Serviceable Addressable Market (SAM): Formula and Examples

Once you have the number, the TAM calculator is where it becomes a list: Clu, Cleanlist AI's GTM agent, builds it in People Search, where searching costs 0 credits, and verifies each work email at 1 credit.

What is a serviceable addressable market?

A serviceable addressable market is the slice of a market you could sell to now, with the product you have, in the places you sell, at the price you charge. TAM asks how big the market could ever get. SAM asks how much of it is open to you this year.

The name comes in several versions. Serviceable available market, served available market and served addressable market all mean the same thing, and all shorten to SAM (Wikipedia, read November 4, 2026). Pick one name and use it in every plan.

SAM has two units, and RevOps needs both:

• Accounts: the number of companies you can serve. Territories, rep capacity and list building run on this.
• Revenue: those accounts times what each would pay a year. Board decks and fundraising run on this.

A SAM is defined by its filters, and each filter should name its reason, such as "English-only product" or "no single sign-on yet". When the reason goes away, the filter goes too. Open a region or ship an integration, and the SAM grows on the day it ships.

The SAM formula

SAM = serviceable accounts × average annual contract value (ACV)

Contract values usually differ by segment, so add the segments up:

SAM = (accounts in segment 1 × ACV in segment 1) + (accounts in segment 2 × ACV in segment 2) + ...

For a product priced per seat, ACV is the expected seats × the price per seat × 12.

The last row is the one most guides skip. Some filters have no search field in any tool, such as the CRM a company runs. Check a random sample of 100 accounts by hand, count how many pass, and apply that share to the total.

A worked B2B example

The company below is made up, and so is every number. It is here to show the arithmetic at each step.

Say you sell coaching software for SDR teams, priced per seat. It integrates with HubSpot and Salesforce only, the product is in English, support runs on North American hours, and it has no single sign-on yet.

Step 1: Start from TAM. There are 100,000 companies worldwide with three or more SDRs, and the average customer would pay $15,000 a year. TAM = 100,000 × $15,000 = $1.5 billion.

Step 2: Apply the filters, one at a time.

Step 3: Price each segment. The closed-won deals show that smaller customers buy fewer seats.

The SAM is $152.2 million, held by 11,340 accounts. That is about 10% of the $1.5 billion TAM.

Step 4: Check it against capacity, which gives the SOM. The team has 8 account executives. Once ramped, each closes about 2 new customers a month, so the team adds about 190 customers a year. Allow for ramp time and churn, and the team ends year three with about 500 customers: 500 × $13,418 is about $6.7 million in annual recurring revenue, from 4.4% of the SAM's accounts. If the plan says $20 million in three years, it needs about 1,490 customers, 13% of the SAM, from the same 8 reps. Either the team grows or the plan changes.

What the example shows:

• Each filter is a reason. When the product adds single sign-on, the size filter can widen to 5,000 employees and the SAM grows that day.
• Contract value by segment matters. One blended $15,000 ACV would have put the SAM at $170.1 million, 12% too high.
• The SAM is an account count first. 11,340 named companies is something a RevOps team can build, assign and track.

Three more SAM examples

Each company is made up. The point is which filter does the most work.

TAM vs SAM vs SOM

SAM does the most work of the three. TAM is too broad to build a list from, and SOM is a forecast. SAM is the market you can name account by account. For how an ICP fits next to these numbers, see what ICP means and how it differs from TAM.

Top-down vs bottom-up SAM

Run both and compare them. Say a report puts the category at $2 billion and your filters keep 45% for region and 40% for size: top-down gives $360 million, against $152.2 million bottom-up. A gap that wide usually means the category figure includes products you do not sell, such as full conversation intelligence suites. Trust the bottom-up count, and write down why the two differ.

Seven SAM mistakes to avoid

1. Calling TAM a SAM. If no filter was applied, the number is a TAM. Every SAM should list its filters.
2. Counting markets you cannot reach this year. No rep in the region, no translation, no integration: leave the segment out, and add it back when that changes.
3. One ACV for every segment. In the worked example, a single average overstated the SAM by 12%.
4. Leaving in your customers and open deals. For a new-customer SAM, take out current customers and accounts with an open opportunity. Report the total and the remainder side by side.
5. Mixing company location with personal location. Count accounts by where the company is headquartered. A buyer who lives in Toronto may work for a company based in London, and the two give different counts.
6. Counting subsidiaries twice. Roll subsidiaries up to the parent when the parent buys for all of them, or one deal shows up as three accounts.
7. Never updating it. Re-run the SAM every quarter, and after any launch, new region or price change.

