will be to pay during the peak hours. You have to consider table service on top of that, since you bring the bill at the table.

Once that you have the capacity of the bar, you can divide the daily working day into time bands and make some assumptions on the "occupancy" of the bar in the different moments of the day. After this, you have just to multiply the number of bills by the average expenditure per client.

Please notice that:​

1. Even when your bar will be "full", you don't have to consider all the seats of the bar, because you can have for example 2 people sitting at a table for 4.
2. You could consider apart weekends and seasonality within a year. Don't forget to ask if you will be open 365 days/year.

Hope it helps,
Luca

Hi,

1) First of all, there are 2 ways to structure market sizing:

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

2) You should learn the key market sizing techniques:

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

3) You should learn the key tools:

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

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

5) Practice 10-15 cases and you'll be fine. Almost all casebooks have good market sizing examples and the solutions

Feel free to PM for clarifications

Good Luck!

Hi,
I would approach it with a household perspective:
- umbrellas per household
- life of umbrellas
- umbrellas purchased because forgotten at home
- umbrellas purchased in the US by foreigners (tourists, businessmen)

Best,
Antonello

The number of hairdressers depends on three main factors: # of haircuts done per year, time spent per haircut and capacity of each hairdresser.

Number of haircuts: Number of haircuts depends on US population that gets a haircut and how often people get a haircut. I will first tackle the population. Obviously, people aged between 0-10 do not get haircuts. Even if they do, it is exceedingly rare. I assume health expectancy is 80 years and population is evenly distributed among each groups. That means, only 87.5% of the US population gets haircuts. I assume there are 160m males and 160m females in the US. That means, 140m males and 140m females gets haircut each year. I also assume males get a haircut once every month and females get once every three months, which means 12 haircuts per year for males and 4 haircuts per year for females.

This means, total number of haircuts per year is 140m * 12 = 1.68 billion for males and 140m * 4 = 560 million for females.

Time per haircut: I will assume a haircut per male will take 30 minutes while it will take 1 hr for females.

Capacity per hairdresser: I will assume there will be on average four hairdressers per shop and each will work for 40 hours per week. Assuming they will work for 50 weeks per year, capacity of each shop is 8000 worker hour.

Obviously, the number of hairdressers is total hours worked / capacity per hairdresser.

Total hours worked: 1.68 billion * 30 minutes (for males) + 560 million * 1 hr (for females) = 0.84 billion worker hours (for males) + 0.56 billion worker hours (for females) = 1.4 billion worker hours

Capacity per hairdresser: I calculated it above to be 8000 worker hours per year.

Number of hairdressers: 1.4 billion worker hours per year / 8000 worker hours per year = 175,000.

There are approximately 175,000 hairdresser shops, or 700,000 individual hairdressers working in the US.

Hi,

I would split Spain into 3 buckets and allocate % of the area:

1. Urban areas
2. Rural areas
3. National parks / uninhabited areas

1. Urban - you take an average size of the block and find out the number of roads between them (e.g. total Urban length from East to West of 1000km, each block is 1 km, thus 1000 roads from E to W. Same with S and N)

2. Rural - you just assume much lower density than in Urban (e.g. 50x lower)

3. National Parks - will be even lower

Best

Hello!

I would calculate how many balls are purchased per year by two segments: (1) individual players and (2) academies/centers.

1. Calculate # of players and # of academies (estimation: X% of the people play tennis, there is X academies every Y people)
2. Calculate, on average, how many matches they play monthly/yearly (estimation)
3. Calculate how much a ball lasts (estimation)

Cheers,

Clara

Hi,
I think the last approach is better. Market sizing cases like this are thought to be solved in 20 minutes, therefore you have the time to be more detailed in the estimation and add at the end qualitative considerations to adjust the number with percentages. I would:

• Start from the London population
• Calculate the households estimating the n. of people per household (e.g. 2)
• Calculate the number of residential buildings assuming the n. of houses in a building (e.g. 20)
• Calculate the number of blocks estimating the n. of buildings per block (e.g. 12)
• Calculate the number of intersection
• Calculate the number of traffic lights, assuming 4 main traffic lights per intersection + 8 pedestrian traffic lights (if needed to be included)

Hope it helps,
Antonello

Hello!

