How to Size a Market

Market sizing questions. They’re universally feared and despised by aspiring consultants, investment bankers, and venture capitalists (not to be dramatic or anything!). Today’s article is meant to give you the tools to approach any market sizing question that ends up being thrown your way.

What is a Market Sizing Question?

Market Sizing questions are used in interviews for certain types of professional services firms (consulting, investment banking, and venture capital in particular) to help test “how you think.”

The general idea is to figure out how large the market is (in terms of customers, units sold, or revenues) for a particular product, whether it be bike helmets or gas stations, without using any outside information or a calculator.

Market Sizing estimations are also used by entrepreneurs and venture capitalists to figure out if they want to develop / invest in a theoretical product as well.

How do I Approach Market Sizing Calculations?

We can break up the approach up into six steps:

1. Define the Problem.

2. Choose Whether to Size by Supply or Demand.

3. Choose to Proceed Top Down or Bottom Up.

4. Choose an Appropriate Proxy.

5. Segment Your Market.

6. Calculate.

And perform a sanity check (are my assumptions reasonable?) at each step.

Define The Problem

Clarify exactly what market you’re trying to size:

1. Is it by revenue, number of units sold, or customers?

2. Is it daily, weekly, monthly, or annually?

3. Is it restricted to a particular region, country, or are we looking for the worldwide market?

4. Are we looking at the total addressable market for this product, or just the portion of the market we can capture?

Make sure you understand exactly what you’re trying to size.

Supply or Demand?

Note* You can skip this step if you’re asked to calculate a total market size. This is only relevant if you’re asked to calculate the size of the market you think you can capture.

Let’s say you are asked to size the market and figure out how much you can capture. The first thing you need to do is figure out whether you’re limited by supply or demand. Why?

It doesn’t matter how many units customers would want (demand) if you have limited factory capacity (supply) and vice versa.

So what you do is ask your interviewer if you’re assuming you can produce enough of the product to meet the demand. Usually they’ll say yes, but if they don’t you can try to calculate the total possible annual production. Usually that will be something like:

Total Annual Production = # of Factories * Annual Production Per Factory

Annual Production Per Factory = Amount Produced Daily * Days / Year

Amount Produced Daily = Total # of Machines * Daily Capacity / Machine

or something like that… (the exact calculation depends on context).

Anyway, once you figure out how much you can supply, you can go ahead and do a market sizing by demand and figure out how much of the market you think you can take. Then you compare the two figures and see what’s bigger.

If the market size you think you can capture > amount you can produce, then the amount you can produce will be the maximum you can capture.

If the amount you can supply > amount you think you can capture, then go with the amount you think you can capture.

Top Down or Bottom Up?

The next thing to figure out is if you want to proceed top down or bottom up with your market sizing.

A top down market sizing would be figuring out, what is the total addressable market for my product and then making an assumption of what percentage I could capture of that total.

They usually look something like this (market sizing for a premium subscription based job board):

Top Down Market Sizing

(credit: The Business Research Blog)

A bottom up market sizing would be the opposite; instead of figuring out the total addressable market you’d figure out what channels you’re going to go through and figure out how much you could sell in each channel. Usually they look like this:

“Here’s how a bottom up analysis might work in real life. Let’s pretend you have developed a prototype for a display case for rare coins and you want to determine if there is a market for your case – a profitable market that will sustain a real business. Let’s walk through the steps:

1. Where are coin display cases typically sold? Most are sold, fairly obviously, in coin and antique shops, but many are sold on the Internet, directly to the end user. You decide to focus on coin shops for now, since your customers already go to coin shops.

2. How many coin shops are in the U.S.? (We’ll assume you don’t want to try to sell and ship overseas, at least at this point.) With some research you find there are approximately 6,000 coin shops. (Keep in mind I made up that number.)

3. How many of those shops would be willing to stock your display case? Here’s where it gets tricky. If your display case is innovative, you may be hesitant to show it to others – you may even hesitate to share photos or drawings. While the risk someone could “steal” your idea does exist, you will face a bigger risk if you spend money setting up a business and creating inventory without determining if there really is a market for your product. If you’re nervous, you could have people sign a confidentiality agreement before you share any details.

Now talk to at least twenty or thirty coin shops to see if they would be willing to carry your display cases. If six say they would, cut that number in half. Always be conservative. (While they may agree in principle today, they may not come through in fact. Plus you may be physically unable to get product into every willing coin shop’s hands.)

Then extrapolate. If six out of thirty say they would carry your product, and you cut that number in half, then 10% of coin shops may be willing to carry your product. 6,000 times 10% equals 600. That means your display cases could potentially be sold in 600 coin shops.

