The Serviceable Obtainable Market (SOM) represents the portion of the Serviceable Addressable Market (SAM) that can realistically be served within a specified geography, market segment, and time period, subject to operational, commercial, and market-capacity constraints.
The Weighted Utilization Capacity methodology calculates SOM by comparing four independently derived constraints:
The methodology uses the smallest of these four quantities as the resulting SOM.
The underlying principle is that obtainable demand cannot exceed the most restrictive applicable constraint, even when other constraints indicate a larger market opportunity.
The purpose of Weighted Utilization Capacity is therefore not simply to estimate the size of the market. It is to establish a defensible, capacity-constrained estimate of the market that can be obtained and to identify the constraint preventing further expansion.
Weighted public proxy =
weighted mean(CBP proxy, ECN proxy, NES proxy)
Operational cap =
open capacity × utilization rate
Public cap =
weighted public proxy × utilization rate
SOM units =
min(
SAM demand units,
campaign forecast units,
operational cap,
public cap
)
All four inputs to the final minimum must use the same unit of measurement, geographic scope, market segment, and time period.
For example, an annual SOM measured in completed service transactions cannot directly compare annual transactions with establishment counts, monthly campaign responses, or physical venue capacity until those quantities have been converted to comparable units.
The model combines internal commercial and operational estimates with public economic information.
| Input | Definition | Source or derivation |
|---|---|---|
SAM demand units |
Total eligible demand within the serviceable market | Upstream TAM/SAM model |
Campaign forecast units |
Expected demand captured through planned campaigns | Campaign forecasting model |
Open capacity |
Gross operational capacity available for the applicable market and period, before the utilization adjustment | Internal operations or capacity model |
Utilization rate |
Proportion of gross capacity expected to be practically usable or realized | Operational assumptions, empirical data, or analyst estimates |
CBP proxy |
Capacity estimate derived from County Business Patterns | U.S. Census Bureau |
ECN proxy |
Capacity estimate derived from the Economic Census | U.S. Census Bureau |
NES proxy |
Capacity estimate derived from Nonemployer Statistics | U.S. Census Bureau |
cbp_weight |
Relative influence assigned to the CBP capacity estimate | Analyst/governance configuration |
ecn_weight |
Relative influence assigned to the ECN capacity estimate | Analyst/governance configuration |
nes_weight |
Relative influence assigned to the NES capacity estimate | Analyst/governance configuration |
The public producer-capacity component uses three economic data sources.
County Business Patterns (CBP)
CBP provides establishment-level economic statistics covering most U.S. employer businesses, classified by industry and geography. Its establishment counts, employment measures, and other applicable variables can support estimates of the business activity or productive capacity associated with a particular market.
Economic Census (ECN)
The Economic Census provides detailed periodic measurements of economic activity across covered industries, including establishment counts, receipts or revenue, employment, and other industry-specific characteristics.
ECN can provide a complementary estimate of market production or commercial activity, particularly where industry-level economic detail is useful.
Nonemployer Statistics (NES)
NES provides information on businesses without paid employees, including establishment counts and receipts.
It supplements the employer-business information found in CBP and much of the Economic Census by representing economic activity that might otherwise be omitted.
The values called CBP proxy, ECN proxy, and NES proxy must represent comparable estimates of capacity, rather than necessarily being the raw measurements received from their respective sources.
The upstream transformation depends on the economic measures available and the type of SOM being estimated.
For a count-based source, a general transformation might be:
Source capacity proxy =
eligible source count
× capacity conversion factor
× period adjustment factor
Where:
This is a general description of the required normalization, not an assertion that every source uses the same conversion formula.
Revenue-based estimates, for example, require a different transformation from establishment-based estimates.
The source-specific transformation should be defined and versioned in the upstream model before the resulting proxies are used in the weighted arithmetic.
All three proxies must represent the same target quantity. A proxy representing establishments cannot be averaged directly with a proxy representing dollars of annual receipts or completed transactions.
Important: CBP and ECN cover overlapping economic populations. Consequently, their derived capacity estimates should not ordinarily be added together. The weighted method treats them as alternative measurements of substantially overlapping market activity, not automatically as separate additive markets.
The Weighted Public Proxy combines the three source-derived capacity estimates into a single public-data estimate.