First of all, I would clarify if we are talking about big/olimpic stadiums (e.g., Allianz in Munich or Bernabeu in Madrid, so the ones of big soccer teams) or total numer of them (including the ones from schools, etc.). This changes the whole story.

For the 1st scenario, I would think how many 1st league clubs you get in your country and divide it into the total population, so you can have a ratio to be applied.

For the 2nd one, more complex, I would make some buckets (e.g., 1. big stadiums for 1st league clubs, 2, medium stadiums for non-professional clubs, 3. schools and municipal, etc.) and repeat a similar approach for each of them. Like this you break the problem into "solvable" little pieces.

Hope it helps!

Cheers,

Clara

It seems to be a standard market sizing that you can tackle starting from total american population, as suggested by others.

If you want to impress you interviewer, once that you have done your estimation, propose to do a sanity check of the result following a bottom up approach. One way could be to list the K-12 schools of your town, calculate the ratio schools/population and apply this ratio to the national population.

Hope it helps,
Luca

I would follow the following process:

• US population: 330M
• US population "per year": 330M / 80 years = 4.1 M / year Here you are making the hypothesis that the average life expectancy is 80 years
• Babies born in a day: 4.1 M / 365 = 11 k

Then you can state some aspects that cour estimation and add more complexity. For example:

1. You could consider that natality is going down, adding a factor to decrease your number
2. You can consider the natality distribution within a year: historically most of the brths are between July and September

Hope it helps,
Luca

Ciao valentina,

The website doesn't let you to post a personal email, the message is crypted. Feel free to contact me if you need more materials.

Buona serata,
Luca

A few thoughts on this:

• Price differs by airline: full service carrier vs. low cost airline
• Price differs by flight duration: short flight vs. long-haul flight
• Price differs by booking class: first class, business class, coach
• Price is dependent from available alternatives (trains etc.) and geography
• Price is connected to free disposable income and willingness-to-pay

Calculation scheme:

• Let's assume any European country.
• Price indicated as round-trip tickets.
• European flights:
  • Coach:
    • 50 EUR for low-cost airlines (Ryanair)
    • 150 EUR for full service carriers (Lufthansa, BA)
  • Business:
    • Let's assume low-cost airlines do not have business class.
    • 300 EUR
  • No first class on European flights
• Medium/long-haul flights:
  • Coach:
    • Low-cost airlines: 600 EUR
    • Full service carriers: 1,200 EUR
  • Business:
    • Full service carriers: 5,000 EUR
  • First:
    • Full service carriers: 10,000 EUR
• Let's talk about proportions:
  • Percentages by Coach/business/first:
    • European flights: Assume an average jumbo jet has 40 rows, 5-6 rows are dedicated for business travellers. Which leads to:
      • Economy: 85%
      • Business 15%
      • First 0%
    • Medium flights:
      • same as above
    • Long-haul flights: Consider 777,787,A380: 600 passengeer seat capacity on average per flight. 40-60 passengers in the business class (10%). 5% in the first class. Which leads to:
      • Economy: 85%
      • Business 10%
      • First 5%

• European flights are dominated by low-cost airlines, therefore:
  • Average coach price will slightly above the price for the low-cost airlines, e.g. 80 EUR
  • Low cost carriers provide no business class, therefore the business class ticket will cost on average 300 EUR
  • The average price on European flights will be therefore: 80 * 0.85 + 300 * 0.15 = 68 + 45 = 113 EUR
  • Sanity check: average price on German railways is between 50-200 EUR
• Medium and long-term flights are more dominated by full service carriers
  • Average coach price will be somewhere between 600 and 1,200 EUR. More on the upper bound: 1000 EUR
  • Business class ticket will cost on average 5000 EUR
  • First class: 10000 EUR
  • Which leads to 1000 * 0.85 + 5000 * 0.1 + 10000* 0.05 = 850 + 500 + 500 = 1850 EUR
• Consider now the split between European and medium/long-haul flights. Local flights have significantly higher frequency at the airports, therefore we assume 70% of flights are European flights. Which leads to 113 * 0.7 + 1850 * 0.3 = 79.1 + 555 = 634.1 EUR

