Operator reference
The 18 KPIs that run a telehealth business.
Every metric below comes with its definition, formula, and the levers that move it, because a KPI you cannot compute is a slide, not a system. One honesty note up front: rigorous public benchmarks barely exist for cash-pay telehealth, so this page defines how to measure, carries the few sourced reference curves that do exist, and refuses to invent the rest. Current as of September 2026.
Acquisition
| KPI | Definition | Formula | What moves it |
|---|---|---|---|
| Customer acquisition cost (CAC) | Fully loaded cost to acquire one paying patient, by channel and blended. | Acquisition spend ÷ new paying patients | Owned-audience share, creative quality, certification unlocking cheaper channels |
| Visitor-to-intake rate | Share of storefront visitors who begin the medical intake. | Intake starts ÷ unique visitors | Offer clarity, price transparency, state-gating accuracy, page speed |
| Intake completion rate | Share of started intakes that are submitted for clinical review. | Completed intakes ÷ intake starts | Form length, progress saving, question ordering, mobile quality |
Conversion & activation
| KPI | Definition | Formula | What moves it |
|---|---|---|---|
| Intake-to-visit rate | Share of completed intakes that reach a clinical decision. | Clinical decisions ÷ completed intakes | Clinician capacity, turnaround speed, follow-up on stalled cases |
| Activation rate (first fill) | Share of new patients whose first prescription is delivered, where clinically appropriate care leads there. Never a target for prescribing itself: clinical decisions belong to clinicians. | First fills delivered ÷ approved treatment plans | Pharmacy routing speed, payment success, shipping communication |
| Time to first fill | Median hours from intake completion to medication delivered. | Median(delivery timestamp − intake timestamp) | Visit turnaround, pharmacy network depth, fulfillment logistics |
Retention
| KPI | Definition | Formula | What moves it |
|---|---|---|---|
| Week-4 survival | Share of a starting cohort still active four weeks in. The category's published early cliff makes this the highest-leverage retention checkpoint. | Active at day 28 ÷ cohort starts | Onboarding expectations, first-fill reliability, early care-team touchpoints |
| Persistence at 3 / 6 / 12 months | Share of a cohort continuously on program at fixed checkpoints, measured with a defined allowable gap (the published studies use 60 days). | Active at checkpoint ÷ cohort starts | Refill continuity, price predictability, adjacent care lines, pause-and-return paths |
| Refill continuity rate | Share of due refills delivered before the prior fill runs out. | On-time refills ÷ refills due | Supply reliability, renewal automation, proactive outreach on exceptions |
| Involuntary churn share | Share of churn caused by payment failure rather than choice. | Payment-failure cancels ÷ total cancels | Dunning quality, card-updater coverage, retry timing |
| Reactivation rate | Share of lapsed patients who return within a window. The published GLP-1 research shows discontinuation is often an interruption, not an ending. | Reactivated ÷ lapsed in window | Respectful off-ramps, winback flows, pause options |
Unit economics
| KPI | Definition | Formula | What moves it |
|---|---|---|---|
| Contribution per patient per month | What one active patient adds monthly after the direct cost stack. The number every other metric exists to move. | Monthly price − (platform rate or medication + visits + processing + support) | Pricing architecture, medication economics, operating efficiency |
| Cohort value at checkpoints | Cumulative contribution per starting patient at months 3, 6, and 12: each month's contribution weighted by the share of the cohort still active that month, then summed. Weighting only the endpoint discards the revenue patients paid before they churned. Use checkpoints, never blended-average LTV: tenure distributions are heavily skewed. | Σ over months t ≤ checkpoint: contribution × share active in month t | Everything in the retention group, times price |
| CAC payback period | Months of contribution needed to repay acquisition cost. | CAC ÷ contribution per patient per month | Both terms; owned audiences shrink the numerator structurally |
| Medication share of stack | Medication cost as a share of the total per-patient cost stack. In recurring-Rx programs it is usually the largest line, which is why per-fill economics belong in writing. | Medication cost ÷ total monthly cost stack | Supply model (included vs pass-through), product mix, market price floors |
Operations & trust
| KPI | Definition | Formula | What moves it |
|---|---|---|---|
| Time to visit decision | Median hours from completed intake to clinical decision, including nights and weekends. | Median(decision timestamp − intake timestamp) | Clinician capacity and scheduling, state routing, escalation paths |
| Support tickets per 100 patients | Monthly support volume normalized by active patients; the earliest leading indicator of operational problems. | Tickets ÷ (active patients ÷ 100) | Shipping reliability, billing clarity, proactive status communication |
| Dispute (chargeback) rate | Card disputes as a share of transactions. Card networks monitor this against thresholds, and sustained breaches endanger the merchant account. | Disputes ÷ transactions | Honest descriptors, findable cancellation, refund policy, delivery reliability |
The benchmarks that actually have sources.
