Operations
Part of Patient Acquisition: The Version That Survives Contact With Reality
Patient Acquisition Benchmarks: Build Your Own Baseline
How to build patient acquisition benchmarks from your own records: eligible inquiries, completed entries, cost per entry, and source fit.
What to take away
- Build an internal baseline of eligible inquiry, completed care entry, cost and capacity before you adopt an outside target.
- Recalculate every borrowed rate from its cited source with your own eligibility rule and denominator.
- The CMS Medicare Current Beneficiary Survey covers a defined Medicare population, not your service line.
- CDC PLACES estimates describe areas and prove no demand for a named provider.
- Record population, period, method and maturity window with the number, or nobody can reproduce it.
Build the internal baseline first
Patient acquisition benchmarks answer a question you have already framed. Frame it in your own records first.
| Measure | Definition | Guardrail |
|---|---|---|
| Eligible contact rate | Eligible inquiries divided by reviewed inquiries | Wrong-service volume |
| Completed care-entry rate | Completed first visits divided by eligible inquiries | Wait time and no-shows |
| Full cost per entry | All acquisition cost divided by completed entries | Staff load and margin |
| Correction closure | Verified repairs divided by material defects | Severe open issues |
Track these measures by service only when the sample and workflow make the split useful. Keep late appointments inside a maturity window. Annotate any mid-period change to intake rules. Never compare a selectively captured campaign list against an all-contact denominator.
Sending and calling restrictions also decide whom you may contact, as set out in the TCPA and CMS outreach rules — the Telephone Consumer Protection Act and the Centers for Medicare & Medicaid Services. Those rules shape the pool any benchmark can describe.
Example: recalculating a borrowed rate
The figures below are illustrative. Repeat the arithmetic with a real source.
Suppose a campaign summary reports 30 completed first visits per 100 leads. Suppose your own quarter shows 400 inquiries and 120 completed first visits, also 30 percent.
The match is false. Your eligibility rule removes 90 inquiries for services you do not offer and 40 duplicate records. Eligible inquiries fall to 270, so the rate rises to 44 percent.
A maturity window then removes 30 eligible inquiries that are 9 days old and still open. Resolved eligible inquiries total 240, and the rate reaches 50 percent.
Acquisition spend of $30,000 for the quarter divides across 120 entries. That gives $250 per completed care entry, with staff load beside it.
The borrowed 30 percent was not wrong. It was measured on a different denominator. Set the plan target inside the 44 to 50 percent band and name the definition you used.
Before any figure enters a plan, apply the population, period, denominator and eligibility checks below. For additional context, see medical practice marketing benchmarks.
Check whether a source fits
The CMS Medicare Current Beneficiary Survey public-use file reports access, health status, satisfaction, demographics and coverage for beneficiaries in its stated universe. It is updated annually. That makes it the right authority for population context, not for a conversion target.
Check the survey year, universe, variables, collection mode and weights. Check missingness and uncertainty. Use it to frame questions or to supply context.
CDC PLACES methodology builds small-area estimates from BRFSS, Census and American Community Survey data. It publishes the geographic levels and validation for each measure.
It is the right authority for area context because it states how each estimate was modelled. These are population estimates, not individual records. They prove no demand for a particular provider.
State the release, geography, measure and limitations. Never infer a person's health status from a tract or ZIP estimate.
Make the comparison reproducible
Write the record in this order so a reviewer can repeat the calculation. Use the data and privacy risk checklist for a separate privacy review.
- Name the population, the service, the period and the inclusion rule once, at the top.
- Recalculate the borrowed figure from the cited source and log any mismatch against your baseline.
- Record the currency, the maturity window and the missing records behind both numbers.
- Keep market context separate from the internal target.
- Set the next review date and name the change that would force an earlier check.
A rate without its population, period and denominator is not evidence.
The U.S. Government Accountability Office (GAO) evaluation design guide ties an evaluation question to evidence needs and design choices, which makes it the right authority for separating a comparison from a causal claim. Experimental design selection guidance from the National Institute of Standards and Technology (NIST) starts from the objective and the practical constraints in the same way.
Set the target and keep it honest
Set the review date when you set the target. Name the trigger that would force an earlier look, such as a shift in referral volume or payer mix.
The denominator problem appears when every inquiry or lead is counted as a patient, even if the person is ineligible or never completes care. Avoid the mistake described in counting every lead as a patient.
Common questions
What makes a benchmark valid or comparable?
A valid benchmark names its population, period, eligibility rule, numerator and denominator. It is comparable only when those definitions match yours closely enough for the decision in front of you.
Is there a good patient acquisition rate?
No universal figure exists. Build a stable internal baseline first, then treat an outside rate as context rather than as a target.
Can two locations share one benchmark?
Only when services, populations, workflows, capacity and data quality are comparable. The combined result must still answer the question you are asking.
What should I document?
Population, period, inclusion rule, calculation, currency, maturity window and the source citation. Add rejected options, because the original constraint can change.
