Consulting Guesstimate Cases

The questions in this chapter are representative examples of Guesstimate Cases. In essence, these questions involve a short case or analysis in which you must try to estimate a value you probably had never considered estimating before. The key to these types of problems is to break them down into logical pieces, walking the interviewer through how you are approaching the problem, and then try to determine appropriate assumptions for the value of key figures in each of those pieces.

Guesstimate cases are generally used as part of a first-round interview to warm a candidate up for a full Business Situation Case later in the interview process. Guesstimate cases are also used as a way to test quantitative ability, numerical thinking and judgment, and an to arrive at reasonable assumptions.

With some practice, guesstimate cases are not difficult and once you get accustomed to them, they are actually quite fun, even though at first they might seem a little scary. Guesstimate cases will also form part of many full case interviews, so it is important to master them.

Here are some quick examples of Guesstimate Cases so you can get a better sense of what we are dealing with:

Guesstimate Case Study Examples

What is Air France’s weekly revenue for the Paris-to-New York leg?
What volume of beer is sold at an average LA Lakers basketball game?
What is the monthly revenue of your hair salon?
How many school teachers are there in Chicago?
What is the monthly profit of your favorite restaurant?
How many flat screen televisions have been sold in Australia in the past 12 months?
How many iPhones are currently being used in China?
Are there more iPhones in operation in the U.S. or in China?
How many trees are there in New York City?
How many avocados are used to make guacamole per day in Mexico?
What is the Revenue for the smartphone market in India?

Here are a few key points to keep in mind when answering these questions:

There is rarely, if ever, a verifiably correct answer or one way to tackle a Guesstimate question. The goal is to make reasonable, logical assumptions and if you are wrong on your assumptions, that is totally fine—after all, some of the figures used to make the estimation are quite obscure and you do not have access to the data during the interview. The interviewer is aware of that. If you can logically explain how you arrived at your assumptions, and highlight which of your assumptions seem the most important to research, you’re doing well. (As an example, for the Chicago teachers question, you might not know the population of Chicago to start with—it is about 2.7 million. However, even if your estimate is 1 million or 10 million, your logic in estimating the correct answer can still be sound.)
As you progress through the case, round assumption numbers up and down so that the calculations are less difficult (also use percentages where applicable that are easy to calculate without a calculator). In the above example, use 3 million even if you know that Chicago had 2.7 million people. Keep the numbers simple, as the interviewer will be quite impressed if you get a result that is reasonably close to his or her estimate.
Importantly, as with all cases, be very structured in your approach and write out your numbers and assumptions in a clear manner. Be sure to clearly label the units and variables. This is also important if the interviewer follows up with the question “What if we changed that assumption?” This question is not an indication that you’re bombing the Case—it might simply mean that your initial assumption was off, or the interviewer might be testing your ability to think flexibly under pressure.
Stay composed, as these cases should not trip you up, especially with sufficient practice. They are all quite similar after you have done a few practice questions.
Communicate your process and answer succinctly and articulately. Do not be afraid to acknowledge mistakes during the process—it shows maturity and composure.
Virtually all of these questions can be grouped into Top-Down or Bottom-Up (often referred to as Ground-Up) questions. In some situations, they can even be both.

Top-Down Questions: Questions that involve starting with an entire population (in other words, the “top” level) and then breaking it down until you arrive at an answer. For example, consider the question about the number of schoolteachers in Chicago. A simple way to approach it would be to start with the population of Chicago, then estimate what percentage of the population is of student age, and then estimate the number of students per class. Using this, you would arrive at an estimate of the number of school teachers, because at any given time there is (generally) exactly one teacher per class. To show your ability to be creative and think outside the box, you could also attempt to account for retired teachers and substitute teachers.
Bottom-Up (Ground-Up) Questions: For these questions, 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. For example, consider the question pertaining to the monthly revenue of a hair salon. In this case, we’d recommend you work out the revenue for a week and then multiply that by four (or if you are quick at multiplication, 4.3). You could start with an assumption regarding the average price per client visit, and then estimate weekly volume by assuming the number of chairs in the salon, the number of hours it is open per week, and the average number of clients chair per hour. To show your ability to be creative and think outside the box, you could also add revenue for hair salon products sold. You may also want to break the estimate into male and female clients, as male clients at hair salons tend to spend less money per visit but also take less time on average.

