How to Calculate TAM, SAM, and SOM Without a Consulting Firm
TAM, SAM, and SOM are three shrinking circles of opportunity: total addressable market is every dollar spent on your category by anyone, anywhere; serviceable addressable market is the slice you could actually reach given your business model and geography; serviceable obtainable market is what you can realistically win in the next few years. You size all three by working top-down from industry data, bottom-up from your own unit economics, and then checking that the two answers agree.
Quick Answer: Calculate TAM top-down by narrowing an industry report's total spend down through your segment and geography, and calculate it bottom-up by multiplying a realistic count of buyers by what you'll actually charge them. SAM is TAM filtered to the customers your current product and go-to-market can serve; SOM is the fraction of SAM you can capture given your resources, price, and competition. Run both methods independently, then check the two numbers land in the same neighborhood — a wide gap means one of your assumptions is wrong, not that you get to pick whichever number is bigger.
Prerequisites: A Defined Customer Segment and Price Point
Neither method works until you can describe your buyer in one sentence and name a price. "Small businesses" or "everyone who manages a team" isn't a segment — it's a category so broad that any market-size number you attach to it will be too large to mean anything. A usable segment names an industry, a company size band, a geography, and often a specific role: "marketing agencies with 10-50 employees in the US that currently manage client invoicing by hand" is narrow enough that you could plausibly go find and count those companies.
The price point matters just as much and gets skipped more often. TAM is a revenue figure, not a headcount, so you need a number to multiply units by — even a rough one. If you haven't settled on pricing yet, use the price of the closest competing product or the number from your own pricing experiments, and say explicitly that it's provisional. A market-size calculation built on a provisional price is still useful as long as you label it that way and revisit it once real pricing data exists.
This is the same discipline the broader market research process for technical founders starts from: you can't size a market you haven't defined, and defining it is a research step in its own right, not a formality before the "real" math begins.
Top-Down: Start From Industry Reports
Top-down sizing starts with a big, credible number and narrows it down to your segment. Find a total figure for your category's spend — from an analyst report, a trade association, a government statistics agency, or a public company's investor filings that discloses market size as part of its own pitch to shareholders — then apply a series of narrowing filters: the percentage of that spend that falls in your geography, the percentage in your target company-size band, the percentage relevant to the specific problem you solve rather than the whole category.
Each filter should be justified by something more concrete than a guess, even a rough one: a company-size distribution from a business-statistics agency, a geographic split from the same report you pulled the top-line number from, a "% relevant to us" filter based on which of a category's sub-segments your product actually addresses. Stack three or four filters like this and a broad category figure narrows into something that resembles your actual TAM.
The weakness of top-down is that it's only as good as the report you started from — a stale number, a category defined more broadly than your product, or a filter you couldn't justify all creep into the final figure without announcing themselves. Where to actually find the reports and public filings this method depends on is covered in more depth in how to research market size using only free public data; the short version is that paid analyst subscriptions are rarely necessary for an early-stage sizing exercise.
Bottom-Up: Build From Units and Price
Bottom-up sizing starts from your own unit economics and builds up, instead of narrowing down. Count how many realistic buyers exist in your defined segment — using a business directory, a professional network's company search, industry association membership counts, or your own outreach data if you've already started selling — then multiply that count by the price you'll actually charge, adjusted for how often they'd pay (monthly, annually, per seat).
The formula is simple on purpose: number of addressable buyers × price × purchase frequency = market size. What makes it credible or not is entirely in how you counted the buyers. "There are 30 million small businesses in the US" is not a buyer count for a product that only fits 10-50 person marketing agencies — that number has to come from actually filtering down to companies that match your segment definition, the same segment you defined in the prerequisites above.
Bottom-up numbers tend to run smaller and more defensible than top-down ones, because every input traces back to something you can point to and explain. Its weakness is the mirror image of top-down's: a buyer count that's too narrow (you only searched one directory) or a price that's optimistic relative to what the market will actually bear can quietly understate or overstate the real number without a report to check it against.
