Serviceable addressable market

Editorial approval scope: The team checked current source fidelity, scope boundaries, dates, arithmetic, reader usefulness, practical workflow, and general-information limitations.

SAM sits between TAM and SOM

Total addressable market (TAM) describes the full defined opportunity under a broad set of assumptions. SAM narrows that universe to the segment the particular model could address. Serviceable obtainable market (SOM) estimates the portion the practice could realistically capture within a time horizon.

The relationship should usually be:

SOM ≤ SAM ≤ TAM

Each term needs the same base unit and date. A SAM measured in people cannot be compared directly with a TAM measured in annual revenue.

Define the service before filtering

Name the service, setting, modality, population, geography, funding route, and period. “ABA in Texas” is too broad. “Weekday center-based assessment for school-age Medicaid members in three counties during 2027” creates a testable boundary.

Specify whether the unit is people, referrals, service starts, authorized hours, delivered hours, or revenue. People and hours answer different capacity questions. Revenue also depends on service mix, allowed amounts, utilization, and collections.

Apply filters in an auditable order

Useful filters may include:

• actual service and clinical scope
• entity, professional, facility, and modality authority
• locations within lawful and feasible reach
• payer products, contracts, enrollment, and self-pay routes
• published population criteria tied to the service model
• settings and schedules the model supports
• accessible communication, transportation, technology, and site requirements
• exclusions supported by current law, clinical scope, and operating design

Keep the source, date, definition, owner, and count remaining after each filter. Overlapping filters require person-level or otherwise deduplicated data; multiplying independent percentages can create false precision.

Capacity belongs in SOM

Current clinician headcount, recruiting pace, supervision bandwidth, rooms, travel hours, and working capital determine what the practice can obtain, not what the model could address in principle. Use those constraints later when building SOM.

A hard legal, payer, facility, or clinical-scope boundary belongs in SAM. A temporary vacancy or six-month marketing budget belongs in SOM. Some constraints need judgment, so document the rule and test both classifications when material.

Public data is only one input

The SBA market-research guide recommends examining demand, market size, location, saturation, and pricing. Census data can supply population and geographic context with stated releases and margins of error where available.

Population or diagnosis estimates do not equal demand, eligibility, service choice, payer coverage, or hours. Combine public context with current boards, payers, provider availability, community input, accessible research, and first-party inquiry data.

A fictional SAM calculation

Blue Elm ABA, a fictional organization, begins with a TAM estimate of 2,400 annual service-start opportunities for a broadly defined offering. Its chosen model and authority filters reduce the count to 1,500 opportunities in approved counties. Published service criteria reduce it to 900. Verified payer or self-pay routes reduce it to 620. Setting and modality fit reduce it to 480.

The SAM is 480 annual service-start opportunities under that model and evidence version. The practice keeps each subtotal: 2,400, 1,500, 900, 620, and 480. It does not multiply 480 by a guessed fee and call the result guaranteed revenue.

Current staff can support only 72 starts during the year. That 72 is a capacity input for SOM rather than a reason to shrink SAM.

Test sensitivity and uncertainty

Use base, downside, and severe cases for uncertain filters. Show ranges for payer access, family choice, schedule fit, provider availability, and estimated service mix. Avoid one decimal place when the underlying inputs are rough counts.

Run a reconciliation check: every excluded segment should have a reason, source, and count. Guard against double exclusion. Preserve people with access needs in the relevant market while modeling the resources needed to serve them.

Use SAM as a planning boundary

SAM can guide facility search, contracting priorities, community research, workforce planning, and product focus. It should never replace case-level clinical decisions or become a promise to investors, payers, referral partners, or families.

Refresh the model after legal, payer, service, site, modality, population, or data changes. Record the owner, approval date, version, and next review trigger.

Before approving a planning decision, have a second reviewer rebuild the filter sequence from the cited sources and reconcile every subtotal. Record any judgment call and its sensitivity effect, then verify the final approved model version.