Retention reference curves, from insured GLP-1 populations rather than cash-pay cohorts (use the shape, not the level): 58% of weight-loss GLP-1 patients historically discontinued before completing 12 weeks, with more than 30% gone in the first month (Blue Health Intelligence, 2024); roughly a third of 2021-2022 starters remained on therapy at one year (Prime Therapeutics); and one-year persistence roughly doubled to about 63% for early-2024 starters as supply and prices normalized (JMCP, March 2026). Our retention benchmarks guide carries every citation.
Price anchors for the revenue line: manufacturer cash channels and the major programs publish their prices, and our sourced pricing map tracks them. Everything else in this category (activation rates, CAC, ticket volumes) is unpublished at any rigor worth citing, which is precisely why the definitions above matter: measure your own cohorts with study-grade definitions and your benchmarks become real while your competitors' stay vibes.
Telehealth KPIs: FAQ
What KPIs should a telehealth business track?
Eighteen metrics across five groups cover the machine: acquisition (CAC, visitor-to-intake, intake completion), conversion (intake-to-visit, activation, time to first fill), retention (week-4 survival, persistence checkpoints, refill continuity, involuntary churn, reactivation), unit economics (contribution per patient per month, cohort value at checkpoints, CAC payback, medication share of stack), and operations (time to decision, tickets per 100 patients, dispute rate). Contribution per patient per month is the one the rest exist to move.
What are typical telehealth retention benchmarks?
Rigorous public benchmarks barely exist for cash-pay DTC telehealth; the sourced reference curves come from insured GLP-1 populations: 58% of weight-loss patients historically discontinued before completing 12 weeks (Blue Health Intelligence, 2024), roughly a third of shortage-era starters remained at one year, and recent cohorts persist near 63% at one year (JMCP, March 2026). Use them as reference shapes, measure your own cohorts with the same definitions, and distrust any vendor quoting category benchmarks without a source.
Why measure persistence at checkpoints instead of average LTV?
Because tenure is heavily skewed: a large early cliff plus a durable tail means a blended average describes almost nobody, flatters models, and hides the week-4 problem where the cheapest retention gains live. Survival at week 4 and months 3, 6, and 12, with a defined allowable gap, is how the published research measures and how honest operators should too.
What is a good CAC for telehealth?
The honest answer is relative: a good CAC is one your contribution per patient per month repays inside your risk tolerance (the payback-period KPI), which is why owned-audience brands operate at CACs paid-acquisition brands cannot. Model payback per cohort against honest tenure checkpoints rather than chasing a category number nobody can source.
Which KPI matters most for a new telehealth program?
Week-4 survival, once patients exist: the category's published early cliff means the first month decides more of your economics than any later intervention, and it is fixed by onboarding expectations, first-fill reliability, and early touchpoints rather than spend. Before patients exist, intake completion rate is the cheapest funnel fix available.
Definitions and formulas are operational guidance, not financial advice; the cited reference curves come from insured populations and are reference shapes, not targets or promises. Clinical decisions belong to licensed clinicians and are never metrics to optimize.
Want your program's KPIs knowable before launch? Get the partner overview.
One email, no sequence. A human follows up with your program scoped.