Market sizing is the process of estimating the total revenue, unit volume or number of customers that a defined market can support within a specific period, usually one year. In market research it turns a vague sense of opportunity into numbers that can be tested, compared and defended in business plans, investment cases and go-to-market decisions.
A market size is never a single fact. It depends on how the market is defined, which customers and geographies are included, which price assumptions are used and which data sources feed the model. Good market sizing therefore states its assumptions as clearly as its results.
How market sizing works
Market sizing starts with a precise market definition: the product or need, the customer group, the geography and the time frame. An estimate for “project management software” means little until it specifies, for example, small businesses in the European Union buying cloud subscriptions in a given year.
Analysts then build a structured calculation that multiplies the number of potential buyers by their expected purchase volume and price, or that breaks a known aggregate figure down to the relevant slice. The result is usually expressed in currency per year, sometimes also in units or customer counts.
Most estimates are layered into three nested levels, which separate theoretical potential from realistic ambition.
| Level | Meaning | Typical use |
|---|---|---|
| Total addressable market (TAM) | Total demand if every possible customer bought | Long-term vision and investor framing |
| Serviceable addressable market (SAM) | Portion reachable with the current product, channels and geography | Strategic planning and segment choice |
| Serviceable obtainable market (SOM) | Share realistically winnable in the next few years | Sales targets and financial forecasts |
Top-down and bottom-up methods
The two classic approaches are top-down and bottom-up sizing, and strong analyses use both and reconcile the gap. A top-down estimate starts from a large published figure, such as total industry revenue, and narrows it with percentage filters. A bottom-up estimate starts from individual customers and multiplies upward.
| Method | Starting point | Strength | Main weakness |
|---|---|---|---|
| Top-down | Industry totals from secondary research | Fast and easy to communicate | Filters are often guesses, results tend to be inflated |
| Bottom-up | Customer counts multiplied by price and volume | Transparent and tied to real buying behavior | Requires reliable unit data, more effort |
| Value theory | Value the product creates for a buyer | Useful for new categories with no history | Highly sensitive to willingness-to-pay assumptions |
| Analog comparison | Size of a similar market elsewhere | Works when local data is missing | Markets may differ in maturity and behavior |
When top-down and bottom-up results differ by more than a factor of two or three, it usually signals a flawed assumption. The reconciliation step, where analysts trace and fix that gap, is often where the most valuable insight emerges.
Key inputs and assumptions
Every market sizing model rests on a small set of driver variables, and the quality of the result depends on how each one is sourced. Inputs come from primary research, such as surveys and customer interviews, and from secondary research, such as government statistics, industry reports and company filings.
| Input | Example | Common source |
|---|---|---|
| Number of target customers | Small firms with 10-49 employees in a region | Business registers, census data |
| Adoption or penetration rate | Share of firms that already use such a product | Surveys, interviews, industry studies |
| Purchase frequency | Annual subscription renewed once per year | Customer interviews, pricing models |
| Average price | Annual contract value per customer | Competitor pricing, conjoint analysis |
| Growth rate | Expected annual change in demand | Historical trends, analyst forecasts |
A disciplined model runs a sensitivity analysis on these inputs, typically showing low, base and high scenarios. This reveals which assumptions move the result most, so that further research can focus on the variables with the greatest leverage.
Market sizing compared with related concepts
Market sizing is often confused with forecasting and segmentation, but each answers a different question. Sizing asks how big the opportunity is; forecasting asks how much a specific company will sell; segmentation asks which groups within the market behave differently.
- Total addressable market is one output of market sizing, not the whole exercise.
- Market segmentation often precedes sizing, because each segment is sized separately.
- Sales forecasting uses market size as an upper limit but adds company-specific capacity and win rates.
- Competitor analysis helps convert SAM into SOM by showing how share is currently divided.
Common mistakes in market sizing
The most frequent error is presenting TAM as if it were achievable revenue. Investors and executives quickly discount such estimates, and the credibility of the whole plan suffers. Other mistakes involve definitions, double counting and outdated data.
| Mistake | Effect | Correction |
|---|---|---|
| Defining the market too broadly | Inflated, meaningless totals | Specify customer, need, geography and period |
| The “one percent of a huge market” argument | No explanation of how share is won | Build SOM from channels and sales capacity |
| Double counting across segments | Overstated total | Use mutually exclusive segment definitions |
| Mixing revenue and unit figures | Inconsistent results | Keep one unit of measure per model |
| Relying on a single outdated report | Stale or biased baseline | Triangulate with several current sources |
A further warning concerns new categories. When customers do not yet know a product type exists, historical data cannot show demand, and estimates should be framed as ranges with explicit uncertainty rather than precise totals.
Understanding it simply
Market sizing answers a basic question: how big is the pie? Before a company bakes a new product, it wants to know how many people might buy it and how much they would spend in total.
Imagine opening a coffee stand. You would count how many people walk past each day, guess how many buy coffee and multiply by the price of a cup. That rough calculation is market sizing. The answer is only as good as the guesses behind it, so careful planners check their numbers in more than one way.
Applied in practice
A startup develops scheduling software for independent physiotherapy practices and needs a market size for an investor presentation. A top-down estimate takes published healthcare software spending for the country and applies filters, producing a figure that looks very large but rests on several guessed percentages.
The team then builds a bottom-up model. National registers list about 6,000 independent practices. Interviews suggest that roughly 40 percent already pay for scheduling tools and would switch for better features, and the expected price is 900 euros per practice per year. That gives a serviceable market of about 2.2 million euros per year. With two sales representatives, the team estimates it can win around 300 practices within three years, an obtainable market of roughly 270,000 euros annually.
The investors find the smaller, transparent figure far more convincing than the top-down total, because every assumption can be checked and challenged.