Sales Territory Data Aside—Total Addressable Market (TAM): How to Size Your B2B Market and Prioritize the Accounts Worth Chasing
By Rick Elmore ·
Last quarter I sat in on a pipeline review where the VP of Sales confidently claimed the company's market was "basically unlimited." Every software company says this. It's almost always wrong, and it's expensive to believe. When your market feels infinite, you spread reps across every industry, fund ABM campaigns that spray budget at accounts you'll never win, and cut territories based on headcount instead of where the actual money sits.
The fix isn't a bigger spreadsheet. It's doing the math on your total addressable market honestly, then using that number to decide where your team spends its hours. Done right, TAM sizing is less about the big top-line figure the board wants to hear and more about the ruthless prioritization it forces underneath.
- TAM, SAM, and SOM answer three different questions — total demand, the slice you can actually serve, and the slice you can realistically win this year.
- Bottom-up beats top-down. Starting from your real pricing and account counts produces a number you can defend and act on; starting from a Gartner headline number produces a fantasy.
- The segmentation is the deliverable, not the dollar figure. A TAM model that can't tell you which 400 accounts to chase next quarter didn't do its job.
- Feed it back into territories and ABM spend. Winnable account density, not geography, should drive how you carve up the team.
What is total addressable market, really?
Total addressable market is the full revenue opportunity if every company that could possibly buy your product did buy it, at your pricing. That's the honest definition, and the word doing the heavy lifting is could. Not "every business on earth." Companies that have the problem you solve, are the right size to need a solution like yours, and operate in markets you can legally and practically sell into.
Underneath TAM sit two tighter numbers. SAM — serviceable addressable market — is the portion you can actually reach given your product, languages, integrations, compliance posture, and go-to-market motion. If your product only integrates with Salesforce, companies running HubSpot aren't in your SAM no matter how much they'd theoretically benefit. SOM — serviceable obtainable market — is the slice you can realistically capture in a defined window given your competition, brand, and sales capacity. SOM is the one that should keep you up at night, because it's the only one tied to a forecast.
Most leaders obsess over TAM because it's the number investors react to. Operators should obsess over SOM, because it's the number your reps live in.
Top-down vs bottom-up: why the method decides the quality
There are two ways to size a market, and they produce wildly different outcomes in credibility.
Top-down starts with an analyst's estimate of a category's total spend and shaves it down with assumptions. "The global CRM market is $X billion, we think we can get 2% of it." I've never once seen this number survive contact with reality. It's a round figure built on someone else's definitions, and the 2% is pulled from thin air. Boards nod at it. Sales teams can't do anything with it.
Bottom-up builds from units you control. You count the actual companies that fit your profile, multiply by what they'd realistically pay you per year, and add it up. It takes more work, but the output is something you can segment, sort, and hand to a rep. When someone on the exec team challenges the number, you can walk them straight down to the account list.
Here's the bottom-up formula I use, stripped down:
TAM = (number of companies that fit your qualifying criteria) × (average annual contract value you'd charge them)
If you sell a product with tiered pricing, you run this per tier and sum it. A 50-person company and a 5,000-person company aren't paying you the same, so don't model them as if they are.
How to calculate TAM with real data instead of guesswork
Start with your qualifying criteria — not your full ICP yet, just the hard filters that define whether a company could buy. Industry, employee count band, region, and any non-negotiable technographic requirement (they run the stack you integrate with, they have a sales team of a certain size, whatever the gate is for your product).
Then pull the actual company counts. This is where teams get lazy and reach for a top-down shortcut. Don't. Use firmographic databases — the same data sources feeding your sales intelligence tools — to query how many companies match each filter. Most platforms will give you a count before you ever export a contact. Cross-check it against a second source if the number feels off; data providers disagree more than you'd expect, especially at the edges of employee bands.
Now attach pricing. Take your real average contract value per segment from closed deals, not your list price and not your aspirational enterprise tier. If you're early and don't have enough closed data, use your pricing model and discount it to what deals actually close at. If your public pricing and packages vary by company size, model each band separately — this is where a single blended ACV quietly distorts everything.