Rather than accepting one economic data source as definitive, the method allows the model to assign different levels of influence to each source.
The weighted mean is calculated as:
Where:
CBPProxy is the CBP-derived capacity estimate.ECNProxy is the Economic Census-derived capacity estimate.NESProxy is the NES-derived capacity estimate.w_CBP, w_ECN, and w_NES are their assigned weights.The weights are configured using:
cbp_weight
ecn_weight
nes_weight
In the current Rust implementation, only strictly positive weights contribute to the denominator.
For an internally consistent weighted mean, the numerator must likewise include only source terms with eligible positive weights.
The weights determine how strongly each source influences the combined capacity estimate.
They can reflect factors such as:
Weights are governance assumptions, rather than properties automatically established by the underlying Census data.
A weight of 0.50 does not mean CBP has a statistically established 50% probability of being correct. It means the model gives CBP half of the combined influence when the effective weights sum to 1.
The weights should therefore be documented, justified, and version-controlled.
Assume three public-data proxies have been converted into comparable annual market-capacity units.
CBP proxy = 10,000 units
ECN proxy = 8,000 units
NES proxy = 2,000 units
CBP weight = 0.50
ECN weight = 0.35
NES weight = 0.15
The weighted mean is:
First, calculate each weighted contribution:
| Source | Proxy | Weight | Weighted contribution |
|---|---|---|---|
| CBP | 10,000 | 0.50 | 5,000 |
| ECN | 8,000 | 0.35 | 2,800 |
| NES | 2,000 | 0.15 | 300 |
| Total | 1.00 | 8,100 |
The calculation becomes:
The combined public-data estimate is therefore 8,100 annual capacity units.
This does not establish that the market contains precisely 8,100 obtainable units. It establishes that the weighted public-data evidence supports an estimated gross capacity of 8,100 units, before utilization is applied.
Weights do not have to sum to 1.0.
For example, the following weights produce the same result:
CBP weight = 5.0
ECN weight = 3.5
NES weight = 1.5
The numerator becomes:
The denominator becomes:
Therefore:
The relative proportions, rather than their absolute numerical values, determine the weighted mean.
A source with a weight of zero has no influence on the resulting estimate.
A source with a negative weight should not be used in a conventional weighted mean. Negative weights should be rejected or excluded according to the configured validation policy.
If every effective weight is zero, the denominator is zero and the weighted mean is undefined. The implementation must explicitly handle this situation rather than perform division by zero.
Missing source data must also be distinguished from a valid source measurement of zero.
For example:
The calculation should use only valid, available source proxies with eligible weights.
The Weighted Public Proxy represents an estimate of gross market capacity.
Not all gross capacity is expected to be available, productive, accessible, or otherwise realizable during the relevant period.
The model therefore applies a utilization rate:
Public cap =
weighted public proxy × utilization rate
Mathematically:
Where U is the utilization rate, expressed as a decimal between zero and one.
Using the previously calculated weighted proxy:
Weighted public proxy = 8,100 units
Utilization rate = 70%
Convert the utilization rate into decimal form:
Apply the utilization rate:
Under the assumed utilization rate, the public economic data supports an estimated effective market capacity of 5,670 annual units.
The 70% rate is an assumption about realizable capacity. It should not automatically be interpreted as an observed utilization rate or as a measure of capacity already occupied.
The Operational Capacity Cap estimates the quantity the business can realistically deliver using its own operational resources.
Unlike the Public Capacity Cap, which derives from external economic indicators, the Operational Capacity Cap comes from internal operating assumptions and measurements.
The calculation is:
Operational cap =
open capacity × utilization rate
Mathematically:
Open capacity represents the gross quantity of deliverable units available during the measurement period before applying the utilization rate.
Depending on the business model, it might be based on:
If open capacity is derived from several operating resources, its calculation must be defined by the applicable upstream operational model.
For example, a simple location-based model might calculate:
Open capacity =
participating locations
× eligible units per location per period
That is an illustrative capacity transformation. More complex operational systems may require separate constraints for facilities, staffing, resources, scheduling, and throughput.
For this SOM variant, open capacity should represent capacity before utilization has been applied. Otherwise, multiplying it by the utilization rate would discount the same capacity twice.
Assume:
Open capacity = 11,000 units
Utilization rate = 70%
Then:
The business is estimated to have the operational ability to deliver 7,700 annual units.