Reconcile the Two Numbers
If your top-down and bottom-up TAM land within roughly the same order of magnitude, that agreement is itself evidence the numbers are sound. If they're wildly apart — one method says the market is $50 million and the other says $2 billion — the gap is a diagnostic, not a nuisance to smooth over by picking the bigger number.
Work through the usual suspects in order: check whether your top-down segment filters were narrow enough, since it's the more common failure mode — an unfiltered "% relevant to us" step routinely overstates a market by an order of magnitude. Check whether your bottom-up buyer count actually covers the full segment or just the slice you happened to find in one directory. Check whether the two methods are quietly using different prices or different purchase frequencies without you noticing.
Once the two numbers are within a defensible range of each other, use the bottom-up figure as your primary SAM and SOM basis — it's built from assumptions you can defend individually — and keep the top-down figure as your TAM headline and as a sanity check whenever you update either one.
Top-Down vs. Bottom-Up: Which Data Sources to Use for Each
The two methods pull from different kinds of sources and fail in different ways, which is exactly why running both catches mistakes that either one alone would miss.
| Top-Down | Bottom-Up | |
|---|---|---|
| Starting point | A total category spend figure | Your own unit economics |
| Typical sources | Analyst and trade-association reports, government statistics agencies, public company investor filings | Business directories, professional-network company search, industry association membership counts, your own sales/outreach data |
| Main strength | Fast, and lends outside credibility if the source is reputable | Every input is traceable and defensible on its own |
| Main failure mode | An unfiltered or stale category number overstates the market | An incomplete buyer count or optimistic price understates or overstates it |
| Best used for | The TAM headline and a sanity check | SAM and SOM, where precision matters more |
Read the table as a division of labor rather than a choice between two competing methods: top-down gives you a fast, citable ceiling; bottom-up gives you a number you can defend line by line when someone asks how you got it. A sizing exercise that only does one is missing the check the other provides.
Worked Example: Sizing a B2B Expense-Automation Tool
The numbers below are invented for illustration only — a hypothetical product, hypothetical buyer counts, and a hypothetical price, worked through both methods so the mechanics are concrete. None of these figures describe a real market; substitute your own segment, count, and price before trusting the shape of the output.
The product: an expense-automation tool built for marketing agencies with 10-50 employees in the US that currently reconcile client-billable expenses by hand.
Bottom-up. Say a business directory search for US marketing agencies in that size band returns roughly 18,000 companies. Assume the tool sells for $200/month per company (not per seat), and that essentially all of them would pay monthly if they bought at all. That's 18,000 × $200 × 12 = $43.2 million as a bottom-up TAM.
Top-down. Say an industry report puts total US marketing-agency software spend (project management, billing, expense tools combined) at a hypothetical $2 billion annually. Filtering to the 10-50 employee band (assume that's roughly 20% of agency headcount, based on a business-size distribution) narrows it to $400 million. Filtering again to just expense/billing tools specifically, rather than the whole software category (assume roughly 10-12% of agency software spend), lands around $40-48 million.
| Sizing method | Result |
|---|---|
| Bottom-up (buyer count × price) | $43.2 million |
| Top-down (category spend, filtered) | $40-48 million |
The two methods land within a few million dollars of each other — close enough to trust the shape of the number, even though every input on both sides was a stated assumption rather than a verified statistic. That's the TAM. SAM narrows further to whatever slice this specific company can actually reach: if the go-to-market plan only covers agencies findable through two specific marketing-agency directories and a handful of industry Slack communities, SAM might reasonably be a third to a half of that TAM, not the whole thing. SOM narrows once more to what's realistic in the first 18-24 months given one founder-led sales motion and no existing brand recognition — commonly a low single-digit percentage of SAM, not of TAM, and even that requires a stated reason (a specific sales capacity, a specific conversion assumption) rather than a round number picked because it sounds achievable.
