Calculate TAM from the full, clearly defined customer universe and annual revenue per customer. Narrow that figure to the customers your product and go-to-market model can serve to estimate SAM, then estimate SOM from the customers you can plausibly reach and win within a stated period. The figures are useful only when their boundaries, evidence and assumptions are clear.
Define TAM, SAM and SOM
These are connected layers of a market estimate, not three independent guesses. Keep the customer unit, revenue basis, currency, geography and time period consistent as you move from the broad opportunity to the realistically obtainable portion.
- TAM (total addressable market): the full revenue opportunity if every potential customer for the defined product were served. Amazon Ads describes the calculation as potential customers multiplied by average revenue per customer.
- SAM (serviceable addressable market): the part of that opportunity the business can serve under its product, geographic, audience and channel constraints. Some sources use “serviceable available market”; state which expansion you use. Amazon Ads, GOV.UK and Startups.com discuss these terms.
- SOM (serviceable obtainable market): the portion of SAM the startup can realistically capture given its resources, strategy, traction, competition and timeframe.
Build the estimate from the customer unit up
A bottom-up model makes its drivers visible: how many eligible buyers exist, what they may pay, and what share the company can actually reach and convert.
- Choose the unit and buyer. For a business-to-business product, the unit may be a qualifying account; for a consumer product, it may be an individual or household. Define what makes a buyer eligible.
- Set the revenue basis. Use a consistent period, commonly annual revenue per customer. Explain whether the input is list price, observed contract value or another defensible measure; do not mix annual revenue with lifetime value without explicitly changing the model.
- Count potential customers. Use a traceable source or documented customer research. Record the source date, geography and any exclusions.
- Calculate TAM. Multiply the full relevant customer count by annual revenue per customer. Amazon Ads describes this basic structure as potential customers multiplied by average revenue per customer: Amazon Ads’ TAM guide.
- Filter for SAM. Apply explicit limits imposed by the current product, geography, customer profile, regulation and channels. The result should describe customers the business can serve, not merely a segment the founder prefers.
- Estimate SOM from an acquisition plan. Identify target accounts or buyers, sales capacity, expected conversion or win rates, competitive conditions and a time period. GOV.UK gives target accounts × expected average contract value × win rate over 24–36 months as one sales-based framing; it is guidance, not a universal formula.
- Make the model reproducible. Record formulas, sources, dates, units, currency, geography and assumptions so another person can follow the calculation.
The layers should connect logically: TAM → SAM → SOM. A broad industry total multiplied by an unexplained percentage does not establish a startup-specific forecast. Top-down market data can still provide context or a reasonableness check, but the customer and pricing inputs should support the startup’s own estimate.
#1 Best Overall
Choose between top-down and bottom-up sizing
| Dimension | Top-down | Bottom-up |
|---|---|---|
| Starting point | Published industry or category total, then segmentation | Defined customer units and revenue per unit |
| Speed | Usually faster when a relevant report exists | Requires customer counts, pricing and scope assumptions |
| Traceability | Depends heavily on the report’s scope and the assumptions used to segment it | Makes customer and price inputs explicit |
| Best use | Market context and cross-checking | Startup-specific serviceable estimates and sales logic |
| Common failure | Applying an unsupported share of a broad category | Counting customers too broadly or assuming unrealistic pricing or conversion |
Startups.com describes these trade-offs, while GOV.UK emphasizes evidence, a logical connection between the market layers and a credible strategy. See Startups.com’s TAM, SAM and SOM explainer and GOV.UK’s investor-readiness guidance.
Make SOM credible with evidence and a time horizon
SOM is not a percentage chosen to make the opportunity look compelling. It is a reasoned estimate of what the business can win, based on execution capacity and a defined period. GOV.UK’s 2025 investor-readiness guidance uses 3–5 years as an example horizon for an evidence-based SOM target and says a 1–5% share of SAM may support credibility when assumptions are explicit. These are examples of guidance—not measured market outcomes, universal benchmarks or guarantees. The same guidance’s 24–36-month sales framing is another illustrative option, not a required forecast period.
Rank #2
For a sales-led startup, a model can begin with target accounts, expected average contract value and win rate over the chosen period. Check that the target volume fits the team’s sales capacity and that the assumed wins reflect competition, the sales cycle and actual traction. For other business models, use a similarly explicit acquisition logic suited to the channel and buyer.
Document inputs and test uncertainty
Give each material input an owner, date and source: customer counts, market boundaries, pricing or contract value, adoption assumptions and conversion rates. Prefer original government or industry data and actual customer evidence where available. Keep units and geography consistent, and label observed facts separately from estimates.
Use scenarios to show how the result changes when customer count, price, reach or conversion assumptions change. Label them as scenarios rather than facts. A precise-looking total is not more reliable if its underlying inputs cannot be checked.
Common market-sizing mistakes
- Calling an adjacent industry’s total the startup’s TAM without defining the product and eligible buyer.
- Claiming the company will capture a fixed percentage without evidence for its buyers, geography, channel, sales capacity and timeframe.
- Mixing annual and lifetime revenue, currencies, geographies, customer types or periods across TAM, SAM and SOM.
- Treating SAM as a preferred customer segment instead of the market the current product and go-to-market model can serve.
- Calling an aspiration SOM without connecting it to acquisition capacity, win rates, competition and a period.
- Presenting a precise total built on customer counts or assumptions that cannot be verified.
What a useful estimate should show
A reader should be able to see who might buy, what revenue basis is used, how the product and operating model narrow the market, and why the proposed share is attainable within the stated period. GOV.UK’s investor-readiness guidance says: “Being able to clearly evidence the size of your market, through credible, data-backed analysis of the Total Addressable Market (TAM), Serviceable Available Market (SAM) and Serviceable Obtainable Market (SOM), is essential for any business seeking investment.” Read the guidance in full.
Quick Recap
Best Value
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
