Market size research: a bottom-up TAM, SAM and SOM from public data

Market size research estimates how much revenue a product could earn in a market. The most useful version is bottom-up: count the potential buyers from public data, multiply by what each would pay a year, then narrow to the buyers you can actually serve and the share you can realistically win. Those three steps give the three numbers people ask for:

• TAM (total addressable market): revenue if every potential buyer paid you.
• SAM (serviceable addressable market): the part of TAM your product, channels and geography can serve.
• SOM (serviceable obtainable market): the part of SAM you can realistically win in the next few years.

A top-down figure, such as a report's estimate of a whole industry, is useful as a rough check. But "1% of a huge market" hides every assumption, and a bottom-up number shows them.

A worked example: an app for HVAC technicians

Suppose you are building a mobile app that heating and air-conditioning technicians use on service calls to look up equipment history and write up jobs. You plan to charge contractors per technician per month. The price, sales capacity and results below are hypothetical. The employment figures come from the U.S. Bureau of Labor Statistics and the arithmetic is shown so you can repeat it.

Decide what unit you are counting

Count the unit you charge for. That choice decides which data source to use:

County Business Patterns covers establishments with paid employees, with counts by industry, employment size and county. Statistics of U.S. Businesses reports firms, establishments, employment, payroll and receipts, broken down by enterprise size. Receipts are included only for years ending in 2 and 7, and each year's data are released about two to two and a half years later.

Our app is priced per technician, so we count technicians.

Count the buyers and build TAM, SAM and SOM

The BLS profile of heating, air conditioning and refrigeration mechanics and installers says the occupation held about 440,900 jobs in 2025. Of those, 68% worked for plumbing, heating and air-conditioning contractors and 6% were self-employed; the rest were spread across other employers such as wholesale trade, education and retail.

Assume a price of $40 per technician per month, or $480 a year.

Three notes on the steps:

• SAM should reflect real limits. Self-employed technicians and technicians at schools or retailers are left out here because a contractor-focused product and sales process would not reach them. If you can only sell in certain states, or only to contractors above a certain size, narrow further using County Business Patterns by county or Statistics of U.S. Businesses by firm size.
• SOM comes from your capacity, not a percentage you picked. The SOM above is under 1% of SAM because it is based on how many deals two people can close, which is usually the real constraint early on. Change the inputs and the result changes in a way you can explain.
• Churn lowers it. The example assumes no customers leave. Real SOM will be lower.

Test the assumptions that move the answer

The count came from public data. The inputs that change the result most are the ones you guessed: price, how contractors buy, and how fast you can sell. Check them with people who run the work.

• Price and pricing unit. Do contractors pay for software per technician, per office or as a flat fee? A flat fee would change the whole calculation.
• Share that would use it. Some technicians may not carry a phone for work or may share one device. Ask contractors how their crews work today.
• Sales cycle. How long does a contractor take to decide, and who signs off?

Five or six conversations with contractor owners and service managers will tell you more about these inputs than another report. How to research willingness to pay covers price questions, and market research interviews with customers, suppliers and operators covers whom to ask.

Check the result against other numbers

Once you have a bottom-up estimate, compare it with any figure you can find from the other direction:

• Public competitors' annual reports. Investor.gov's guide to reading a 10-K explains that Item 1 describes a company's products and markets, and the financial statements show its revenue. Filings are public on the SEC's EDGAR website. Revenue often covers broader segments than yours, so treat it as a rough ceiling or floor, not a match.
• Industry reports. Use them to see whether your number is in a reasonable range. If a report's figure is far from yours, find out what it counts before trusting either.

If your bottom-up SAM is larger than the entire revenue of every company selling to that market, check your price or count.

Common mistakes

• Counting people who are not buyers. Every technician is not a buyer. The contractor is.
• Mixing units. Multiplying a count of workers by a per-company price inflates the result many times over; a count of companies times a per-seat price shrinks it just as badly.
• Using a big headline number. A TAM in the billions says nothing about whether you can reach the first hundred customers.
• Mixing years. Note the reference year of every input, since public data sets are updated on different schedules.

Secondary market research lists more free data sources, and market research for a new product explains why market size should come after checking that the problem exists.

Your next step

Write your pricing unit and price on one line. Find the public count for that unit, multiply, then narrow to the buyers you can serve and the number your team can close in a year. Circle the two inputs you guessed, and test them with five conversations.

If you do not know anyone who buys in the market, Instant Expert can find people who match a description you write, such as "service managers at residential HVAC contractors." You review who it finds, it sends your invitations, and you pay only for calls that get booked. The directory pages for sales professionals in construction and operations professionals in facilities management are one place to start.