Hi Jorgie,

I like the way you approached the problem but I would suggest an approach that is a mixture of the 2:

1. Calculate the population that lives in cities that have the facilities to charge electric cars (Your idea of considering only cities is similar, but this condition could be even more restrictive)
2. Calculate how many cars have been replaced since when electric car have really entered the market (similar to what you did in the second approach)
3. Try to estimate the percentage of electric cars bought, dividing the population per age groups (It's more likely that an electric car is bought by a 25 years old than a 60 years old)

Thank you for sharing your experience!

Hi,

Your question is unclear, could you pls clarify?

Thanks

Hi Anonymous

This seems to be an interesting ask and can be a bit intriguing. You may get this if you are an experienced candidate and have something like this on your CV.

I hope you are aware of what Internet of Things (IoT) is. If not then , IoT is a system of interconnected interrelated computing devices that measure specific paramters/function and have the ability to transfer data over a network without requiring human interaction.

Basis this a candidate should first identify the areas of application of IoT and then monetization of IoT services:

1. Application of IoT : This can be very diverse and exhaustive. But it would be good for a candidate to detail the segments and appraoch and then check with the interviewer which segment he would prefer to focus on. Also note that the adoption of IoT services would vary across each segment based on the benefit of having IoT services.

Consumer Segment : (a) Individual : Wearable Devices (b) Home : Home Solutions like Smart LIghting, Surveillance, Smart Lock etc

Enterprises Segment : (a) Manufacturing (b) Logistics and Transportation (c) Healthcare (d) Smart Cities - Government (e) Automotive (f) Real Estate and many more

2. Monetization of IoT : There are broading 3 ways to monetize IoT services :

Device Revenue : Generated from the sales of IoT devices as per the identified applciatoin

Connectivity Revenue :The revenue from internet connectivity. Usually this is very nominal less than 8% and declining, also many devices are connected to already present Wifi so for ease the candidate should just mention this and not include this.

Value added Revenue : this may include revenue from value added services like monetizing data captured through devices as data as a service (DaaS), for up sell and cross sell or Platform as a Service (PaaS). During the interiew it would be good to mention these and not calculate the value impact as these too can have tiered monetization strategy - Freemium, Transactional, Volume Based etc.

So for easy the candidate should look at revenue from the sales of devices only.

Thus you can calculate the Market Size of IoT by application or function = # of Devices Sold Annually * Price of Device

Hope this help. For practice I would suggest (1) you work of calculting the size of few segments and (2) do some research on IoT. Please feel free to reach out to me in case of any queries.

Hi

Your questions is a very valid one. However, the answer lies in how you address the problem.

1. Addressing a Sizing Problem

A typical market sizing problem can be addressed from either Demand Side or Supply Side. For either of the approach you need to indentify it's drivers.

In the demand side appraoch you would have to address the drivers of demand of consumption/user/buyer base. This appraoch is good in straight forward cases how many LV bags are sold in Indonesia. Size of beer market in the US etc.

In the supply side approach you would have to look to estimating the total capacity/production output. This appraoch is good for tricky cases like how many Red BMW cars in Germany. Number of Passengers communting daily via London Undergroung.

Candidates perform well when they are asked to size the market for conventional industries like consumer goods, food and packaged, automotive, electronics etc. However they face issue with industries like business services, digital services, technology related new age services.

One of the main reasons, I have encountered while interviewing candidates is that they are not well aware of the business and operating models of industries like these.

Thus, the first thing as a candidate you should do is to learn at a high level about how different industries operator. There is also no harm in asking the interview to tell you about the industry in question - revenue streams, operating model, etc.

