# Best Practices for Market Sizing for Case Interviews: Part 3

So far in our series of Casing 101 blogs, we have reviewed structuring the problem via an initial framework and navigating case math. In this edition, we will cover another problem-solving exercise that is common in case interviews: market sizing. At first, market sizing may seem daunting – how can we possibly estimate the number of televisions sold annually in Boston, for example? But once you learn the technique and get comfortable making certain assumptions, you will find that it’s actually an area where well-prepared candidates can differentiate themselves.

# Best Practice #1: Start at the Top

Many market sizing exercises will ask you to find, for a given country, the number of units sold or the number of people participating in an activity. At first, it undoubtedly will feel daunting to come up with a reasonable estimate. But it is actually quite easy to do if you start with the population of the market and then make a series of defensible assumptions to drill down.

To help you start at the top, it is helpful to commit the following figures to memory. Note that some of these statistics are approximations, but ones that will make your math much easier.

• US Population: 300 million
• China Population: 1.4 billion
• India population: 1.3 billion
• Europe population: 750 million
• Average people per household: 2.5 or 3 people, depending on which makes the math easier

Note that the people per household figure is important, because many products for which you are sizing a market are bought at the household level, and not the individual level. For instance, if you were asked to find the number of microwaves sold annually in Europe, you would want to calculate the number of households in Europe, because it is very unlikely that most households would have more than one microwave.

Best Practice #2: Segment the Population

Frequently in market sizing exercises, you will be dealing with a product that is likely to appeal differently to various segments of the population. Accordingly, it is important to show your interviewer that you understand this dynamic, and make assumptions accordingly.

For instance, if you were asked to size the number of coffeemakers sold annually in the US, calculating the number of households in the US is not enough. You also need to show your interviewer that you understand that not every household likes to drink coffee. Accordingly, your calculations to get to the number of households that drink coffee might look like as follows:

In other cases, you may go as far as to segment the population into two separate groups for which you will size the market. For example, if you are asked how many video games are sold in the US, you might want to break the US population down into individuals above and below the age of 35, as the number of people playing video games in each segment likely differs. Once you break the population into these two segments, you need to make assumptions for different penetration rates in each demographic.

# Best Practice #3: Sizing the Population Isn’t Enough; Make Further Assumptions to Get to The True Addressable Market

A common mistake that candidates make is to size a broad sub-segment of the population and stop there. Often, there are additional assumptions that need to be made to calculate size of the annual market into which the client is selling. For example, in the calculations above, many candidates arrive at 90 million coffeemakers in the US, and determine that this figure represents the annual size of the market. However, is it reasonable to assume that all 90 million households with a coffeemaker replace their coffeemaker each year?

When you arrive at a number that you think represents the market size you are trying to calculate, take an extra moment to sanity check in your head whether the figure you’ve arrived at really makes sense. Do you really think it’s reasonable that 90 million coffeemakers are sold in the US every year? If the answer is no, there is probably an extra assumption or two that are necessary in order to further drill down to a reasonable answer. Accordingly, always be cognizant of what additional assumptions might be necessary. In this instance, we have 90 million coffeemakers across the US. How often are these replaced? Any assumption between 5 and 10 years seems reasonable and defensible, allowing you to arrive at 9-18 million coffeemakers sold annually. Your full set of calculations on the page should look as follows: