Business model and economics

Proven rule. Every body claim traces to a source. Inferred or single-sourced-shaky claims live in section 4.

Kicker. MEDvidi publishes enough of its own numbers to reconstruct the whole P&L shape, and the picture is this: roughly 120,000 paid video visits a year at $195 first / $159 after, sold to roughly 35,000 patients a year, which is about 3.4 visits per patient per year and about $20 million of visit revenue. The clinician is a 1099 contractor paid $30–$60 per visit, so the gross margin on a visit is around 73–75%, and paid search is effectively switched off — 1,400 paid sessions in July 2026 against 153,850 organic ones. That combination, cheap acquisition plus a fat per-visit margin, is what lets a $20M clinic carry a 55-plus person product and engineering org. The fragility is not cost and it is not clinical capacity, which is running at roughly a third of the contracted panel's minimum hours. The fragility is that 3.4 — the business collects a monthly-medication cadence's worth of clinical obligation and only about a quarter of the revenue that cadence could produce. MEDvidi already knows this: the patient portal shipped in August 2026 contains a fully built, Stripe-backed membership called MEDvidi Prime priced at $159 recurring that "covers your monthly visits", complete with a cancellation-save flow, sitting behind a feature flag called Membership_Launch_Test_1 and a targeted offer that fires at the exact moment a patient is about to pay a $50 late-reschedule fee. The public FAQ still says, in the present tense, "we have no subscription model". That gap between the marketing site and the shipped code is the single most important thing in this Area, because it means the repricing decision is already made and the VP of Product is being hired into its execution — a move that, if it lands, takes revenue per retained patient from about $581 to about $1,908.

Source legend

1. What we found

1.1 The price architecture, from primary sources

MEDvidi sells one thing: a video appointment with a prescriber. There is no medication margin, no lab margin, no device, no add-on. Its own FAQ is explicit — "Will MEDvidi charge me extra for online prescriptions? No, we don't ask for any extra fees. You only pay for an appointment" [WEB]. Everything else in this Area is a consequence of that sentence.

The published prices, as of 2026-08-18:

Four details in that table matter more than the headline numbers.

The follow-up is priced at 82% of the initial. $159 / $195 = 0.815. In a business whose clinical reality is recurring visits, that is an unusual choice — see section 1.9 for what competitors do with the recurring leg.

The weight-loss bundle is the only volume discount in the business. The page says "$249 … Save $105 with our package" [WEB], and the arithmetic checks: $195 + $159 = $354, minus $249 = $105, a 29.7% discount for pre-committing to two visits. The portal confirms this is a real first-class plan type, not marketing copy — the service-plan enum in the booking chunk reads INITIAL, FOLLOW_UP, MEMBERSHIP, WEIGHT_LOSS_BUNDLE and there is a plan code initial_and_follow_up_weight_loss_bundle [PORTAL].

Discounting is a stated policy, not an omission. MEDvidi maintains a whole page whose job is to tell you no coupon exists: "At this time, MEDvidi does not offer promo codes or discounts to maintain transparent and fair pricing for all patients," and, pointedly, "Third-party websites may publish promo codes without MEDvidi's approval … the codes you may find elsewhere are not valid" [WEB]. That page is a defensive SEO asset aimed at coupon-affiliate sites that would otherwise intercept branded search. The Terms and Conditions reinforce it from the other side, prohibiting patients from distributing referral links "on sites where you are not the primary owner, including but not limited to Wikipedia or coupon websites" and from "promoting your referral link via search engine marketing, including but not limited to AdWords, Yahoo, or Bing" [LEGAL]. So: zero discount leakage on the marketing site, and a referral programme with "referral credits" that exists in the contract but is not advertised.

Payment is taken before the service is delivered, and in two bites. "The payment might be divided into two parts. First part charges at the moment of the appointment booking. The rest charges just before the appointment starts" [LEGAL] Payment Terms. Combined with the fee regime, MEDvidi has built a fairly hard-edged collections posture for a healthcare brand: "Any outstanding No-Show or Late Cancellation/Rescheduling fees must be paid before a new appointment can be scheduled" [LEGAL]. For a patient on a Schedule II stimulant who needs a new prescription every 30 days, that clause has real force.

1.2 The fee regime is a product surface, not fine print

The Refunds & Fees Policy grants a full refund in six enumerated situations — timely cancellation at 8+ hours, provider cancellation, provider-side technical failure, a patient wrongly booked with a provider not licensed in their state, provider no-show or arrival 15+ minutes late, and no suitable slots available after payment [LEGAL]. It refuses a refund once the appointment has begun, charges $50 for late cancel/reschedule and $100 for a no-show, processes approved refunds in 2–3 business days, and caps the request window at 15 days from payment [LEGAL].

