How to calculate TAM, SAM, and SOM (with a worked example)

Foundations · ~9 min read · public-methodology, reproducible on paper

TAM, SAM, and SOM are three progressively narrower cuts of the same market. Most decks show them as three shrinking circles and stop there. The value is entirely in the arithmetic behind each step — so here it is, worked end to end on a warehouse-robotics example, using the exact chain the SectorIntel engine runs.

The three layers, defined

• TAM — Total Addressable Market. Annual revenue if every buyer who could adopt the category actually did, at a realistic price. It's the ceiling, not a forecast.
• SAM — Serviceable Addressable Market. The slice of TAM your specific offer can serve — narrowed by geography and by the customer segment that actually fits.
• SOM — Serviceable Obtainable Market. The realistic near-term (typically three-year) share of SAM one player can actually win, given competition and go-to-market limits.

The relationship is strictly nested: SOM ≤ SAM ≤ TAM. If your SOM ever exceeds your SAM, the model is broken — that's the first automated check in the section below.

Step 1 — TAM, counted bottom-up

The most defensible way to build a TAM is bottom-up: count the buyers, multiply by what each pays. The formula:

Take warehouse robotics. Using public, illustrative figures you'd cite to census / trade-body data and vendor pricing pages:

These are the same inputs the worked sample uses, so this cut lands on the same $7.56B bottom-up TAM you'll see there. The buyer universe here is already drawn to the ~180,000 large facilities in scope — how tightly you draw that count is the single biggest lever, so it's the one figure you cite most carefully.

Bottom-up is the most honest method because every factor is a countable, citable number — but it's also the easiest to lowball if you draw the buyer universe too tightly. That's why it's never used alone; see the top-down vs bottom-up guide and the triangulation guide for the two cross-checks that anchor it. In the worked sample, triangulation lands the reported TAM at a $9.00B median, with bottom-up ($7.56B) on the low side of the three methods.

Step 2 — SAM, narrowing TAM to what you can serve

TAM is the whole category. SAM asks: of that, how much can this offer actually reach? Two multipliers do the narrowing — served geography and served segment.

Suppose your offer covers a multi-region footprint (US + EU + China + Korea ≈ 70% of the global market) and fits about 55% of facilities within those regions (the segment your product actually serves — say, mid-to-large fulfilment centres, not cold-storage or micro-DCs). On the $9.00B triangulated TAM:

The two most abused SAM levers are geography and segment. State them explicitly, cite the geography split, and define the segment in one sentence — "mid-to-large fulfilment centres with >X throughput" — so a reader can challenge each multiplier on its own.

Step 3 — SOM, the reachable slice

SOM is the number a founder or corp-dev team actually gets held to. It's SAM times a realistic near-term capture share:

A reachable share of 12% of the SAM over three years — credible for a strong entrant against real incumbents — gives:

The sanity check that stops a wishful SOM

A SOM is only credible if it's a modest share of its SAM. Capturing more than ~30% of a served market in three years implies near-monopoly, and an investment committee will discount it on sight. In the worked example:

This is one of five automated gates every figure clears before a report ships — full list in the sanity-checks guide.

Try it on your own market

The calculator on the landing page runs a simplified version of this exact chain — top-down / bottom-up / value-theory, then the SAM/SOM narrowing — on your own parent-market size, sector archetype, and geographic breadth, in about ten seconds.