TL;DR
Count acquisition using a defined relationship-start event, then report evaluations and subsequent encounters separately. A PT series can generate repeat bookings without generating new people. Keep new-to-clinic relationships, returning patients with new episodes, reevaluations and follow-ups distinguishable. Use clinical record classifications supplied by the clinic; a marketer should not decide what kind of care occurred.
Begin with the reporting question
An owner sees a dashboard labeled “new patient appointments,” but the scheduler includes evaluation appointments, continuing visits and rescheduled visits. The total cannot answer how much new demand arrived until the underlying records are classified. Do not solve the ambiguity by choosing whichever number makes the campaign look strongest.
Ask whether the report is meant to describe new relationships, evaluation access or workload. Those are different business questions. New relationships concern people who newly enter the clinic. Evaluation access concerns whether an evaluation was requested, scheduled and attended. Workload concerns encounters using staff and schedule capacity, including ongoing care.
This article proposes a management dictionary. It is not a clinical coding guide, a billing rule or a recommended care frequency. The practice's authorized clinical and billing staff must determine the actual encounter classifications from appropriate records.
Adopt an evaluation-cohort dictionary
Choose a review period and a cohort anchor before calculating anything. For acquisition, a useful proposed anchor is the attended initial evaluation for a new-to-clinic relationship. Record inquiries and bookings as earlier states, while leaving incomplete journeys open. Decide explicitly how the clinic handles people whose historical relationship cannot be established.
| Reporting term | Proposed operational definition | Exclusion or caution |
|---|---|---|
| New-to-clinic relationship | Person with an attended initial evaluation and no prior clinic relationship under the approved history rule | An incomplete history is unknown, not automatically new |
| Returning person, new episode | Existing relationship entering a separately documented episode | Report as returning episode demand, not newly acquired person |
| Evaluation request | Received request for an evaluation | May never become a scheduled or attended appointment |
| Scheduled evaluation | Evaluation appointment confirmed in the clinic system | Reschedules should not create extra acquisition credit |
| Attended evaluation | Attendance confirmed for the evaluation event | Does not establish completed care or collected revenue |
| Reevaluation | Encounter classified by qualified clinic staff as a reevaluation | Do not reclassify it as a new relationship for marketing |
| Follow-up encounter | Subsequent encounter classified by the clinic within the relevant care episode | Workload and care activity, not an additional acquired person |
Add the data owner, system field, history coverage and unresolved-state rule to each row. If the scheduler uses “evaluation” for several appointment types, request a mapping from the people who understand that configuration. Preserve the original field so the mapping can be audited later.
Keep person, episode and encounter records at their proper level
Ask the clinic's reporting owner to document the links among the records it already uses. A person-level acquisition question, an episode-level service-demand question and an encounter-level workload question should not inherit the same counting rule.
| Record level | Link the clinic must establish | Reporting decision |
|---|---|---|
| Person | Approved identity and relationship history | Has this person already met the clinic's relationship-start definition? |
| Episode | Clinic-assigned episode linked to that person | Is this a separately documented episode for a returning person or continuing activity in an existing episode? |
| Encounter and its scheduling history | Clinic-classified care encounter linked to its episode, plus the associated appointment record | Was care delivered as an initial evaluation, reevaluation or follow-up, or does the appointment remain cancelled, rescheduled or unconfirmed? |
A returning person's separately documented episode can increase episode demand without increasing newly acquired people. A reevaluation within an existing episode can change encounter workload without starting either a new episode or a new relationship. Have qualified staff establish those links; do not manufacture episode boundaries from a gap between visits, a changed campaign tag or a different appointment label.
Use an internal mapping record: approved person key ___; approved episode key ___; encounter key and state ___; relationship-start event ___; unresolved linkage ___. These are field prompts for the clinic, not a request to export identifying keys to an agency. If episode linkage is unavailable, preserve that limit instead of counting every evaluation label as fresh acquisition.
Keep the relationship cohort stable as encounters accumulate
Within approved clinic systems, link the qualifying relationship event to later encounters without adding new acquisition credit for each booking. A reschedule changes the appointment history. It should not silently produce another relationship. A cancelled evaluation remains a cancellation unless a later attended event meets the cohort definition.
Use a separate encounter report to show evaluations and follow-ups delivered during the reporting period. That report can include care associated with relationships that began earlier. Its workload total should not be compared as though it were the same population as the current acquisition cohort.
Google Ads lets advertisers choose how conversion actions are counted after an interaction. The platform setting is distinct from the clinic's person-and-episode definition. Google Ads counting options. Limiting repeated ad conversions does not by itself deduplicate a person across the clinic's relationship history. Ask the analyst to document what the configuration actually observes.
Reconcile the ambiguous records before presenting a rate
Have the clinic classify unresolved records as unknown, pending or excluded with a reason. Examples include missing prior history, an evaluation booked outside the reporting period, an encounter with an unclear appointment type and a duplicate booking record. Marketing should receive only the findings approved for that recipient.
If calculating an attendance proportion, divide attended evaluations from the defined scheduled cohort by that cohort's scheduled evaluations using an agreed observation cutoff. State how cancellations and reschedules are handled. If no eligible scheduled evaluations exist, show unavailable rather than a percentage. This is a definition-based calculation, not a performance benchmark.
Do not divide this month's advertising spend by this month's follow-up visits and label the result cost per new patient. The denominator and the buying question do not match. The general measurement dictionary explains event boundaries; this PT dictionary adds the relationship, episode and encounter distinctions needed to keep a visit series from inflating acquisition.
Ask the vendor to deliver a reconciled reporting note
Require the report to name the cohort anchor, history rule, cutoff, exclusions and unresolved records. Show newly attended relationships, returning episodes and encounter workload as separate series. Record any tracking change that breaks historical comparison.
Judge reporting quality by whether an owner can trace each label to the clinic's approved definition. A software screenshot is not sufficient when the field meanings remain disputed. If the agency only measures requests, retain that useful narrower measure and label the later outcome gap.
Frequently asked questions
Is a returning patient with a new concern worthless to marketing?
No. Returning episode demand can matter to the clinic. Give it an accurate separate label, and leave assessment and care decisions with qualified staff. Its value does not require calling the person newly acquired.
Should we report follow-ups at all?
Yes, when the report addresses delivered workload, schedule demand or financial context. Keep them separate from new relationships and avoid setting visit quantity as a marketing success target.
What if our software cannot connect episodes reliably?
Use the definitions that can be supported and mark the missing connection. Have the clinic resolve its internal record process. Do not upload patient histories to an agency simply to complete a dashboard.