The portal shows this is instrumented, not aspirational. The payment-type enum is INITIAL, FOLLOW_UP, MEMBERSHIP, NO_SHOW_FEE, LATE_RESCHEDULE_FEE, REQUEST_APPOINTMENT, the appointment-fee taxonomy is no_show_lead, no_show_lead_reschedule, no_show_doctor, no_show_both, late_reschedule, and the cancellation-reason enum includes PATIENT_NO_SHOW, UNEXPECTED_CIRCUMSTANCES and LATE_CANCELED [PORTAL]. There is a refund reason code verification_issue_refund alongside lead_verification [PORTAL] — which lines up with the FAQ's disclosure that after you pay, "the Care Team will process your Intake form and complete a PDMP (prescription drug monitoring program) verification" before the appointment is confirmed [WEB]. Money is collected, then eligibility is checked, then some of that money is refunded under a named code. That is a measurable leak with a name in the schema.

There is also an isBlacklisted flag on the patient object [PORTAL].

1.3 MEDvidi Prime: a $159 recurring membership is already shipped

This is the load-bearing finding of the Area, and it is entirely from public unauthenticated static assets.

The Personal Portal bundle served from join.medvidi.com on 2026-08-18 (bundle last-modified 2026-08-14) contains a complete membership subsystem [PORTAL]:

• Routes: /membership, /membership/cards, /membership/history, /membership/manage.
• Brand: the header logo swaps to assets/images/medvidi-prime-logo.svg when hasMembership() is true — a gold (#E0A800) badge mark that resolves with HTTP 200. The product is called MEDvidi Prime; menu items read "Cancel MEDvidi Prime" and "Reactivate MEDvidi Prime".
• Price: the billing-history subtitle renders, verbatim, `Next charge is on ${v}, $159` where v is formatted from subscription.dateTo. When cancelled it renders "Auto-renewal is off".
• What it covers: the dunning notice reads "Your bank indicated insufficient funds. Add a different card to keep your monthly visits covered." The cancellation screen lists what the patient loses: "You'll have to pay for the visits separately" and "You'll lose access to free late rescheduling ($50 per visit)".
• Billing rails: stripeService, paymentApiService.getSubscriptionBilling(), addCard()/updateDefaultCard(), a subscription object with isCancelled, dateTo, hasErrorInLastPayment, defaultCard, plus card-expired and card-soon-to-expire states. This is a Stripe subscription with a real dunning surface.
• Churn machinery: 28 distinct PostHog events under PP — Membership — …, including "Membership Cancelled", "Stayed With Membership", "Moved Next With Membership Cancellation", "Membership Reactivated", "Went To Membership Reactivation" (reachable from the dashboard) and a cancellation-reason picker with a free-text comment. Someone has built a save flow and a win-back flow.
• Feature flags: Membership_Launch_Test_1, Targeted_release_late_reshedule_membership_offer, HSA_FSA_test1, patient_session_recording_split, patient_anonymous_signup_session_recording, is_pharmacy_text_search_enabled. Flags are resolved through PostHog per leadId.

The offer targeting is the most interesting engineering in the bundle. Verbatim logic: the membership offer is suppressed unless all of the following hold — the current screen is the appointment-fee view, the fee type to pay is LATE_RESCHEDULE, the user has no existing membership, the user is not blacklisted, the service isMentalHealthGroup, and the service plan type is FOLLOW_UP; only then is the PostHog flag Targeted_release_late_reshedule_membership_offer consulted, and the offer shows only when it returns the literal string "test" [PORTAL].

Read in plain language: MEDvidi is A/B-testing a $159/month membership pitched to a mental-health patient at the precise instant they are about to be charged $50 for rescheduling a follow-up late. That is textbook behavioural design — the offer arrives at peak willingness to pay to make a penalty disappear — and the job description happens to require "experience with behavioural design — building product experiences that drive user behaviour" [JD].

Meanwhile the public FAQ says: "we have no subscription model … you may book one appointment at a time" [WEB]. Both statements are current as of 2026-08-18. The site describes the business as it was; the code describes the business as it is becoming.

1.4 The recurrence mechanic, and why it is the only number that matters

Controlled-substance treatment forces visits. MEDvidi states the mechanism itself: "Health & Safety Code Section 11200 (c) prohibits the refilling of a Schedule II controlled substance. A new prescription should be issued every time" [WEB]. Urgent refills without a visit are possible for "established patients … (excluding controlled substances)" [WEB]. The pharmacy FAQ pins the clock: "refills can be requested only 25 days after the previous prescription's fill date, and pharmacies usually dispense medication 30 days after the last fill date" [WEB]. The portal copy for the refill product is blunter still: "Plan in advance: No forgetting to schedule that follow-up next month!" [WEB].

So the clinical clock is 30 days. The commercial clock is not. MEDvidi's own About Us page states "35,000+ patients & 120,000 online visits every year" [WEB]:

Visits per patient per year
  120,000 visits ÷ 35,000 patients = 3.43 visits per patient per year
                                                              [WEB], company's own figures

Three point four, against a clinical cycle of twelve. The gap is the entire product opportunity, and the entire risk.

Annual revenue per patient, at the realised cadence:

Realised annual value per patient  [REASON]
  Assume each counted patient has 1 initial + (3.43 - 1) = 2.43 follow-ups in the year.
  195 + (2.43 x 159) = 195 + 386.37 = $581.37
  Assumption: patient-year composition is one initial plus the remainder as follow-ups.
  This overstates initials for patients carried over from the prior year (see sensitivity below).

Sensitivity, so the number can be argued rather than asserted:

Realised versus theoretical  [REASON]
  581 / 1,944 = 29.9%
  MEDvidi realises about 30% of the revenue a strictly monthly stimulant patient would generate.

Two honest explanations compete, and they have different product implications. Either patients churn part-way through the year — in which case the fix is retention — or the median stable patient receives sequential/post-dated prescriptions covering roughly 90 days, in which case the $672 row is the real ceiling and the realised $581 is close to structural. MEDvidi's own FAQ supports the second reading at least partially: post-dated prescriptions "may be considered for long-term, stable patients who have been regularly monitored" [WEB]. The likely truth is a mix, and separating the two is a data question a new VP of Product can answer in a week with the portal's own event stream — which is exactly why the answer is not in this document (section 4).

Either way, Prime reprices the gap rather than closing it:

Prime versus pay-per-visit, per retained patient-year  [REASON]
  Prime:            12 x 159            = $1,908
  Pay-per-visit:    195 + 2.43 x 159    = $581
  Ratio:            1,908 / 581         = 3.28x

And Prime is accretive in every consumption scenario, which is the part that should make a product leader both excited and careful:

Prime contribution, by how much care the member actually uses  [REASON]
  Clinician cost per follow-up, blended 50/50 MD/PMHNP: (40 + 30) / 2 = $35   [JOBS]
  Card processing per charge at 2.9% + $0.30:            159 x 0.029 + 0.30 = $4.91
  Annual processing on 12 charges:                       12 x 4.91 = $58.92

  Member uses 12 visits:  1,908 - (12 x 35) - 58.92 = 1,908 - 420 - 58.92 = $1,429  (74.9%)
  Member uses  6 visits:  1,908 - (6  x 35) - 58.92 = 1,908 - 210 - 58.92 = $1,639  (85.9%)
  Member uses  3 visits:  1,908 - (3  x 35) - 58.92 = 1,908 - 105 - 58.92 = $1,744  (91.4%)

  Assumptions: MD/PMHNP split 50/50 per the company's own "50% of healthcare providers are MD"
  [ARC]; Stripe's published standard US card-not-present rate as the processing proxy [REASON];
  no allowance for chargebacks, refunds or partial months.

Margin improves the less care a member consumes. That is the gym-membership dynamic, imported into psychiatry. It is a legitimate business — patients genuinely value never thinking about the $50 fee again — but it puts the product organisation's incentives and the clinic's incentives on opposite sides of a line, and the person who owns PM, Design, Data and Engineering under one umbrella [JD] is the person who will have to hold that line explicitly.

How much conversion it takes to matter:

Break-even conversion for a $5M revenue lift  [REASON]
  Incremental revenue per converted patient: 1,908 - 581 = $1,327
  Members needed:                            5,000,000 / 1,327 = 3,768
  As a share of the 35,000-patient base:     3,768 / 35,000 = 10.8%

Alternative lever, same money, no membership:
  One additional follow-up per patient per year: 35,000 x 159 = $5,565,000 revenue
  At 74.9% contribution:                        $4,168,000

Those two lines are the same size. A VP of Product therefore has two routes to the same $5M and should be able to say which one they would run first, and why.

1.5 Scale, triangulated four ways

Method 1 — visits times price. The strongest, because both inputs are the company's own and the price spread between the two SKUs is only $36, which makes the answer nearly insensitive to the initial/follow-up mix.

Method 1: visit revenue for the 120,000-visit year  [REASON] on [WEB] inputs
  Upper mix (every counted patient has an initial):
     35,000 x 195 = $6,825,000
     85,000 x 159 = $13,515,000
     total        = $20,340,000

  Lower mix (25,000 initials; 10,000 patients carried over from the prior year):
     25,000 x 195 = $4,875,000
     95,000 x 159 = $15,105,000
     total        = $19,980,000

  Range: $19.98M - $20.34M.  Blended revenue per visit: $166.50 - $169.50.

Add the fee line:

No-show and late-fee revenue  [REASON]
  Let f = share of booked slots that end as no-show or late cancel/reschedule.
  Booked slots = 120,000 / (1 - f).
  At f = 5%:  booked = 126,316; fee events = 6,316
  Fee mix 50/50 between $100 no-show and $50 late = $75 average
  Gross fees = 6,316 x 75 = $473,700
  Collected at 70% (MEDvidi gates re-booking until fees clear [LEGAL]) = $331,590

  Assumption: f is unobservable from outside. MEDvidi's own recruiting pitch claims "the lowest
  no-show rate in the market" [JOBS], so f is plausibly 3-8%. Range of fee revenue: $0.15M-$0.7M.

So the 120,000-visit year is roughly $20.2M–$21.0M of collected revenue. Which year that is matters: 120K+ is the company's 2025 figure [WEB]. Applying its own 2024→2025 growth:

2026 run-rate  [REASON]
  2024: 90,000 visits -> 2025: 120,000 visits = +33.3%  [WEB], About Us timeline
  If sustained: 2026 = 120,000 x 1.333 = 160,000 visits
  At $166.50-$169.50 blended: $26.6M - $27.1M, plus fees
  Range for a 2026 run-rate: $24M - $28M.
  Assumption: growth continues at the 2025 rate. Fragile - see section 4.

Method 2 — prescriber capacity. This one is not a revenue estimate; it is a constraint test, and it comes back negative.

Method 2: is clinical supply the binding constraint?  [REASON] on [WEB]+[JOBS] inputs
  Panel (company's 2025 figure): 117 providers
  Visits per provider per year:  120,000 / 117 = 1,026

  Contracted minimum per the live posting: ~15 hours/week [JOBS]
  15 h/wk x 46 working weeks = 690 hours/year
  A 15-minute follow-up implies 4 slots/hour -> 2,760 slots/year at full use of the minimum

  Utilisation of contracted minimum: 1,026 / 2,760 = 37%
  Panel capacity at full use of minimums: 117 x 2,760 = 322,920 visits/year
  Headroom: 322,920 / 120,000 = 2.7x

  Assumptions: all providers on the ~15h minimum; pure follow-up slot length; no admin time;
  no state-licensure mismatch. All of these push the true ceiling down, so treat 2.7x as an
  optimistic upper bound. Even at half of it, supply is not what is limiting revenue.

Method 3 — demand side. Semrush puts medvidi.com at 268,740 visits in July 2026, of which 153,850 organic search and 1,400 paid search [TRAF].

Method 3: implied funnel conversion  [REASON] on [TRAF]+[WEB] inputs
  New patients per year ~= number of initial visits = 25,000 - 35,000
  Per month: 2,083 - 2,917
  Against 268,740 monthly sessions: 0.78% - 1.09% session-to-paying-patient

  Reverse direction, taking 1.0% as the assumption:
  268,740 x 12 x 1.0% = 32,249 new patients/year
  32,249 x $581 first-year value = $18.7M

  Assumption: one paying patient per new session cohort at a flat rate; ignores ezcareclinic.io
  traffic (152.4K/month as of March 2024 [TRAF], weak), which would push the true conversion
  rate lower. Method 3 does not independently prove revenue; it proves that Method 1's answer
  implies an ordinary DTC-health conversion rate rather than an implausible one.

The 1,400 paid-search sessions deserve their own sentence. This is not a paid-acquisition business. Organic search is 57% of sessions and paid search is 0.5% [TRAF]. Whatever MEDvidi spends on growth, it is not buying clicks on Google at meaningful volume. Paid social is not visible in this data and is not ruled out (section 4).

Method 4 — revenue per employee. This one fails, and the failure is the finding.

Method 4: revenue per head  [REASON] on [WEB]+[JD]+[MKT] inputs
  MEDvidi revenue (Method 1):        ~$20.3M
  Headcount, company's own claims:   "300+" (2023, [ARC]); "200+" (2022, [ARC])
     20.3M / 300 = $67,700 per person
     20.3M / 200 = $101,500 per person

  Listed comparables, FY2025 [MKT]:
     LifeMD:        $194.1M / 389 employees   = ~$499,000 per employee
     Hims & Hers:   ~$2.35B / ~1,350          = ~$1,740,000 per employee

  MEDvidi is 5x-7x below the nearer comparable.

There are only three ways that arithmetic can be true. Either the "300+ people" count includes the 117 1099 clinicians and their "designated Provider Assistants" [JOBS], so the employee denominator is not comparable; or the cost base is priced in low-cost geographies rather than US dollars-per-US-hire; or revenue is materially above $20M. The evidence favours a combination of the first two — a Russian-fluency requirement for the VP of Product [JD], recruiters on @smart-it.io domains [JD], and a 55-plus person product and engineering organisation [JD] against $20M of revenue only reconcile at non-US labour rates. That is the actual business model: US cash-pay prices on the revenue side, offshore engineering and operations on the cost side, joined by organic search. No single source says this; it is what the four methods together force, and it is flagged as inference in section 4.

1.6 Cost structure and contribution margin

Clinician pay is published, per visit, by role [JOBS]:

Everything else the provider needs, MEDvidi pays for: "MEDvidi sponsors your medical and DEA licenses", malpractice coverage, a proprietary EMR with integrated PDMP and e-prescribing, "no software costs", and a "Designated Provider Assistant" who "will handle the majority of administrative tasks" [JOBS]. The older Florida posting adds the tell that this is a cost MEDvidi is deliberately buying down: "Automated admin tasks — forget about manually downloading PDMP reports" and "The lowest no-show rate in the market — ensuring more consistent patient engagement and income for you" [JOBS].

Contribution margin per visit  [REASON] on [JOBS]+[WEB] inputs
  Blended clinician cost, 50/50 MD/PMHNP per [ARC]:
     initial:   (60 + 45) / 2 = $52.50
     follow-up: (40 + 30) / 2 = $35.00
  Card processing at Stripe's published US card-not-present 2.9% + $0.30 [PORTAL confirms Stripe]:
     initial:   195 x 0.029 + 0.30 = $5.96
     follow-up: 159 x 0.029 + 0.30 = $4.91

  Initial:    195 - 52.50 - 5.96 = $136.54   (70.0% margin)
  Follow-up:  159 - 35.00 - 4.91 = $119.09   (74.9% margin)

  Per patient-year at 3.43 visits:
     136.54 + 2.43 x 119.09 = 136.54 + 289.39 = $425.93

  Across the 120,000-visit year (upper mix):
     35,000 x 136.54 = $4,778,900
     85,000 x 119.09 = $10,122,650
     total contribution = $14,901,550 on $20,340,000 = 73.3%

Below that line everything is estimate. Stated as ranges, with the reasoning visible:

Residual for marketing and profit  [REASON]
  Contribution                     $14.9M
  less support/care mid-point      -$2.7M
  less product+eng mid-point       -$4.0M
  less content mid-point           -$0.55M
  less G&A mid-point               -$1.5M
  = $6.2M for all acquisition spend and any profit

Implied CAC if $4M of that is acquisition, against 30,000 new patients/year:
  4,000,000 / 30,000 = $133 per acquired patient

  Compare first-visit contribution: $136.54

That last comparison is the most important structural fact in the cost section. If blended CAC sits at or under about $137, the first visit repays acquisition entirely and every follow-up is close to pure contribution. With paid search effectively off, a mostly-organic funnel plausibly lands well below $137, which is the actual reason a $20M clinic can afford a 55-person product and engineering organisation without going bankrupt. The corollary is uncomfortable and worth saying in the interview: the business's returns are driven by an SEO position, and the product organisation's marginal contribution is mostly about patient lifetime, not acquisition.

1.7 The price history is a record of two strategic reversals

Wayback shows MEDvidi has been three different businesses, priced three different ways.

May 2022 — a membership business with therapy and insurance. The FAQ then read: "If you're on the Medication management plan, your first month's membership will only cost $49! Prescriptions will be billed separately and are covered by insurance." "If you're on the Therapy & Medications management plan, your first month's membership will only cost $219." "The first month of MEDvidi's Therapy plan costs $199. This subscription includes bi-weekly sessions with a licensed therapist, unlimited messaging with our CareTeam…" And the self-description: "MEDvidi is an online mental health treatment platform that offers professional online mental health care and medication management to our patients for a fixed monthly fee" [ARC] 2022-05-25. Three subscription tiers, therapists on staff (the page lists PsyD, PhD, LCSW, LMHC, LPC licences), unlimited chat, and an explicit $49 loss-leader first month.

December 2022 to September 2023 — no published price, and payer relationships. The pricing answer became "The cost of an appointment depends on the service you've chosen and your location" [ARC] 2022-12-08 — i.e. prices varied by state and were not published. And the insurance answer was not "no": "MEDvidi currently collaborates with several insurance providers by state: California: AETNA, Anthem BCBS, BCBS, Cigna Healthspring, Humana, Tricare, UHC; Florida: AETNA, AETNA Better Health, BCBS Florida, Cigna, Florida Health Care Plan, Humana, Molina, UHC; New York: AETNA, Cigna, Emblem Health, Empire BCBS, Excellus, Humana, Molina, UHC; Texas: AETNA, BCBSTX, Christus Health Plan, UHC; Colorado: AETNA, Anthem BCBS, Cigna, Kaiser; Washington: AETNA, Kaiser, Regence/Premera, UHC" [ARC] 2022-12-08. By 2023-09-13 the page title had become "MEDvidi | EZCare Clinic" [ARC] — the merger showing up in the <title> tag.

By 6 November 2023 — the current model. A single flat national price, "The majority of our services cost $195 for an initial appointment and $159 for a follow-up appointment", and the payer list replaced by "Currently, MEDvidi doesn't accept insurance. We work on collaborating with several insurance providers" [ARC] 2023-11-06. That capture also carries the DEA flexibilities "extended until December 31, 2024", which dates the change to the exact window in which the November 2023 DEA telemedicine deadline was being resolved.

Price stability  [REASON] on [ARC]
  $195/$159 present in captures: 2023-11-06, 2024-01-15, 2024-05-22, 2024-09-02, 2025-01-05,
  2025-05-09, 2025-09-04, 2026-01-03, 2026-05-10, and live 2026-08-18.
  2023-11 to 2026-08 = 33 months with no change to either headline price.

Thirty-three months of frozen prices in a category where costs and regulation moved repeatedly. That is either strong pricing discipline or a stale price. Given that the same 33 months contain the sentence "we work on collaborating with several insurance providers" — unchanged — the more likely reading is that pricing simply has not been anyone's project, and Prime is the first move on it since 2023.

Two reversals, then. Subscription → pay-per-visit (2023) → subscription again (2026). And payer relationships → cash-only (2023) → "working on it" ever since. A candidate who knows this cannot be surprised by "why don't you take insurance?" — the honest answer is that they used to, in six states, and stopped.

1.8 The no-insurance decision, interrogated

What cash-pay buys MEDvidi, all of it visible in the operating model: complete price control (a flat $195/$159 nationally, set once and held for 33 months); no credentialing, so a new state costs a licensed provider and nothing else — which is how the footprint went "from 12 to 36 states in 3 years" [WEB]; no claims, no denials, no accounts receivable, no prior authorisation; cash collected before the service (part at booking, part just before the visit [LEGAL]), so working capital is negative in MEDvidi's favour; and the freedom to charge a $100 no-show fee, which no payer contract would let it keep.

What it costs, and where the mitigations are:

• A hard ceiling on the addressable market. Only patients who will pay $195 out of pocket for a psychiatric evaluation are in the market at all.
• HSA/FSA is the partial answer, and MEDvidi has found it. The ADHD medication page advertises "Clear pricing: pay-per-visit; HSA/FSA accepted" [WEB], and the portal carries an active HSA_FSA_test1 flag gating a card-payment label on non-membership payments [PORTAL]. Pre-tax dollars materially soften the sticker price for the segment that has an HSA, and testing the label rather than the capability suggests the mechanism works and the question is whether saying so converts.
• The expensive part is already insured. MEDvidi charges only for the appointment; the medication is filled at the patient's pharmacy on the patient's own benefit. The intake asks for an "insurance ID if applicable" [WEB] and the pharmacy FAQ walks patients through prior authorisation and appeals for the drug [WEB]. So "cash-pay" applies to the consult, not to the therapy. That is a much smaller ask than it first appears.
• Adverse selection toward drug-seeking intent. The site's information architecture is organised by branded drug — Adderall, Vyvanse, Xanax, Ambien, Zepbound [WEB] [SM] — which is superb for capturing medication-intent search and structurally selects for patients who arrive wanting a named controlled substance. MEDvidi's own FAQ has to carry three separate answers on this: booking "does not guarantee receiving a prescription", "Why was I prescribed medication different from what I requested?", and "the necessity for prescribing any medication is determined by your provider" [WEB]. That is the sound of a funnel whose top is wider than its clinical gate — and every rejection is a paid visit that generates a refund request or a complaint rather than a follow-up.
• Zero payer defensibility. No contracts, no network position, nothing a competitor cannot replicate in a quarter.
• Direct exposure to any competitor who does take insurance. Which brings us to 1.9.

1.9 The competitive price squeeze, quantified

The recurring-leg comparison  [REASON] on [COMP]+[WEB]
  Follow-up as a share of the initial price:
     MEDvidi:        159 / 195 = 81.5%
     Circle Medical: 100 / 179 = 55.9%
     Done:            79 / 299 = 26.4%   (monthly, not per visit)

  MEDvidi's follow-up against the field:
     vs Klarity's $51 self-pay entry:      3.1x
     vs Circle Medical cash follow-up:     1.6x
     vs an insured Circle Medical visit:   ~4.0x

MEDvidi is the only company in this set that does not discount the recurring leg. That is defensible if the initial visit is a loss-leader — but it is not: the initial carries a $136.54 contribution [REASON]. So MEDvidi charges near-full price at both ends and realises 3.43 visits a year, while Done charges $79/month and Klarity opens at $51. The most plausible causal story — and it is a story, not a finding — is that the $159 follow-up is exactly what suppresses cadence from ~12 toward ~3.4. Prime at $159/month does not test that story; it monetises around it.

The Klarity line is the sharpest single competitive fact in this Area: a direct competitor for the same ADHD-medication search intent now advertises $51 self-pay and accepts 400+ insurance plans, Medicaid in many states, and HSA/FSA [COMP]. MEDvidi's answer to "why does it cost $195?" cannot be "because telehealth costs that", because a competitor on the same keyword says otherwise.

1.10 ESA letters: a $354 document in a $150 market

MEDvidi sells ESA letters, but not the way the market does. Its own page: "MEDvidi can issue an ESA letter only based on a detailed evaluation … After an initial appointment, you will have at least 30 days to follow the received treatment plan. Then, during a follow-up video visit, a healthcare provider will assess your health dynamics and prepare an ESA letter if deemed appropriate — the letter will be ready in up to 3 business days" [WEB]. No price is published on the page.