This is an upper bound on obtainable demand, not evidence that 7,700 units of customer demand will actually materialize.
The model applies the same utilization rate to operational and public capacity:
This makes both capacities subject to a common realizability assumption.
However, internal operational utilization and market-wide producer utilization are not necessarily identical in the real world.
The use of one shared rate is therefore a modeling assumption that should be documented.
If subsequent versions distinguish internal operating efficiency from market-wide capacity realization, the model may use separate utilization rates for the two constraints.
SAM demand represents the quantity of demand within the market the business is capable of serving, before the additional constraints represented by the SOM calculation are imposed.
This value is generated by the upstream TAM/SAM methodology.
SAM demand units =
eligible demand within the serviceable market
The applicable upstream model is responsible for determining the eligible geography, customer population, product or service category, and measurement period.
The SOM calculation receives the resulting demand estimate as an input.
It does not recalculate TAM or SAM, nor does it independently establish the assumptions used in deriving those estimates.
Consequently:
The SOM cannot exceed the demand available within the modeled SAM.
For example, if SAM demand is 9,000 annual units, the obtainable market cannot exceed 9,000 annual units regardless of campaign expenditure or operating capacity.
The campaign forecast represents the quantity of demand expected to be captured through planned marketing, customer acquisition, distribution, or other commercial activities.
It is a forecast of expected obtainable demand, not a measurement of total market demand.
Campaign forecast units =
expected units obtained through planned campaigns
The campaign forecast is produced upstream and supplied to the SOM model.
A generic campaign forecast might use:
Campaign forecast units =
eligible opportunities
× expected conversion rate
× units per conversion
The actual formula depends on the campaign forecasting model and the definition of an obtainable unit.
Campaign-specific factors may include geographic coverage, channel reach, conversion performance, participation rates, and campaign duration.
For example:
Campaign forecast = 6,200 units annually
Even if the market can support a larger volume and the business has sufficient operational capacity, the modeled obtainable market cannot exceed the campaign forecast under this methodology.
Thus:
The campaign forecast provides a commercially grounded constraint that is independent of overall market size and available production capacity.
After all four constraints have been calculated, SOM is determined by selecting the smallest value.
SOM units =
min(
SAM demand units,
campaign forecast units,
operational cap,
public cap
)
Mathematically:
Where:
| Symbol | Meaning |
|---|---|
| (D_{SAM}) | SAM demand units |
| (F_{Campaign}) | Campaign forecast units |
| (C_{Open}) | Gross open operational capacity |
| (P_{Weighted}) | Weighted public-data capacity proxy |
| (U) | Utilization rate |
Assume the following inputs for a single market and annual measurement period.
| Input | Value |
|---|---|
| SAM demand | 9,000 units |
| Campaign forecast | 6,200 units |
| Open operational capacity | 11,000 units |
| Utilization rate | 70% |
| CBP proxy | 10,000 units |
| ECN proxy | 8,000 units |
| NES proxy | 2,000 units |
| CBP weight | 0.50 |
| ECN weight | 0.35 |
| NES weight | 0.15 |
Step 1: Calculate the Weighted Public Proxy
Step 2: Calculate the Operational Capacity Cap
Step 3: Calculate the Public Capacity Cap
Step 4: Compare the four constraints
| Constraint | Candidate SOM units |
|---|---|
| SAM demand | 9,000 |
| Campaign forecast | 6,200 |
| Operational capacity cap | 7,700 |
| Public capacity cap | 5,670 |
Step 5: Select the smallest value
The estimated Serviceable Obtainable Market is therefore 5,670 annual units.
The binding constraint is the Public Capacity Cap.
Although the SAM contains 9,000 units of modeled demand, the campaign forecast supports 6,200 units, and operational resources can support 7,700 units, the utilization-adjusted public economic evidence constrains the estimate to 5,670 units.
The calculation therefore identifies both an obtainable-market estimate and the reason that estimate is smaller than the available SAM.
The binding constraint is the input responsible for limiting the final SOM result.
Each SOM run should retain all four candidate constraints together with an identifier for the constraint selected by the minimum operation.
