How to build a healthcare territory plan you can defend
A territory plan built on last year's revenue plus a growth factor cannot tell you whether the number is achievable, because it never establishes what the territory contains. Public healthcare data makes the denominator knowable for essentially every county in the country, which changes territory planning from negotiation into arithmetic.
8 min read
Define the decision unit before anything else
County is the workhorse grain in US healthcare data. Metro and CBSA work for market-level framing; ZCTA and tract work for site selection. Pick the grain your decision actually operates at, and accept that some datasets will only be available coarser — that widening should be visible in the output, not hidden.
Size the universe with registries
Count the addressable accounts: clinicians by taxonomy from NPPES, facilities from Care Compare, health centers from HRSA, businesses by NAICS from County Business Patterns. These are counts of the whole population, so 'how many are there' has an answer rather than an estimate.
Example pull from RepVector
Retained population and enrollment rows for a resolved county — the sizing step above.
ACS 5-year estimates for 1 county/counties in Boise, Ada County, Idaho
| County | FIPS | Total population | Median age | Median household income | Median home value |
|---|---|---|---|---|---|
| Ada County | 16001 | 508,052 | 38.3 | 88,907 | 476,000 |
First 1 of 1 retained rows. See the full field list for American Community Survey (5-year)
Attach demand and capacity
Census ACS and Population Estimates give population, age structure, income and insurance coverage. CMS Medicare enrollment gives the covered senior population. HCRIS gives hospital-level financials and utilization. CDC PLACES gives condition prevalence. This is where a large county with low relevant need gets correctly demoted below a smaller county with high need.
Rank, tier, and justify
Tiering is only credible if the rule is stated. Tier 1 = top decile on the ranking metric and payer coverage present; Tier 2 = high need, supply-constrained, unranked on volume; Tier 3 = everything else in the universe, retained so coverage can be measured. Every account keeps the figures that placed it in its tier.
Set quota against the denominator
With a universe of N accounts, an observed conversion rate and an average value, the plan becomes a calculation with named inputs. When the number is missed, you can identify which input was wrong instead of arguing about effort.
What to take away
- Pick the decision grain first; county is usually correct.
- Registries give a true denominator, not an estimate.
- Demote large-but-low-need markets using prevalence and payer data.
- State the tiering rule, and keep the figures that produced each tier.
Put this to work
RepVector assembles this for a named market in one query
Describe the market and the offering. The datasets above are selected, resolved to your geography, pulled, scored, and returned as a ranked list with every figure sourced.
Related guides
- Healthcare referral generation without buying a list
Five steps from a named market to a ranked, sourced referral list — using federal data that is already public.
- Medicare Advantage penetration by county — and what it changes about your sales approach
Payer mix decides who authorizes and who pays. CMS publishes it monthly, by county, for free.
- HCRIS hospital cost reports: reading a hospital's financial position before you pitch it
Every Medicare-certified hospital files an audited annual cost report. It is public, and it tells you who can buy.