4. Historically, how many display cases has each shop sold over the course of a year? That’s a good question, but a better question is, “How many display cases like mine does a coin shop sell in a year?” Your case is a premium case, and that limits your market. A premium product carries a premium price, and some buyers won’t be willing to pay a premium. Always compare apples to apples.

Say the coin shops you talk to say they sell, on average, fifty premium display cases a year. That’s great – but how many can you sell to the store? The answer isn’t fifty. You’ll be competing with other products, so you will have to convince coin enthusiasts to buy your product instead of another product. You go conservative and assume you can sell ten cases a year to each shop.

The math is easy: 600 shops times ten cases per coin shop equals 6,000 cases a year. What’s hard is pulling together the data so you can do the math.”

(credit: Metcalf Bank)

So how do you decide whether to go top down or bottom up? It depends on what sort of background information you have on your product. If it’s a completely new product and you don’t even know how you’d sell it, go top down.

If its something that has been done before, you know where you’re going to sell it, and there’s a limited number of channels to go through, take a bottoms up approach.

Choosing a Proxy

In many cases you’re going to be asked to size the market for either a completely new kind of product or something that is out of your everyday experience.

To size the market you need to pick a proxy. You can think of a proxy as a product or service whose customer base your product would also be valuable.

For example, let’s use the job board example above. The best proxy to use would be job seekers.

What about cigarette lighters? Probably frequent smokers.

What about wireless heart monitors? Probably people who have a history of heart disease.

If you can’t think of a proxy, the most general proxy out there is the entire population or the number of households.

Segmenting Your Market

OK, let’s use that exclusive job board market as an example. First thing you need to do is calculate the working population, which is a subset of the total population.

1. You start with the total population, and then segment it into working and nonworking people. How do you figure out what percentage of the population works?

Well, you can either make some sort of estimate  based on your personal experience (half of everybody I know works, so 50% of the population works) or break the population down by age group (i.e. I’m going to assume life expectancy is 75, the number of people at each age is the same, nobody below 18 and above 68 works full time, so 50/75 (2/3) of people could work, and maybe 80% of those people actually could work once you factor in disability etc. -> ~50% of people work).

2. So now you have the total number of working people, but not all of those people will be looking for jobs. So you need to segment the population again into people who are likely to stay with their jobs and people who are likely to switch. How do you figure out the breakdown?

It’s hard to say. I think the best thing you can do is rely on personal experience (“sometimes I click on people’s LinkedIn’s randomly, and have noticed 1 out of every 5 or 10 people are very interested in new opportunities, so maybe its between 10-20%).

3. Now you have to segment again, because not all job seekers will use free online job boards for finding jobs. Again, I’d rely on personal experience.

(“When my friends and I were looking for jobs, I’d say only about half of us used job boards.”)

4. Now again you have to segment, because there’s a difference between using a free online job board and paying for a subscription. Again, hard to say unless you actually have done a lot of market research, but you could say something like:

(“I’d estimate 1 out of 5 people who used online job boards amongst my peer group really struggled, and I suspect those would be the people who would actually pay for a service.)

And that’s it. And you can do similar segmentations if you’re sizing with a bottoms up approach also.

By the way, are you wondering why I keep advising that you use your personal experience, even if you don’t really have any experience with whatever product you’re dealing with?

The reason is that some interviewers will challenge your assumptions, just to see how you react. They already know its impossible for you to know if your assumptions are reasonable or not. It’s just a test.

As a way to stonewall them you can use your own experiences, even if your experiences may not be representative of the whole population, and tell them you’re doing so. You can caveat it with the fact that your experiences may be different than the population as a whole, but its the best proxy you have.


The calculation part is pretty straightforward, you can see my post here about how to do calculations quickly. I would just recommend a few things:

1. Pick easy to calculate numbers, otherwise you may mess up the calculations.

2. Do a sanity check at each step of segmenting. For example, does it make sense that a completely new company can take 10% of the market? Probably not, maybe 1-2% is more reasonable.

3. Don’t segment too many times. The more complicated you make this, the longer it takes and the more likely you are to screw up.


Market sizing questions can be intimidating, but with practice you’ll be fine. Make sure to practice with other people and call each other out on your assumptions to help improve your process.

The standard way of approaching market sizing questions is to:

1. Define the Problem.

2. Choose Whether to Size by Supply or Demand.

3. Choose to Proceed Top Down or Bottom Up.

4. Choose an Appropriate Proxy.

5. Segment Your Market.

6. Calculate.

And make sure to perform a sanity check (are my assumptions reasonable?) at each step.

Good luck!


3 thoughts on “How to Size a Market

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s