TAM, SAM and SOM are three nested numbers: every company that could buy your product, the part you can serve today, and the part your team can realistically win in the next 12 months. In B2B, get them by counting real companies and real buyers, then multiplying by what one customer pays you a year.
You will get the formulas, a worked example built on US Census and live people counts, and a worksheet you can fill in this afternoon.
What are TAM, SAM and SOM?
TAM (total addressable market) is the yearly revenue you would earn if every company with the problem bought from you. SAM (serviceable addressable market) is the slice your product, price and sales channel can serve today. SOM (serviceable obtainable market) is the slice of SAM you can realistically win in the next 12 months.
Each layer answers a different question. TAM tells you if the market is worth building for, SAM tells you where to aim the product and the team, and SOM tells you what goes in next year’s plan. In B2B, write each one twice: as a count of accounts and buyers, and in dollars.

TAM, SAM and SOM formulas
Every TAM SAM SOM formula is the same multiplication: a count of accounts times what one account pays you per year. The work is in counting the accounts honestly.
| Layer | Question it answers | Bottom-up formula |
|---|---|---|
| TAM | Is this market worth it? | Accounts with the problem × ACV |
| SAM | Where do we focus now? | Accounts you can serve today × ACV |
| SOM | What will we win next year? | Accounts worked in 12 months × win rate × ACV |
ACV is annual contract value: what one customer pays you per year. If your price rises with company size, use one ACV per size band and add the bands, as the example below does. You will also see SAM called the serviceable available market; it means the same thing.
Then add the column the formulas do not show: the number of buyers inside each layer. That turns a slide into a list. For the TAM layer alone, see our bottom-up total addressable market guide.
Top-down vs bottom-up: which method to use
Use bottom-up. Top-down starts from an analyst’s total (“the manufacturing software market is worth X billion”) and takes a percentage of it. Bottom-up starts from a count of companies that fit your customer profile and multiplies by your own price.

Top-down is fine as a check on the order of magnitude. It hides three problems: the report’s category rarely matches your buyer, the percentage you pick is a guess, and at the end you still do not know which companies to call.
A third approach, value theory, prices each customer on the value you create for them. It helps you set ACV, but it still needs an account count underneath.
How to calculate TAM, SAM and SOM bottom-up
Calculating TAM, SAM and SOM bottom-up takes about an hour in five steps: define the account, count accounts, count buyers, price the layers, then cap SOM by capacity.

1. Define the account in one sentence
Write who buys: industry, company size band, country and the job title that signs. “Operations leaders at US manufacturers with 50 to 499 employees” is usable. “Mid-market industrial companies” is not. If you do not have this yet, start from an ideal customer profile template.
2. Count the accounts from a primary source
For US companies, use the Census Bureau’s Statistics of U.S. Businesses (SUSB). It counts employer firms by industry (NAICS code), state and company size. The 2022 tables, released in April 2025, are the latest.
SUSB sizes each firm by its total employees across all locations and leaves out businesses with no employees. TAM and SAM are usually two size bands of the same industry count, so this one file gives you both.
3. Count the buyers inside those accounts
Accounts do not answer the phone; people do. Count the people who hold the buying title at companies in your SAM, then divide by the number of accounts. The ratio tells you how many contacts each company gives you.
In Scalelist, you describe your ICP in one sentence and see how many people match before you spend a credit.
4. Price each layer with ACV
Multiply accounts by ACV for TAM and SAM. Use your real price list, not the price you hope to charge. If you have no customers yet, use what your first pilots agreed to pay.
5. Cap SOM by capacity and win rate
SOM = accounts your team can work in 12 months × the share of worked accounts that become customers × ACV. Capacity is reps × new accounts per rep per month × 12. In founder-led sales, capacity is your own calendar.
Take the win rate from your CRM. With no history, run a 3-month pilot on a few hundred accounts before you commit to a number.
TAM SAM SOM example: a scheduling tool for US manufacturers
Here is the full method on one real market. The company is fictional: Linefold, a Columbus, Ohio startup selling production-scheduling software to manufacturers. Prices, team size and win rate are example inputs. The firm and people counts are real.
TAM: every US manufacturer with 20 or more employees
Linefold’s founder, Priya Raman, decides that shops under 20 people schedule on a whiteboard and will not pay for software. That removes 177,779 of the 239,265 US manufacturing firms, about three in four, and leaves 61,486. Her price rises with company size, so she uses three ACV bands.
| Company size | US manufacturing firms (2022) | Example ACV | Market per year |
|---|---|---|---|
| 20 to 49 employees | 31,593 | 4,800 USD | 151.6 million USD |
| 50 to 499 employees | 25,716 | 9,600 USD | 246.9 million USD |
| 500+ employees | 4,177 | 24,000 USD | 100.2 million USD |
| TAM | 61,486 | 498.8 million USD |
Firm counts: US Census Bureau, SUSB 2022, NAICS 31-33 (manufacturing).
SAM: the size band Linefold can serve today
Linefold handles one to three plants and does not integrate with enterprise ERPs yet. Its serviceable addressable market is the 50 to 499 band: 25,716 firms × 9,600 USD, about 246.9 million USD a year.
Now count the buyers. On 27 September 2026, Scalelist’s people count for US manufacturing companies with 51 to 500 employees returned these numbers (a 1-employee difference from the Census band):
- Over 1.9 million people work at those companies.
- Over 196,000 are decision makers (director level or above, any function): about 7 to 8 per company.
- Over 26,000 are operations leaders (director, head, VP or C-level in operations): about one per company, or 1 in 75 of all the profiles there.
- A title list (COO, VP of Operations, Director of Operations, Head of Operations, Plant Manager) returns over 23,000, which confirms the order of magnitude.