Mistakes: The "We Just Need 1% of a Huge Market" Fallacy
Claiming SOM by taking 1% of TAM and calling it conservative. This is the single most common mistake in market-sizing slides, and investors who've seen enough pitches recognize it instantly: 1% of a market you haven't described how you'd reach is not a conservative estimate, it's a number with no path attached. SOM has to come from something concrete — a sales capacity you actually have, a conversion rate from a channel you've tested, a customer count you can plausibly reach with the team and budget you're describing in the same pitch.
Sizing the category instead of the customer. A market report on "business software" or "SaaS tools" describes a category so broad it includes competitors you'll never actually compete with. If your top-down filters don't narrow all the way down to companies that would genuinely consider your specific product, the resulting TAM is a category size, not a market size.
Treating SAM as if it equals TAM. SAM is supposed to be smaller than TAM — it's filtered by what your current business model, geography, and go-to-market motion can actually reach. A pitch where SAM and TAM are suspiciously close usually means the SAM filter never actually happened.
Using a report number without checking its age or definition. Industry figures get quoted for years after publication, and a category's total spend can shift substantially in that time, especially in fast-moving technology segments. Check the publication date and the report's own definition of the category before you inherit its number as if it were current.
Skipping the reconciliation step entirely. Presenting only whichever of the two methods produced the bigger number, without mentioning the other, is the fastest way to get caught by anyone who runs the other calculation themselves — which, for anything approaching a fundraise, someone usually will. The reconciliation work described above is what turns a market-size slide from a guess into an analysis.
Key Takeaways
- TAM, SAM, and SOM are three progressively narrower slices — total category spend, the portion you can reach, and the portion you can realistically win — not three independent guesses.
- A defined customer segment and a price point are prerequisites, not optional detail. Neither top-down nor bottom-up produces a usable number without both.
- Top-down narrows a big report figure through justified filters; bottom-up multiplies a real buyer count by your actual price. Each checks the other's blind spots.
- A wide gap between the two methods is a signal to investigate, not a chance to pick the bigger number. Work through segment filters, buyer counts, and pricing assumptions until the two agree.
- SOM needs a stated path — a sales capacity, a channel, a conversion rate — not a round percentage of TAM. "1% of a huge market" is the mistake investors spot fastest.
- Report numbers age. Check the publication date and the category definition before inheriting a figure as current.
Frequently Asked Questions
What's the difference between TAM, SAM, and SOM in plain terms?
TAM is every dollar spent on your category by anyone, anywhere. SAM is the slice of that you could actually reach given your current business model, geography, and go-to-market approach. SOM is what you can realistically capture in the next year or two given your actual resources, price, and competition — the three narrow in that order, each one a subset of the one before it.
How do investors actually use TAM SAM SOM numbers?
Investors use the TAM figure to gauge whether the category is large enough to support a venture-scale outcome, and use the SAM and SOM figures — along with how you got them — to judge whether you understand your own business or are pattern-matching a pitch-deck template. A defensible bottom-up SOM with a stated path to reach it tends to land better than a large TAM with no SAM/SOM narrowing at all, which is part of what gets scrutinized in preparing for a Series A raise.
Should I use top-down or bottom-up as my primary method?
Use both, and treat disagreement between them as the useful signal rather than picking a favorite. Bottom-up tends to be more defensible for SAM and SOM specifically, because every input traces back to something you can point to; top-down is faster and lends outside credibility for the TAM headline. A sizing exercise built on only one method is missing the check the other provides.
How often should I recalculate TAM, SAM, and SOM?
Recalculate whenever a core input changes — your price, your defined customer segment, or a new industry report that supersedes the one you cited — and treat a market-sizing exercise more than a year or two old as due for a refresh even if nothing else changed, since category spend and buyer counts drift over time. Early-stage companies often revisit the numbers every funding round, since the segment definition itself tends to sharpen as the product and go-to-market motion mature.