Possible binding constraints are:
| Binding constraint | Interpretation |
|---|---|
SAM demand |
Insufficient serviceable demand is the principal modeled limit |
Campaign forecast |
Commercial acquisition or campaign performance limits obtainable demand |
Operational capacity |
Internal delivery resources limit obtainable demand |
Public producer capacity |
External economic evidence indicates a smaller realizable market capacity |
This identification is one of the most useful analytical outputs of the methodology.
It distinguishes between different obstacles to market growth.
For example, if campaign forecast binds the result, additional operational capacity alone will not increase SOM. Conversely, if operational capacity binds, more effective marketing may not increase the obtainable volume unless delivery capacity also increases.
If two or more constraints share the same minimum value, they are jointly binding. An implementation that records only one binding-constraint identifier should use a documented tie-breaking rule and, preferably, preserve the full set of tied constraints.
The binding constraint also identifies where further investigation may be most useful.
A public-capacity-bound SOM can be tested under different economic-source weights or utilization assumptions.
An operationally bound SOM can be tested against alternative capacity and resource-allocation assumptions.
A campaign-bound SOM can be tested against alternative reach, participation, or conversion forecasts.
However, increasing the currently binding constraint does not necessarily increase SOM indefinitely. Another candidate becomes binding as soon as the original constraint exceeds the next-smallest value.
This behavior is a direct consequence of using the minimum of multiple independent constraints.
An alternative method of estimating public producer capacity uses the following construction:
Public proxy =
positive_min(
CBP proxy,
ECN proxy + NES proxy
)
This method should be distinguished from the Weighted Public Proxy.
The positive-minimum method constructs two alternative capacity estimates:
It then selects the smaller strictly positive valid estimate.
Assuming both are positive and valid:
Using the example source values:
CBP proxy = 10,000
ECN proxy = 8,000
NES proxy = 2,000
Calculate the combined ECN and NES estimate:
Compare it with CBP:
Therefore:
At 70% utilization, the corresponding public capacity cap is:
This differs from the weighted method's public capacity cap of 5,670 units.
The two approaches answer related but different modeling questions.
Positive-minimum methodology:
What is the smaller valid capacity estimate when comparing CBP against the combined ECN and NES estimate?
This selects the more restrictive of two alternative public-data constructions, provided both are valid and positive.
Weighted methodology:
What capacity estimate results when CBP, ECN, and NES are treated as separate measurements with explicitly assigned influence?
This combines evidence using a weighted arithmetic average, rather than choosing the smaller of two aggregate estimates.
| Characteristic | Positive minimum | Weighted mean |
|---|---|---|
| Public-data estimates | CBP versus ECN + NES | CBP, ECN, and NES individually |
| Aggregation | Select minimum valid positive value | Weighted arithmetic average |
| Source weights | Not required | Required |
| Relative source influence | Determined by the minimum comparison | Determined by configured weights |
| Treatment of ECN and NES | Added together | Averaged as separate source estimates |
| Primary objective | Restrictive comparison between alternative estimates | Blended estimate reflecting source influence |
The two results are not necessarily ordered. A weighted mean can be lower or higher than the positive-minimum construction depending on the values and weights.
The expression ECN proxy + NES proxy assumes the two inputs represent additive, non-overlapping components of the same target capacity measure.
This is reasonable only if their upstream transformations make them compatible and additive.
By contrast, the weighted mean assumes each input represents a comparable estimate of the same target capacity quantity.
These assumptions are not interchangeable.
For example, if ECN represents employer-business capacity and NES represents additional nonemployer-business capacity, their sum may be intended to estimate total employer-plus-nonemployer capacity.
However, treating those two components as separate, alternative measurements in a weighted average changes the meaning of the result.
Therefore, the Weighted Utilization Capacity model must explicitly establish whether CBP, ECN, and NES are independent full-market estimates or complementary market components before applying the weighted mean.
If the sources represent different components rather than alternative measurements, their upstream normalization must account for that distinction. Otherwise, the weighted average may systematically understate capacity.
Likewise, positive_min requires a documented policy for zero and missing values. Selecting only positive values is different from taking an ordinary minimum, and a genuine zero should not automatically be treated as missing.
For the resulting SOM to be analytically defensible, the inputs and assumptions must be consistent and auditable.
All inputs must be expressed in compatible measurement units.
At minimum, the model should enforce or document consistency across:
For example, a CBP-derived annual transaction estimate should not be directly combined with an NES-derived business count or a quarterly campaign forecast.