Cost of a MEDvidi ESA letter  [REASON] on [WEB]
  Mandatory initial visit:                     $195
  Mandatory follow-up visit, 30+ days later:    $159
  Minimum total:                               $354
  Minimum elapsed time:                        ~33-35 days

Market comparison [COMP], secondary except CertaPet:
  Typical online ESA letter:                   $99 - $250
  CertaPet, published:                         $149.99 (housing letter)
  MEDvidi versus CertaPet:                     354 / 149.99 = 2.36x, plus a 30-day wait

Read charitably, that is a compliance posture: MEDvidi refuses to sell a letter as a document and insists on a treatment episode, which is both better medicine and a much smaller regulatory target than the ESA-mill category. Read commercially, it is a page ranking for a $150 intent that converts only visitors willing to pay $354 and wait five weeks. The two readings are compatible and both are true; what a VP of Product needs to know is which one the company thinks it is doing, because the product answers diverge completely (harden the funnel and own the compliance story, versus stop paying to rank for a query the business cannot serve).

Note also that the ESA path is structurally identical to the weight-loss bundle: two visits, one outcome, $354 versus $249. The bundle is discounted; the ESA path is not.

1.11 Weight loss: different economics, and a dangerous name collision

MEDvidi's weight-loss line is consult-only. The page prices an introductory consultation at $195, a bundle of introductory plus follow-up at $249, and follow-ups at $159, and describes prescribing "GLP-1 if appropriate" [WEB]. Nothing on medvidi.com offers to supply, compound or ship medication; the FAQ's "you only pay for an appointment" [WEB] and the pharmacy FAQ's entire framing — your pharmacy, your insurance, your prior authorisation [WEB] — confirm the drug never touches MEDvidi's balance sheet.

Weight-loss patient-year, MEDvidi  [REASON] on [WEB]
  Bundle entry + two more follow-ups: 249 + 2 x 159 = $567
  Bundle entry only:                                  $249
  Contribution at ~75%:                        $187 - $425

Contrast the medication-inclusive model. MEDVi — a different company, at medvi.org, which launched in September 2024 and reported roughly $401M of 2025 revenue on ~250,000 customers, implying about $1,604 per customer-year — sells compounded semaglutide at $299/month ($179 on an annual prepay) and tirzepatide at $399/month ($249 on annual prepay), medication shipped included [PRESS] [COMP], both secondary and weak.

So MEDvidi occupies the high-margin, low-ARPU slice of a market whose economics are dominated by drug supply and drug price. Adding GLP-1 to a psychiatry clinic makes sense on the cost side — same prescriber, same 15-minute follow-up, same portal, zero inventory — and it makes sense on the demand side, because weight-loss search volume dwarfs psychiatric search volume. What it does not do is compete with Hims or Ro or MEDVi on the offer; it competes for the evaluation only, at a price those companies bundle into a monthly fee that includes the drug.

And there is a brand problem attached. MEDvidi and MEDVi are different companies, and the public record already confuses them: the BBB complaints tab on the "Medvidi Inc" profile — the one at MEDvidi's own San Jose address, with MEDvidi's own DBAs — carries recent complaints whose text names "MEDVi" and describes tirzepatide auto-refills at $399.01 and a $179/month price stepping to $350, which are MEDVi's published prices, not MEDvidi's [BBB] [COMP]. See section 3; the practical consequence is that MEDvidi's reputational surface, and its branded search, are being polluted by a much larger namesake in an adjacent category.

1.12 Corporate and contractual shape, where it touches the money

• Entity and structure. "Medvidi Inc", 4010 Moorpark Ave #114, San Jose CA 95117-1803; business started 2020-04-08, incorporated 2021-09-06; BBB accredited since 2025-03-07 with an A+ rating; alternate names "EZCare Clinic" and "Mango Clinic" [BBB]. The About Us page gives the same address and names Vasili Razhnou as founder, after "establishing traditional clinics in San Francisco and Miami" [WEB].
• MSO / friendly-PC. The Terms distinguish MEDvidi Health (owns the sites, bills the patient, provides "a certain management services agreement") from MEDvidi P.C., which "alone controls and supervises the Medical Services and clinical decisions", whose Practitioners are "independently insured" contractors [LEGAL]. Standard for the category and materially relevant: the revenue line a buyer or investor would diligence is a management fee from the PC, not patient revenue directly.
• Two live consumer funnels. mangoclinic.com 301-redirects to medvidi.com, consolidating its SEO equity; ezcareclinic.io is still live, titled "MEDvidi | Mental Health Clinic USA", shares the same nine-doctor roster, and has its own signup entry at start.ezcareclinic.io/signup [WEB]. MEDvidi's own funnel is join.medvidi.com/signup [WEB]. Any revenue or traffic figure for "medvidi.com" alone understates the business.
• Legacy contract language contradicting the current model. The Terms still say the patient authorises "MEDvidi Health to submit claims and bill Medical Services to your insurer on your behalf"; still describe a "membership" with fees "automatically charged to your preferred payment method … until you cancel your registration"; and still say, flatly, "We do not offer refunds" [LEGAL] — against a Refunds & Fees Policy that grants full refunds in six situations [LEGAL]. The claims-billing clause is a strategic tell: the contract for taking insurance was never removed. The auto-charge and no-refund clauses are a consumer-protection liability, and they become sharper the moment Prime goes live, because then the terms will finally describe something real.
• Stripe and PayPal are the processors (stripeService, clientSecret, addCard, and PayPal references) [PORTAL]. PostHog carries the analytics and feature flags; Sentry the error reporting; the video call runs on Twilio (the FAQ links patients to the "Twilio Video Diagnostics Test") [PORTAL] [WEB].