That last ratio matters. One operations leader per account means Linefold’s SAM is a list of up to about 26,000 named people, not an abstract market. If manufacturing is your market too, the manufacturing lead generation playbook covers how to reach them.
SOM: what two reps can win in 12 months
Linefold has two sales reps. Each can research, email and call about 150 new accounts a month (example input). Its 3-month pilot turned 2% of worked accounts into customers.
- Reach: 2 reps × 150 accounts × 12 months = 3,600 accounts
- Customers: 3,600 × 2% = 72
- SOM: 72 × 9,600 USD = 691,200 USD in year one
That is 0.28% of the SAM. The usual slide claims “1% of the market”: 257 customers. At a 2% win rate that needs about 12,850 worked accounts a year, which is about 7 reps, not 2.

Fill in your own numbers
Copy this worksheet into a doc or a spreadsheet. Every line has one source, and lines 5, 6, 8, 11, 12 and 13 are simple formulas.
2. TAM accounts (all companies with the problem): ____ Source and year: ____
3. SAM accounts (the ones you can serve today): ____ Source and year: ____
4. ACV per account (one per size band if prices differ): ____ USD
5. TAM = line 2 x line 4 = ____ USD
6. SAM = line 3 x line 4 = ____ USD
7. Buyers in SAM (people count for your buyer title, with date): ____
8. Buyers per account = line 7 / line 3 = ____
9. Accounts your team can work in 12 months (reps x accounts per rep per month x 12): ____
10. Win rate on worked accounts (from your CRM or a pilot): ____ %
11. SOM customers = line 9 x line 10 = ____
12. SOM = line 11 x line 4 = ____ USD
13. SOM share of SAM = line 11 / line 3 = ____ %
Count the buyers in your SAM: describe them in one sentence and see how many match before you spend a credit.
Describe your buyer in one sentenceCommon TAM, SAM and SOM mistakes (and the fix)
Most market sizing goes wrong in the same four places. Each has a simple fix.
Counting companies, not buyers
A 25,716-account SAM means little if only a few hundred have a reachable buyer. Fix: add the people count for your buyer title to every layer.
Picking SOM as a percentage
“We will take 1%” is a guess written as math, and it hides a hiring plan. Fix: derive SOM from reps, accounts per rep and win rate.
Mixing size bands
Census bands (50 to 499) and data-tool bands (51 to 500) rarely line up exactly. Fix: note both, and never add numbers from different bands.
Treating a people count as a contact list
A people count is a count of matching profiles, not of verified emails or mobiles, and it can include people matched on a past role. Fix: treat it as an upper bound, and check the real hit rate on a 100-person sample before you plan capacity.
Best practices for a market size you can defend
A defensible market size shows accounts, buyers and dollars for each layer, with a source and a date next to every count.