Step-by-Step Approach to Guesstimate Questions

Step 1: Ask questions if the request is not totally clear. Take a minute or two to decide how to structure your answer—this is the period in which you determine whether it is Bottom-Up or Top-Down question, how you will break the question down into pieces, and the assumed values you will use for each of those pieces. Though you can interact and direct questions to your interviewer, in these cases it is not useful to ask for help on your assumed values as often his or her guess will be an educated guess, just like your own guess. (And even if you assumed value is off the mark, it is far less important than being able to structure and articulate the response, correctly calculate an answer based on the assumed values, and provide insight for which assumptions you’d ideally like to research further.)
Step 2: Mentally double-check the “Guesstimated” values that you will apply to the pieces of the calculation you developed in the first step. Note that in Guesstimate Cases, a population figure is very commonly part of the solution process (whether it be the U.S. population, another country’s population, a city population, or passengers on an airline). Often you will need to segment this population—in the Chicago schoolteacher example, we would need to estimate the number of children of student age. Keep in mind that any population be segmented in a number of ways. In the context of using a Top-Down approach, the key segments to use typically are:

Gender
Age
Geography
Income & assets

Step 3: Perform the calculations to arrive at an answer. Remember to use estimation to make your math easy. Also, if you find that you are estimating multiple figures to make the math easier, try to balance “rounding up” with “rounding down.” For example, if an answer involves multiplying 44 by 5,300, you will get a more accurate answer with 50 × 5,000 = 250,000 (one rounded up, and the other rounded down) than 40 × 5,000 = 200,000 (both rounded down). (The exact answer is 233,200). This effect tends to get larger if you have 3 or more numbers to round, so always be mindful of the degree and direction to which your estimate might be off due to rounding error.
Step 4: Identify any additional creative elements that could further refine your answer, if relevant. For example, in the hair salon case, we saw that Revenue might be boosted from the sale of hair care products. Additionally, you should tell the interviewer which pieces of your estimate seem most vital to research further—those that seem to have the highest degree of sensitivity in determining the correct answer, or those in which you have the most uncertainty as to whether your estimate was reasonably close. In the Chicago schoolteacher example, if you had no sense whatsoever of the population of Chicago, you might indicate that you’d like to look up that number and that your answer depends heavily upon the value you chose for that assumption.

Guesstimate Case Example #1

How many cups of coffee were consumed in the United States in the past week?

Start by clarifying the question, then identify the variables to apply to this problem.
Number of cups in the past week: This equals number of cups per day × 7 (for 7 days per week). Mention to the interviewer that your assumption treats each day equally, although you would get points for mentioning that there are likely fewer cups consumed on the weekend, as people are not in the office and in general might feel less of a need to drink coffee.
Percent of the population that drinks coffee: this would be an educated guess. Assuming 300 million people in the U.S., we could further assume that 20% are children that (we hope) do not drink coffee. We could also guess that another 20% of the population does not drink coffee at all (perhaps they prefer tea or other beverages, or just water).
Number of cups per day: here our guess is that of the remaining 60% of people, half drink 2 cups per day, a quarter drink 4 cups per day, and a quarter drink 1 cup per day. This averages out to 2 × 0.5 + 4 × 0.25 + 1 × 0.25 = 2.25 cups per coffee drinker per day.
Therefore the calculation is:

60% × 2.25 × 300,000,000 = 405 million cups each day
405 million cups × 7 days per week = 2.84 billion cups per week (you could round it to approximately 2.8 billion cups)

Note: the interviewer could then ask questions around how many cups are drunk at home or the office versus bought from a store, or other similar variations. The interviewer might also ask your thoughts on coffee trends and also how much revenue this would mean, etc.

Guesstimate Case Example #2

What was the revenue for flat screen televisions sold in Australia in the past 12 months?

Your first thought might be, “Why Australia? I am not applying for the Sydney Office of McKinsey!” Interestingly, more and more questions have a global component as Consulting becomes more global in nature (for example, advising firms on entering or performance in foreign markets, or working directly with foreign clients). While you will not be expected to know Australia’s population, your estimate will say something about your ability to think and your logic.
Start by clarifying the question, then identify the variables to apply to this problem.
Population of Australia: Approximately 23 million people. (Editor’s note: I once received a Case Study question involving this figure and my estimate was far too high, but I still did well on the case because my thought process was transparent and otherwise accurate.)
Assume that the average household is 3 people. It is worth noting that families probably have more than 3 people, but this is balanced out by people living alone, such as students and young professionals. Here is a good example of rounding: you can say 8 million households (which is a little more than 23 million ÷ 3).
Assume households replace their televisions every 4 years. The interviewer might say that “seems reasonable” or “you should try a higher/lower number.” You might also note that you believe there was nothing special about last year in terms of television sales—no major product innovations, sluggish but growing economy, etc.
Assume an average of 1 flat screen television per household. Some households might not have any, but others may have 2 or even 3.
Therefore, (8 million households) × (1 TV per household) ÷ (4 years/purchase) = 2 million televisions purchased in the past year.
Assume an average sale price of $600. Again, the interviewer might say that “seems reasonable” or “you should try a higher/lower number.” He or she may even ask you to break this down into groups, such as “high-end flat screens” and “smaller flat screens” with different average prices. However, $600 seems like a reasonable average across higher-end TVs, which might cost more than $1,000, and smaller flat screen TVs, which can sell for $200 or even lower.
Therefore, 2 million × $600 = $1.2 billion annual Revenue for television sales in Australia.
To show your creativity and business thinking you could have also discussed the market for resold used televisions and the Revenue that comes from that (this might reduce your estimate of the market size by reducing the average sale price; if the interviewer insists that the question should only include new televisions, then keep the average sale price constant but reduce the number of purchasing households each year).
The interviewer might then ask you a follow-up question, such as “discuss your thoughts about trends in television sales,” just to see how you think on the spot.

