How to Calculate a Bottom-Up TAM (Step by Step)

A bottom-up TAM is the number of potential customers multiplied by the annual revenue each one generates at your price. Instead of shaving a headline market figure down with guesswork, you build the total from countable units — who the buyers are, how many exist, and what they pay — so every input can be defended.

Quick Answer: Bottom-up TAM = potential customers × annual revenue per customer (ACV). Count the buyers from real sources — registries, associations, platform directories — set a price you can defend from your own deals, and multiply, ideally segment by segment. It beats top-down because investors can challenge each input, not just the headline.

Top-down sizing hands you a big number and dares you to trust it. Bottom-up sizing hands the skeptic a number they can take apart — and that is exactly why it survives due diligence. This guide walks the build end to end: define the unit you're counting, count it, price it, multiply, then narrow the result to the slice you can actually win.

Prerequisites: Define Your Unit of Customer Before You Count Anything

Before you count anything, decide what one "customer" is — the billing account you invoice, not the individual user or the parent company. Bottom-up TAM multiplies a customer count by revenue per customer, so a fuzzy unit corrupts both numbers at once.

The unit is whatever you send an invoice to. For a per-seat SaaS tool it might be a company; for a marketplace, a merchant; for a prosumer app, one individual. Pick the entity that signs the contract and pays the annual fee, because that is the thing you will both count and price.

Define the buyer tightly enough to be countable. "Small businesses" is not a unit; "independent physiotherapy clinics with two to ten practitioners" is. The tighter the definition, the easier it is to find a real source for how many exist — and the harder it is for a reviewer to accuse you of hand-waving. That precision is the whole reason bottom-up beats top-down for a startup, and it anchors any credible startup market sizing exercise.

Fix the revenue unit to a year. Bottom-up TAM is an annual figure, so revenue per customer means annual contract value (ACV) — what one account pays across twelve months, not a monthly price or a one-off transaction. If you bill monthly, multiply by twelve; if customers buy sporadically, estimate annual spend per account. Slipping a monthly price into an annual TAM understates the market by 12×, a mistake that is embarrassingly common on early decks.

The Bottom-Up TAM Formula: Count Customers, Set Price, Multiply

The formula is simple: bottom-up TAM = number of potential customers × annual revenue per customer (ACV). The difficulty is not the arithmetic; it is sourcing each factor honestly and splitting the market into segments whose prices actually differ. Run it in three steps.

Step 1: Count the Potential Customers in Your Defined Segment

Count every entity that matches your buyer definition and could plausibly buy — not everyone in the broad category, only those who fit. This is a headcount of accounts, sourced from data rather than rounded off a hunch.

Start from a real registry, not a round number. Business counts come from census and statistical agencies, industry associations, professional licensing bodies, business registries, and platform directories. A round "let's say a million" is the tell of a made-up market; a sourced count is the tell of a real one.

Count the addressable subset, not the universe. If your tool only works for firms already using cloud accounting, count those firms, not every firm. Narrowing here is legitimate as long as you can name the filter — and it keeps you honest later when you narrow TAM to SAM.

Step 2: Set the Annual Revenue Per Customer (ACV)

Set ACV to the annual price a customer would realistically accept — anchored in your own signed deals, competitor pricing, or willingness-to-pay conversations, not the number you wish you could charge.

Ground the price in evidence you already hold. Your pricing page, closed deals, signed letters of intent, and even lost-deal feedback are the strongest sources. Absent your own data, competitor pricing for a comparable product is the next-best anchor.

Use a distinct price per segment. A ten-person firm and a thousand-person firm do not pay the same, and blending them into one average distorts the total. Price each segment on its own, then stack the segments — which is exactly what Step 3 does.

Step 3: Multiply, Then Stack the Segments

Multiply each segment's customer count by its ACV, then add the segments together. A single blended multiplication hides the fact that most markets are really several small markets with very different economics.

Segment-level math beats one big multiplication. Splitting by company size, vertical, or plan tier turns one fragile number into several defensible ones. If a reviewer challenges your enterprise price, only the enterprise segment moves — the rest of the model stands.

The sum of the segments is your bottom-up TAM. That total is a ceiling: the annual revenue if every reachable customer in every segment bought at your price. It is not a forecast, and the next job is to narrow it — but first, source the inputs so the number holds up.

Where to Source Each Input: Customer Counts and Price

Every factor in a bottom-up TAM has a place you can source it from, which is what separates a defensible model from a fabricated one. The table below maps each step to the input it needs and where founders actually find it.

Takeaway: If you cannot name a source for a number, it is not ready for a slide — replace the guess with a sourced figure or a clearly labeled estimate. When your own customer count is thin, the analog method using comparable companies borrows numbers from businesses that resemble yours, which is often steadier than a solo guess.

Worked Example: A Hypothetical Bottom-Up TAM, Segment by Segment

Here is a full multi-segment build. Every figure below is invented purely to demonstrate the method — these are not researched market numbers, and you should never quote a teaching figure like this as a real market size.