- Label every assumption as one: your ACV, rep capacity and win rate are inputs, not facts.
- Recount when your ICP, price or team size changes, and at least once a year.
- Make the SOM real: build a target account list of the first 300 accounts and their buyers, then segment the market with firmographics to pick where to start.
Where to get each number (and what to look for in a tool)
A US bottom-up count needs four inputs: accounts from Census SUSB, buyers from a people search, ACV from your price list, and capacity and win rate from your CRM. Company size and industry are firmographics, so any segment you can describe that way can be counted. Track the outbound sales metrics that feed your win rate.
For the buyer count, pick a tool that:
- Shows the count at no cost, before you spend credits.
- Understands title, seniority, industry, company size and location from plain English, and shows what it understood.
- Returns work emails and mobile numbers for the same people when you are ready to reach them.
- Charges only for contacts it finds.
How Scalelist helps you turn SOM into a list
Scalelist, an AI Lead Finder, lets you size each layer at no cost, then pull the SOM as named people with verified work emails and mobile numbers.
You describe your buyer in one sentence. Scalelist’s Lead Finder shows what it understood and how many people match before you spend a credit. For Linefold’s buyer, that count was 26,308 on 27 September 2026.

For Linefold’s plan, 300 new accounts a month with one operations leader each costs at most 6,600 credits a month: 1 per lead, 1 per email, 20 per mobile, 0 for a miss. That fits the Scaler 10K plan at 349 USD a month (see Scalelist pricing).
Scalelist finds and verifies contact details on demand rather than serving a pre-compiled list.
Size your SAM for free, then reveal verified emails and mobiles only for the buyers you keep.
Get 20 credits to test itFAQ
What is the difference between TAM, SAM and SOM?
TAM is every account that has the problem your product solves. SAM is the part your product, price and sales channel can serve today. SOM is the part your team can realistically win in the next 12 months. Each is a subset of the one before, and in B2B each should be written as accounts, buyers and dollars.
How do you calculate TAM, SAM and SOM?
Multiply accounts by annual contract value (ACV). TAM = accounts with the problem × ACV. SAM = accounts you can serve today × ACV. SOM = accounts your team can work in 12 months × win rate × ACV. Count accounts from a primary source such as the US Census, and use your own price and win rate.
What is a good SOM as a percentage of SAM?
There is no standard percentage, and picking one is the most common mistake. Derive SOM from capacity instead: reps × accounts each rep can work × win rate. In our manufacturing example, two reps produce a SOM of 72 customers, 0.28% of the SAM. Claiming 1% would have needed about seven reps.
Should I use top-down or bottom-up market sizing?
Use bottom-up as your answer and top-down as a sanity check. Bottom-up counts real companies and buyers and multiplies by your own price, so every input can be checked and the result doubles as a target list. Top-down starts from an analyst’s category total, which rarely matches your exact buyer.
Where can I find how many companies are in my market?
For US companies, start with the Census Bureau’s Statistics of U.S. Businesses (SUSB), which counts employer firms by industry, state and employee size; the 2022 tables are the latest. For the people inside those companies, use a people search that shows a count before you pay, such as Scalelist’s Lead Finder, and note the date.
Is TAM, SAM, SOM only for investor pitch decks?
No. The same numbers tell you which segment to sell to first, how many reps you need and how long your list will last. In the example, 25,716 accounts worked at 3,600 a year is about seven years of new accounts for two reps. That is a hiring decision, not a slide.
How often should I update my TAM, SAM and SOM?
At least once a year, and whenever your ideal customer profile, price or team size changes. Census business counts come out yearly with a lag of about two to three years, while people counts move month to month. Date every number you show so readers know how fresh it is.
Does Scalelist charge to count a market?
No. Sizing a search in Scalelist is free: you see how many people match before you spend a credit. Credits are charged only when something is found: 1 per lead, 1 per email and 20 per mobile number. A miss costs 0, and the free trial includes 20 credits.
Sources
- US Census Bureau: 2022 SUSB annual data tables by establishment industry
- US Census Bureau: SUSB 2022, US and states by detailed employment size (xlsx)
- US Census Bureau: SUSB glossary
- US Census Bureau: SUSB methodology
- Scalelist people count (POST /api/v1/people/count), queried 27 September 2026, 0 credits.