Multiply and sum. That's your TAM. Then start cutting toward SAM: remove the companies you can't actually serve. No integration, wrong geography for your support hours, regulatory barriers, segments where your product genuinely underperforms. Be honest here. Every account you leave in your SAM that you can't service is an account a rep will waste a week on.
Finally, pressure-test toward SOM. Look at your win rate by segment, your competitive density, and your sales capacity. If you have eight reps who can each run 40 active opportunities a year with a 25% close rate, your obtainable market this year is bounded by that math, not by how many companies exist. SOM is the collision between demand and your own capacity.
A simple worked example
Say you sell a sales automation platform priced for mid-market teams.
| Layer | Filter applied | Companies | Avg ACV | Value |
|---|---|---|---|---|
| TAM | B2B companies, 50–1,000 employees, has a sales team | 120,000 | $18,000 | $2.16B |
| SAM | Of those, run a CRM you integrate with, in your supported regions | 38,000 | $18,000 | $684M |
| SOM | Realistic reach + win rate given current capacity (next 12 months) | 900 opportunities | $18,000 | ~$16.2M pipeline |
The numbers are illustrative, but notice what happens. The TAM is a headline. The SAM cuts it by two-thirds the moment you apply honest serviceability filters. And the SOM is a different kind of number entirely — it's bounded by your team, not the market. That SOM figure is the one you build a plan around. The $2.16B is the one you put on slide three for the board.
Turning the segmentation into territories and ABM spend
This is where TAM work actually pays off, and where most teams stop too early. Once you've built the account list underneath your SAM, you have a scored, segmentable universe. Don't flatten it back into a single number and move on.
Sort your SAM accounts by fit and winnability — how closely they match your best customers, whether there's a competitor entrenched, whether you have a warm path in. Now cluster the high-fit, winnable accounts and look at where they concentrate. Sometimes it's an industry vertical. Sometimes it's a company-size band. Rarely is it a clean geographic region, which is exactly why geography-based territory carving leaves money on the table.
Build territories around account density, not map lines. A rep whose patch contains 300 high-fit accounts and a rep whose patch contains 60 are not running the same job, and paying them on the same quota is how you lose the good one. This connects directly to ICP and territory planning — your ICP tells you what a great account looks like, your TAM tells you how many of them exist and where they cluster, and your territory design puts the right rep in front of the right concentration.
For ABM, the logic is sharper still. ABM budget is finite and expensive per account. Spending it on your full SAM is lighting money on fire. Spend it on the top slice of your SOM — the winnable, high-ACV, high-fit accounts — and let lower-fit segments run on lighter-touch automation. The TAM model gives you the defensible line between "worth a custom ABM play" and "worth a sequence." Without it, that line gets drawn by whoever shouts loudest in the pipeline meeting.
One operator warning: a TAM model is not a one-time document. The moment you ship a new integration, enter a region, or launch a tier, your SAM shifts. Rebuild it at least annually, and re-run it before any major territory or ABM budget decision. A stale TAM is worse than none, because people trust it.
Frequently asked questions
How is TAM different from my ICP?
Your ICP describes the characteristics of a great-fit customer — the qualities that predict a fast close and long retention. TAM counts how many companies matching those characteristics actually exist and what they're worth in revenue. ICP is the definition; TAM is the sizing and the dollar value built on top of it. You need the ICP first, then you size against it.
What data do I need to calculate TAM accurately?
Three things: clear qualifying filters (industry, size, region, required technographics), firmographic company counts from a reliable data provider, and your real average contract value by segment from closed deals. If you lean on a top-down analyst figure instead of actual account counts, you'll get a number that sounds impressive and tells you nothing about which accounts to chase.
How often should I rebuild my TAM model?
At minimum once a year, and always before you redesign territories, set annual quotas, or allocate ABM budget. Any product change that expands who you can serve — a new integration, a new region, a new pricing tier — changes your SAM and should trigger a refresh. Treat it as a living input to planning, not a slide you built once for a fundraise.
If your market feels "unlimited" and your territories are carved by zip code, you're almost certainly spreading your team across accounts you'll never win. We'll build you a defensible TAM, SAM, and SOM model, segment it into a prioritized account list, and wire it straight into territory and ABM decisions. Book a Revenue Systems Audit.