2. Making Assumptions

Once you have identified your approach and the drivers, it should be easy for your to determine parameters for that driver and respctive degree of assumption

However, the trick lies in your logic and reason of assuming a certain parameter. Easier said than done

- Always try to classify the parameter into categories of reasonings - economic, customer preferences, substitutional, complementary, operational.

- Leverage your knowledge across industries to arrive at a suitable number

- Always cite examples to substantiate your assumption.

Now lets take your example of consulting market in the US. This can be aptly sized using the supply side appraoch. And I would use my set of assumptions which may vary candidate to candidate.

Revenue stream of a management consulting firms is mainly billing of the consultants on client engagements. Thus lets determine the number of consultants and the revenue they would generate.

- Assuming Top 10 firms command 70% - 80% of the Management Consulting market in the US. MBB + Big 4 + Niche Firms.

- On an average each of these firm has 2 offices across each state in the US.

- Each office houses close to 350 management consultants on an average. Big 4 firms would have larger number compared to niche.

- Thus, total consultant strength across top 10 firms = 2* 52*350*10 = 364,000 consultants.

- Average industry utilization per consultant - 80% with an average billing rate of $ 300 dolalrs per hour with 40 hours a week for 50 weeks/year.

Why 80% utilization, because junior consultants have 95% target and as you rise up the ladder that target reduces so a blended 80% is relatively close to actuals. Why $300 /hr - juniors are billed around $100 - $200 full time where as principles and above are way to high but they work across engagements. So blended $300.

The idea is to communicate your thinking not focus on the accuracy of the number, once you are in the real words you would have benchmarks, actual data, and excel sheet to capture the averages :-)

- Therefore, Market captured by top 10 firms = 364,000 * 80% * 40 hr/week * $200/hr *50 weeks/year = 174 Billion

- Thus, the total market value of management consutling in the US = 174 Billion /70% = ~USD 250 Bn

I hope this helps providing you the approach and essense of market sizing problems. In case you would like to know more about it please feel free to contact me.

All the best.

Hi,

The general framework for pricing is: Cost base - Value based - Competitor based - pricing strategy

1. Cost based - you actually check what are the costs and apply standard industry / target markup

2. Value based pricing can be done in 2 ways:

• For existing products you compare the value proposition and features of your product vs. the VP of your competitors. If you have a significant difference in value prop - you have to define how much value you propose to the customer in $ terms.
• For new products you have to calculate the value of the closest alternatives and think how much additional value we provide by replacing them. Basically it's the customer WTP - the willingness to pay. Think of the discount airlines compared to trains or buses

3. Competitor based pricing - basically it's benchmarking against competitors. Make sure you take into account the segment (i.e. in premium higher price may be the proxy for quality)

4. Pricing strategy - here you define how you will price the product taking into account 1,2,3 and your company strategy. Maybe you decide to have a zero margin if you can crossell other services. Or maybe you would like to subsidize to win the competition. Also think of price differentiation and having different pricing tiers (e.g. basic, premium or even fremium) and how it helps to drive price perception and fulfill strategic goals

Good luck!

The question has a bit of a disconnect. The amount of money given to grow trees does not have to correlate to the number of trees that can be planted.

To answer the question on number of trees that can be donated, here is an alternative approach

1. Value donated to plant trees cannot exceed value of all charitable donations in the US
2. Total charitable donations (personal) can be a % of GDP. So assuming GDP of 20 Trillion, let's say 1% of GDP is donated to charitable causes - that is ~200Bn dollars a year
3. Next we can say that of that, causes related to the environment make up ~5% - 10% or so (most donations are to religious organizations), have some sort of a rationale for that
4. At 10% this is roughly $20Bn a year
5. Of that people donating specifically to plant trees would be a very small sliver - again 1%-5% is not an unreasonable number when you are competing with things such as national parks, WWF, PETA etc.
6. So that gives is ~500M per year. Divide that by the cost of planting a tree - maybe $25 seems reasonable - giving us 20M trees a year potentially.
7. Of course this does not mean you can plant 20M trees, but the question asked for number of trees donated not planted
8. Also this assumes you get 100% of all the donated money whereas in reality you will get a very small fraction of it

Hope this helps,

Udayan