Guesstimate Case Example #3

How many iPhones are currently being used in China?

Start by clarifying the question. In this case you might want to confirm whether the question is how many iPhones are in operation, or how many are being used at that this current moment. (The interviewer would probably explain that he or she means “in operation”.) Once the question is clarified, identify the variables to apply to this problem.
Population of China: Approximately 1.4 billion people.
There are several different approaches from this point; one approach is to make assumptions around the number of people that can afford iPhones rather than considering the number of households.
Based on very basic knowledge of China, even though the country is experiencing extraordinary economic growth, you might assume that the majority of the population is still very low-income and cannot afford an iPhone. Thus, you might estimate that 20% of the population could afford an iPhone.
Therefore, the total potential market size is 20% × 1.4 billion = 280 million iPhones.
What percent of this total market size is penetrated? There are many competing products that are cheaper, but perhaps you’ve read that the Chinese are very brand-focused and that Apple has an extremely trusted and desirable brand in China. Therefore you estimate that 20% of this segment is currently using an iPhone.
Using these estimates, 20% × 280 million = 56 million iPhones are currently being used in China.
A follow-up question might be something along the lines of “Are there are more iPhones in operation in the U.S. or in China?” We’ll leave you to try and figure that one out yourself.
Another follow-up question might be how you would check the accuracy of your assumptions and response (in this case, perhaps by reviewing the Apple annual report or telecom industry reports).

Guesstimate Example #4

What is the revenue of Peugeots sold in France per year?

Start by clarifying the question, then identify the variables to apply to this problem.
Population of France: Approximately 60 million people.
Assume an average household is 3 people. This leads to 20 million households (60 million ÷ 3).
Assume 20% of households have no car, as they are in urban cities such as Paris or Lyon.
Of the remaining households, assume an average of 1.5 cars per household.
Therefore, there are approximately 80% × 1.5 × 20 million households = 24 million cars in France.
Assuming a replacement rate of every 6 years, there will be (24 ÷ 6) = 4 million cars replaced per year.
Of these 4 million, how many are Peugeot brand? You could suggest that the French are quite patriotic, so perhaps 20% of the 4 million cars purchased each year are Peugeot.
Therefore, you estimate that (20 × 4 million) = 800,000 Peugeot cars are purchased in France per year.
Of the 800,000, assume 70% are new cars and 30% are used cars.
Assume that the average price is $30,000 for new cars, and used is $10,000 for used cars (this is assuming similar pricing, currency-adjusted, to that of U.S. cars).
Using these assumptions, (560,000 × $30,000) + (240,000 × $10,000) = $16.8 Billion + $2.4 Billion = $19.2 Billion.
Therefore, total Revenue of Peugeot cars sold in France per year is approximately $20 Billion.

Guesstimate Case Example #5

What is the Revenue for the board game Monopoly® sold in India per year?

Start by clarifying the question. In this case, intelligent questions might be:

Is there an Indian version of Monopoly®? (Interviewer: “Yes.”)
If yes, how long has it been on the market in India? (Interviewer: “The Indian version has been on the market since 2006. Prior to that, the British version was sold in India.”)

From this, perhaps the interviewer might further explain that in India there is a children’s version and the regular version, and she would like you to estimate the revenue of the children version, which targets 8-15 year old children.

Once the question is clarified, identify the variables to apply to this problem.
Population of India: Approximately 1.2 billion people.
Percent of population that is aged 8-15 in India: Assume 15%.
Total population of 8-15 year old children: 180 million.
Assumed percent living in areas where the board game is available for sale: 50%
Assumed percent of such children playing board games: 20%
Assumed number of board games purchased per child in this age range per year: 2
A potential question at this point for the interviewer: How popular is Monopoly compared to other games? (Interviewer: It is gaining a lot of interest and is a popular board game with an estimated 10% share.)
Average price for Monopoly®: Assume Rs 600 (The interviewer will probably provide the number; otherwise, you might need to estimate based on U.S. prices and dollar-rupee exchange rates)
Using these estimates, the annual Monopoly® sales in India are as follows: 180 million × 50% × 20% × 2 × 10% × Rs 600 = Rs 2.16 billion annual Revenue.
Interviewer: If the exchange rate is U.S. $1 to 60 Rupees, how many dollars is that?
2.16 billion ÷ 60 = $36 million.
Total Revenue for the children’s version of Monopoly® in India per year appears to be approximately $36 million. Note that because of the large chain of assumptions made, this estimate could be off significantly; in particular, the estimate is highly sensitive to the percentage breakdown assumptions for the relevant demographic (percent of 8-15 year olds living where the game is available; percent of those individuals who play board games; number of board games purchased by those customers annually; etc.)

←Consulting Business Situation Cases Consulting Brainteasers→

STREETOFWALLS

SOW