# Guesstimation

The guesstimation questions and answers in this lesson will help you understand the types of guesstimation questions you can expect in data science interviews.

## We'll cover the following

## How to answer guesstimation questions

A lot of people ask how to answer guesstimation questions. At the time of the interview, you will not get much time to answer guesstimates. So, the interviewer expects that you are good with numbers. The most important thing you must understand is that the interviewer is not looking for a correct answer. They are looking for the closest value to the actual estimate, and they might ask you how you found that value.

**The best way to prepare for guesstimation questions is as follows:**

Study the company business where you're going to interview. For example, suppose they are a taxi service provider like Uber. In that case, they are more likely to ask you guesstimates related to taxi services, such as the number of cars on the road, number of taxi passengers in a day, number of car accidents, and so on. Once you clearly understand the business, move on to the second step.

Search on Google for an approximate value of the population the company targets and round off that value. For example, after searching on Google, you get an estimated 1.32 million, then round it to 1 million.

After this step, start making segments. The number of segments should not be very large because this will take much of your time, and it would be difficult to explain how you found the estimate. You should find the simplest way to find the estimated value. For example, to find the number of taxi passengers, you can make segments by using age groups and give a percentage to each group, then use the whole population and calculate the final estimate.

Note:For any calculations to get your estimate, ensure that you round off the values because if you don't round off, you will run out of time.

## Guesstimation questions and answers

**How many cars are there on New York City roads?**

Consider that the population of New York city is 10 million. 50% of people use cars, and the rest use buses, the metro, and so on. Therefore, we are left with 5 million people. Suppose each family in the city has an average of 5 people. Therefore, there are 1 million families in the city where a car is used, whether they're using a car, a yellow taxi, an Uber, or any other taxi service.

Now suppose that among these 1 million families, 60% own a car, and say 10% of the families have at least two cars. Therefore the total number of cars is 600,000 + 60,000 = 660,000. The rest of the families use taxi services; therefore, we have 400,000 more cars in the city. Hence, the total number of cars is 1.06 million.

Now assume that 50% of these people do not travel by car daily. Therefore, we are left with 530,000. That means the city uses 530,000 cars daily. Now, if it is the opening time or closing time for most offices and colleges, we can say that there are 500,000 cars on the city's roads. If it is midday, we can reduce it to 50% and say that there are 250,000 cars on the roads of New York City.

**Can you estimate the total number of English-speaking learners in India?**

The first value we can consider here is India's population, which is 1.3 billion. Round it to 1 to make further calculations easier. Now we need to make segments based on education, so the next thing is the literacy rate, which is around 70%. Hence, the population decreased to 70% of 1 billion, which is 0.7 billion.

There are many English-speaking people in India because almost 50% of Indian schools use English for teaching, and students from those schools generally don't take speaking classes. Hence, we can reduce the population to half, and the resulting value is 0.35 billion. Round it to 0.4.

We estimated the number of English-speaking people in India, but we want to know English-speaking *learners*, so we are left with 0.4. Among 0.4, consider that 50% are babies, old people, and working people, so we are left with 0.2. Again, among these 0.2, only 20% are English-speaking learners. Therefore, the target population is 0.04 billion or 40 million. It is a reasonable estimate because India's official languages are English and Hindi.

**Can you estimate the total number of IELTS applications from Germany annually?**

Germany is a small country. Assume that the population of Germany is 100 million and literacy in Germany is very high, so we can expect that almost everyone is literate. Since most people in Germany are patriotic, let's assume that only 1% of people want to leave their country. Therefore, we are left with 1 million people who want to leave Germany. Again, among these 1 million people, only 30% are eligible for IELTS. So, we are left with 300,000 people.

We know that IELTS is not easy, and many people fear it. We can remove 50% of people due to the difficulty of the IELTS exam. Now we have 150,000 people. Again, consider that most people think about giving the IELTS but do not appear because they are not prepared or they are not interested in giving IELTS in a particular year. Now let's say that 100,000 people are not prepared or interested; hence, we are left with 50,000 people in Germany who appear in IELTS annually.

**How many lipsticks are sold on a weekend in Dubai? **

This can be solved by using the below formula:

Level up your interview prep. Join Educative to access 70+ hands-on prep courses.