Imagine a hypothetical compliance tool sold to accounting firms, billed per firm per year, with pricing tiered by firm size. Suppose — for illustration only — the addressable firms and prices break down like this:

Add the three segment totals and the illustrative bottom-up TAM is $286 million — a made-up teaching figure, not a claim about any real market. Notice what the segmentation reveals: the largest pool of value sits with just 5,000 large firms, even though small firms outnumber them sixteen to one. A single blended average would have buried that insight.

Every input here is contestable — that is the point. Are there really 80,000 small firms that fit, or does that count include firms too small to buy? Would a large firm pay $20,000, or is $12,000 more honest? Change any assumption and the total moves, which lets a reviewer argue with an input instead of dismissing the whole model.

Narrowing the Illustrative TAM to SAM and SOM

Bottom-up TAM is the ceiling; SAM and SOM narrow it to what you can serve and then win. The numbers below stay invented — the method matters, not the figures.

SAM removes what you cannot serve. Suppose your tool only integrates with cloud accounting systems and you launch in one country. Cutting firms outside that geography and on incompatible software might leave roughly 60% of the market — an illustrative SAM near $170 million. Each cut needs a stated reason, never a percentage chosen because it felt realistic.

SOM is the share you can actually win soon. Based on your funnel capacity and win rate, you might capture 3% of that SAM over three years — an illustrative SOM around $5 million. That figure describes how you start: through named channels at a stated conversion rate. For the full narrowing exercise across all three layers, see how to calculate TAM, SAM, and SOM without a consulting firm.

Bottom-Up TAM Mistakes That Undermine the Number

Most bottom-up TAM errors come from one of two habits: sneaking a guess in where a source belongs, or blending segments that should be priced separately. Avoid the handful below and the model survives a hard conversation.

• Using a monthly price in an annual TAM. Revenue per customer must be annual (ACV); drop a monthly figure in and you understate the market twelvefold.
• Counting the universe instead of the addressable subset. If half the category cannot use your product, they are not in your TAM — inflating the count is the bottom-up version of quoting a headline report.
• Blending segments into one average price. A single ACV across wildly different buyers hides where the value actually sits and invites a reviewer to pick the number apart.
• Rounding the customer count off a hunch. "Let's say a million" with no source is the fastest way to lose credibility; every count needs a place it came from.
• Confusing TAM with revenue. Bottom-up TAM is a ceiling assuming everyone buys, not a sales forecast — your projection lives down in SOM.
• Skipping the roll-up to SAM and SOM. A vast TAM with no obtainable slice reads as hand-waving; the narrowed number is what proves you can start.
• Never updating the estimate. Customer counts and prices are hypotheses; refine them as real deals and conversion data arrive, or the number quietly goes stale.

Key Takeaways

• Bottom-up TAM = potential customers × annual revenue per customer. You build the total from countable units instead of shaving a headline figure with guesswork, so every input can be inspected and defended.
• Define the unit of customer before counting. The customer is the account you invoice, defined tightly enough to be countable from a real source — "clinics with two to ten practitioners," not "small businesses."
• Revenue per customer means annual contract value. ACV is what one account pays across a year; slipping a monthly price into an annual TAM understates the market by twelve times.
• Source every factor or label it an estimate. Counts come from registries, associations, and platform directories; prices come from your own deals and competitor pricing — a number with no source is not ready for a slide.
• Segment the market and price each slice separately. Most markets are several small markets with different economics, so stacking priced segments turns one fragile number into several defensible ones.
• TAM is a ceiling, not a forecast. Narrow it to SAM by removing what you cannot serve, then to SOM by the share you can win soon through named channels — each cut needs a stated reason.
• Treat the whole model as a living hypothesis. The strength of a bottom-up TAM is that inputs are contestable, so keep refining the count and the price as real evidence arrives and the figure gets sharper, not louder.

Frequently Asked Questions

What Is the Formula for a Bottom-Up TAM?

Bottom-up TAM equals the number of potential customers multiplied by the annual revenue per customer (ACV). For a segmented market, you run that multiplication for each segment and add the results. The strength of the formula is that both factors are sourced and challengeable, so an investor can question an input rather than dismiss the whole figure.

Is Bottom-Up or Top-Down TAM Better for Startups?

Bottom-up is better for startups because it is built from assumptions investors can inspect — reachable customers times price — rather than a published total shaved with arbitrary percentages. Top-down is useful only as a quick sanity check or upper bound. Lead with the bottom-up number in any deck or go/no-go decision, and use top-down to catch obvious overreach.

How Do You Find the Number of Potential Customers?

Source the customer count from census and statistical agencies, industry associations, professional licensing boards, business registries, and platform or marketplace directories, plus paid company databases for harder segments. Count only the addressable subset that fits your buyer definition, not the whole category. When your own count is thin, comparable companies can supply a defensible proxy.