How to turn a SAM into a list of real accounts and contacts

A SAM in a spreadsheet changes nothing. The work starts when each filter becomes a search.

1. Write each filter as a search field. Region becomes company headquarters country, size becomes an employee band, and industry becomes an include and exclude list.
2. Count the accounts. Run the search and compare the count with your estimate. If they differ by more than 20%, find the filter that is off before you go further.
3. Remove what you already own. Take out customers, open opportunities, partners and do-not-contact accounts by matching company domains against your CRM.
4. Pick this quarter's slice. Tier the accounts by fit, using a model like the one in our account scoring guide, and give each rep what they can work. If 8 reps each work 150 accounts a quarter, the first slice is 1,200 accounts, about a tenth of the SAM.
5. Find 2 or 3 buyers per account. The person who signs and the person who uses the product, by job title and seniority.
6. Add verified contact data. A verified work email for every buyer, and a phone number for the top tier.
7. Mark the SAM in your CRM. A "SAM segment" property on each account lets you report coverage: the share of SAM accounts worked, with a verified contact, and in pipeline.

How to build your SAM account list in Cleanlist AI, step by step

1. Open the TAM calculator. It lays out TAM, SAM and SOM, and its button opens a Cleanlist AI workspace. Every workspace starts with 14 days of Pro: up to 3 seats and 250 credits, with a card required.
2. Describe the SAM to Clu in one sentence. For the worked example: "VPs of Sales and SDR managers at software companies headquartered in the US or Canada with 51 to 1,000 employees." Clu turns the sentence into People Search filters: company headquarters country, employee band, industry, job title and seniority.
3. Read the counts before you spend anything. Searching costs 0 credits, so change a filter and run it again until the account count matches your estimate.
4. Handle the filters no index has. "Runs HubSpot or Salesforce" is not a search field, and neither are revenue and funding (which ICP traits no tool can filter). Add an AI research column that checks each company's website and careers page, at 3 credits a lead, or check a sample by hand and apply the share.
5. Enrich only this quarter's slice. Cleanlist AI looks up verified work emails and phone numbers across 25+ data providers, at 1 credit per verified email and 10 per phone number, and a miss costs nothing. On our own 500-lead benchmark, that returned 98% verified work emails and 85% direct dials.
6. Score and load the list. ICP scoring gives each lead a fit score from 0 to 100 at 5 credits a lead, so reps start with the best accounts. On Pro, the HubSpot and Salesforce sync writes the list into your CRM. On any plan you can export a CSV.

What you get is a named list of the accounts in your SAM, with 2 or 3 buyers per account for the slice you will work. In the worked example, counting all 11,340 accounts and their buyers costs nothing, though Clu uses credits based on the work it does. Enriching the first 1,200 accounts with 2 buyers each is up to 2,400 credits for verified emails, and phone numbers for the top 300 buyers add up to 3,000 more. Starter is $49 and Pro $89 a seat a month, with 750 and 1,500 credits per seat.

Frequently asked questions

Is serviceable available market the same as serviceable addressable market?

Yes. Serviceable available market, serviceable addressable market, served available market and served addressable market are four names for one number, all shortened to SAM. Use one of them in every plan, because teams that write both often end up with two different figures.

Should SAM be measured in revenue or in accounts?

Both. Revenue, accounts times ACV, is the number for a board or an investor. The account count is the number RevOps uses to set territories, check rep capacity and build lists. Report them together, with the filters underneath.

What is a good SAM size?

There is no universal benchmark, because it depends on your price and your plan. The useful test is the share of SAM accounts your plan needs to win. In the worked example, $6.7 million in three years takes 4.4% of the SAM's accounts and $20 million takes 13%, and the higher that share, the more proof the plan needs in win rate, sales cycle and rep count.

What is the difference between SAM and an ICP?

An ICP describes your best-fit accounts, with traits like industry, size, a buying trigger and the disqualifiers. A SAM counts every account you can serve, including weaker fits. The ICP usually sits inside the SAM and decides which SAM accounts get worked first. For building one, see ICP in sales.

How often should you update your SAM?

Every quarter, and whenever a filter changes: a new integration, a new region, a translated product, a new price or a new compliance certification. Keep a dated change log next to the filters so you can explain why the number moved.

Can SAM be larger than TAM?

No. SAM is a subset of TAM, so a larger SAM means the two were built on different definitions, such as a TAM counted in companies and a SAM counted in seats. Rebuild both in the same unit for the same market.

Sources

• Wikipedia, Serviceable available market, read November 4, 2026, for the other names of SAM
• Cleanlist AI credit costs and plans are our own published prices, read from our pricing on November 4, 2026. The worked example and the three short examples are made up and are not customer data.