The model should validate the following conditions:
SAM demand units >= 0
Campaign forecast units >= 0
Open capacity >= 0
Source capacity proxies >= 0
0 <= utilization rate <= 1
Effective source weights > 0
Sum of effective source weights > 0
The source-weight condition applies to weights included in the calculation; individual configured weights may be zero to exclude a source.
Unavailable measurements should be handled explicitly rather than automatically replaced with zero.
Calculation results should remain at full precision until the model's defined output-rounding step.
Where SOM is measured in indivisible units, a documented rounding policy is required. If the result is intended to represent a strict upper bound, rounding upward would not be appropriate.
The following assumptions should be retained with each calculation or its associated model version:
This enables later recalculation, source replacement, scenario analysis, and explanation of changes in reported SOM.
The SOM warehouse output, identified here as WAREHOUSE_SOM_VCT, should preserve sufficient information to reconstruct the calculation and explain the result.
The record should contain or reference:
| Output | Purpose |
|---|---|
| SAM demand units | Original demand constraint |
| Campaign forecast units | Commercial forecast constraint |
| Open capacity | Original operational capacity input |
| Utilization rate | Applied capacity adjustment |
| CBP proxy | Public source-derived estimate |
| ECN proxy | Public source-derived estimate |
| NES proxy | Public source-derived estimate |
| Source weights | Calculation governance |
| Weighted public proxy | Combined public capacity estimate |
| Operational capacity cap | Adjusted operational constraint |
| Public capacity cap | Adjusted public-data constraint |
| Final SOM units | Calculated obtainable market |
| Binding constraint | Identifies the limiting factor |
| Model and source references | Auditability and reproducibility |
These are logical output requirements rather than assertions of existing physical database column names.
A complete record permits the warehouse to support several analytical functions:
Constraint attribution: Identify why a particular market has a smaller SOM than its SAM.
Scenario modeling: Determine how SOM changes when campaign forecasts, operational capacity, source weights, or utilization assumptions change.
Geographic comparison: Compare markets under consistent normalization and source-weight policies.
Historical analysis: Track how obtainable-market estimates and binding constraints evolve across successive measurement periods.
Model validation: Compare forecast obtainable units against observed performance and revise assumptions where necessary.
Weighted Utilization Capacity provides a constrained estimate of market obtainability. It does not establish guaranteed sales, market share, revenue, or realized demand.
The final calculation is a minimum of four modeled upper bounds. Its accuracy depends on the reliability of the inputs and the assumptions used to construct them.
Several limitations are particularly important.
Public data is an indirect capacity measure. CBP, ECN, and NES measure aspects of economic activity rather than directly measuring the ability of the modeled business to capture transactions.
Weights do not eliminate source uncertainty. A weighted average can reduce dependence on one source, but it does not automatically correct systematic errors, incompatible source definitions, or correlated measurements.
Utilization is an assumption unless independently measured. A 70% utilization rate must be supported by operational evidence, empirical benchmarking, or a clearly identified planning assumption.
Constraints may be interdependent. Marketing performance, operating capacity, and market participation can influence one another. Modeling them as independently supplied caps simplifies the analysis.
The minimum is not a probability estimate. The result is the smallest modeled capacity or demand constraint, not a statistically derived confidence interval or a probability-adjusted sales forecast.
The public proxies require compatible economic interpretations. Averaging alternative estimates is different from adding complementary market components. The source transformations must support the selected calculation.
These limitations do not invalidate the method. They define the conditions under which its output can be interpreted responsibly.
The Weighted Utilization Capacity methodology calculates SOM by reconciling market demand, commercial expectations, internal operating capacity, and external economic capacity indicators.
It first constructs a weighted public-data capacity estimate from CBP, ECN, and NES:
It then adjusts internal and public capacity estimates using the utilization rate:
Finally, it selects the most restrictive applicable constraint:
The resulting SOM is accompanied by the identity of the binding constraint and the intermediate calculations used to produce it.
The principal value of this methodology is that it converts a broad estimate of serviceable demand into a constrained estimate of obtainable volume, while making the limiting assumptions explicit, reproducible, and available for subsequent analysis.
The SOM is therefore both a market-sizing output and a diagnostic measure of the capacity, demand, or commercial constraint most immediately limiting the modeled opportunity.