A virtual care company is a healthcare provider with a technology company's investors, and the two frames collide on one line of the income statement: gross margin. Software investors expect 70 to 80 percent; a telehealth company that reports it has usually left the costs that scale with visits — the platform, the providers, the clinical support team — in operating expense, where they flatter the margin and wait for a diligence team to move them. The company that builds its margin honestly from the start negotiates its valuation on numbers that hold.
What belongs in cost of revenue, and why?
The test is whether the cost scales with visits. Provider compensation per visit — employed clinicians' salaries allocated to visits, or contractor fees per visit; clinical support — care coordinators, nurses, intake clinicians, and clinical quality staff whose hours rise with volume; the technology that delivers the visit — video platform fees, electronic health record licensing per provider or per patient, e-prescribing, scheduling, patient-messaging infrastructure, and the cloud and data costs behind them; payment processing and claims-clearinghouse fees; and the malpractice coverage that prices by provider volume; each rises with the next visit, and each belongs above the gross-margin line.
What stays in operating expense. Product and engineering building the platform rather than running it (with the judgment call documented), sales and marketing, general and administrative, and the corporate functions that do not move with visit volume; the boundary is written down as a policy, applied consistently, and defended to the auditors and the board once.
The restatement. A company reporting 71 percent gross margin with platform fees in technology expense, provider cost in "clinical operations" below the line, and care coordination in general and administrative will see a diligence team reallocate all three and arrive at 55 to 60 percent — and reprice the deal on the lower number; the company that reports 58 percent from the start is valued on 58 percent without the credibility loss.
How is contribution per visit built by payer and state?
Net revenue per visit by payer, by state, from remittances. Commercial plans reimburse telehealth behavioral and primary care visits at parity with in-person in many states and not in others, under policies that have shifted repeatedly; Medicare's telehealth rules have their own history and conditions; Medicaid varies by state program; employer and health-plan contracts pay per member per month or per engagement; self-pay at the company's price; the state dimension matters because parity, licensing, and credentialing differ by state, and a company serving fourteen states has fourteen payer environments.
Cost per visit by state. Provider cost at the compensation model; clinical support allocated per visit; technology per visit; and the state-specific costs — licensing and renewal for each provider in each state, payer credentialing by state, state-specific compliance and telehealth registration, and the malpractice differentials — allocated to the visits in that state; a state with 300 visits a month carrying four provider licenses and three payer credentialings has a per-visit state cost that a state with 3,000 visits does not.
Contribution per visit by payer and state is the map. Which payer-state combinations earn, which are marginal, and which lose — and the usual finding is that several small states are negative after their licensing and credentialing load, carried by two or three large ones; the decisions are state entry and exit, payer prioritization by state, and provider licensing concentrated where the volume is.
What are patient acquisition cost and visits per patient, and why do they decide valuation?
Patient acquisition cost by channel. All spend to acquire a patient who completed a first visit — paid search and social, content, affiliates, health-plan and employer referral programs, provider-referral development, and the intake staff time per converted patient — divided by first visits, by channel; the range across channels is typically three to five to one, with health-plan and employer referrals at the low end and paid social at the high.
Visits per patient and cohort retention. The number of completed visits a patient produces over an episode or a year, by acquisition channel, by payer, by clinical program; cohort retention curves — patients active at 30, 60, 90, 180 days — because a behavioral health patient who completes two visits and a chronic-care patient who completes twelve are different economics on the same acquisition cost.
Payback in visits. Acquisition cost divided by contribution per visit gives the visits to break even on the patient; against visits per patient by channel, the company sees which channels acquire patients who pay back and which acquire patients who leave before they do; a channel at US$310 per patient with a US$46 contribution per visit needs 6.7 visits to break even, and if its cohort averages 4.1, every patient from that channel is a loss.
The investor's unit economics table. Contribution per visit, patient acquisition cost by channel, visits per patient and retention by cohort, payback, and lifetime contribution per patient — by program and by payer type; it is the first page of any diligence request and the page the company should build before anyone asks.
How do provider models change the economics?
Contractor networks versus employed clinicians. Contractors paid per visit put utilization risk on the provider and cost per visit on a fixed basis; employed clinicians on salary put utilization risk on the company and lower cost per visit at high utilization; classification exposure exists where contractors are scheduled, assigned, and managed as employees; investors read the model for its scalability and its exposure, and the finance function models both at the company's actual utilization.
Provider utilization is the operating metric. Completed visits over available hours by provider, by state, with no-show rates — because telehealth no-shows run high in some programs and the provider's hour was paid either way.
What is burn, runway, and the bridge to breakeven?
The board's first question. Monthly net cash burn, cash on hand, months of runway, and the milestones — visits per month, contribution per visit, patient acquisition cost — at which the company reaches operating breakeven; the 13-week cash forecast for the near term and an 18-month runway model for the board; hiring and state expansion gated on runway milestones rather than on plan dates.
Why is this CFO work?
The cost-of-revenue policy is an accounting design with a valuation consequence; contribution by payer and state requires remittance and licensing data joined by state; the acquisition and cohort economics require marketing, scheduling, and billing data read by channel; the provider-model analysis is a compensation and classification design; and the investor table is the finance function's product. The clinical leadership owns the care model and the quality; the CFO owns the numbers investors price.
Worked example
A virtual behavioral health company in fourteen states, US$9.1 million revenue, venture-backed, preparing for a growth round.
| Item | As reported | Restated |
|---|---|---|
| Gross margin | 71% | 58% (platform fees, care coordination, and contractor provider cost moved to cost of revenue) |
| Contribution per visit (blended) | US$61 | US$46 |
| State (by visit volume) | Monthly visits | Net revenue / visit | State-specific cost / visit (licensing, credentialing, compliance) | Contribution / visit |
|---|---|---|---|---|
| Three largest states | 2,100–3,400 each | US$108–119 | US$2–4 | US$44–58 |
| Seven mid-size states | 400–900 each | US$96–112 | US$7–12 | US$18–36 |
| Four smallest states | 90–220 each | US$94–105 | US$29–41 | −US$6 to −US$17 |
| Channel | Patient acquisition cost | Visits per patient (12 mo) | Payback (visits) | Verdict |
|---|---|---|---|---|
| Health-plan referral | US$95 | 7.8 | 2.1 | Scale |
| Employer program | US$120 | 6.4 | 2.6 | Scale |
| Content and search | US$180 | 5.1 | 3.9 | Keep |
| Paid social | US$310 | 4.1 | 6.7 | Cut or restructure |
Illustrative figures for a hypothetical organization; not a client's data.
The company had been preparing investor materials on a 71 percent margin and a blended acquisition cost; the restated margin was 58 percent, four states were losing money on every visit after licensing and credentialing, and the largest acquisition channel by spend acquired patients who left before paying back. The plan: cost-of-revenue policy documented and the deck rebuilt on restated numbers; exit or pause in the four smallest states with providers' licenses allowed to lapse; acquisition spend shifted toward health-plan and employer channels; provider utilization program; and hiring gated on runway milestones. The round closed on the restated numbers — at a lower headline margin and with no restatement in diligence.