1.13 Self-reported growth series, and the outcome claims

From About Us, as of 2026-08-18 [WEB], with the 2024-vintage version for corroboration [ARC]:

The appointments column is monotonic and the 2022 figure agrees between the old and new page versions, which is why it is used above. The states and providers columns are not monotonic (34→23 states, 91→71 providers across 2023→2024) — see section 3.

Quality and satisfaction claims, all self-reported: "4.85/5 Our patients rated their providers", "89% of our patients rated their appointments 5 out of 5", "33k+ patients rated their appointments" [WEB] /providers/ /reviews/; "89% of our clients achieve clinically significant results in 8 to 12 weeks" [WEB] /services/online-assessment/; "Appointments within 24 hours" and "24/7 patient support" [WEB].

The stated 2026 goal is worth reading twice: 70,000 patients per month. Against 35,000 patients per year, that is a 24x step, and it is paired with re-adding psychotherapy — the service line the company cut in 2023. Whether that goal is live or is stale copy from the 2023-era page (it appears verbatim in the 2024-01 capture [ARC]) is itself a question for the CEO.

2. Capability / object table

3. Reconciliation notes

Reconciliation note — "no subscription model" versus MEDvidi Prime. [WEB] states, live on 2026-08-18, "we have no subscription model … you may book one appointment at a time." [PORTAL] contains a shipped, Stripe-backed, $159 recurring membership with routes, dunning, billing history and a 28-event cancellation funnel, in a bundle last modified 2026-08-14. Resolved in favour of [PORTAL]: shipped code that renders a charge amount is stronger evidence of what the business does than marketing copy, and the two are reconcilable if Prime is flag-gated to a test cohort (Membership_Launch_Test_1) whose existence the public FAQ has not yet caught up with. The correct statement is therefore: pay-per-visit is the published model; a membership is shipped and in test.

Reconciliation note — BBB complaints are not MEDvidi's dispute rate. The BBB profile for "Medvidi Inc" at MEDvidi's own address, carrying MEDvidi's own DBAs, reports 159 complaints in three years and 58 closed in the last twelve [BBB]. But the three most recent complaint texts name "MEDVi", describe a $399.01 tirzepatide auto-refill, a shipment of GLP-1 medication that "arrived melted", and a $179/month price that stepped to $350 [BBB] — every one of which matches MEDVi's published GLP-1 price ladder [COMP] and none of which matches anything MEDvidi sells, since MEDvidi supplies no medication at all [WEB]. Resolved: the complaint counts on that profile are contaminated by brand collision and cannot be used as a MEDvidi dispute proxy. Parked in section 4 with the evidence that would settle the true dispute rate. The collision itself is a real finding and belongs in the competitive and brand discussion.

Reconciliation note — the About Us states and providers series. [WEB] lists, in DOM order matching the year tabs 2025→2019: 2025 = 36 states / 117 providers, 2024 = 23 / 71, 2023 = 34 / 91. Appointments in the same order are monotonic (35K → 80K → 90K → 120K) and the 2022 figure matches the older page version [ARC], which validates the ordering for that column. The states and providers columns are not monotonic under any pairing. Not resolved. Two readings survive: a genuine 2024 contraction (plausible — 2024 was the year of the federal indictment of Done Global executives and widespread pharmacy refusal of telehealth stimulant prescriptions), or an authoring error in the Elementor tab widget. Parked. The appointments series is used in this Finding; the states/providers series is not.

Reconciliation note — how many providers. [WEB] About Us says 117 providers for 2025 and "a team of 100+ licensed medical providers"; the 2024-vintage page said "45+ medical providers" [ARC]; /providers/ publishes nine named physicians and /reviews/ adds one nurse practitioner; sm_team.xml contains 35 team profiles, of which roughly 10 are clinicians and the rest are health writers and medical reviewers [SM]. Resolved as consistent rather than contradictory: the panel is ~117 contracted clinicians of whom ~10 are published as patient-facing marketing faces. The capacity arithmetic in 1.5 uses 117 and flags its optimism.

Reconciliation note — "35,000+ patients … every year". [WEB] reads "35,000+ patients & 120,000 online visits every year", which this Finding takes as both figures being annual, giving 3.43 visits per patient-year. An alternative reading — 35,000 as a cumulative patient count — would make the sentence internally incoherent with "every year" and would put visits per patient at 11.4, i.e. essentially the monthly cadence, which contradicts the $159-follow-up price pressure and the very existence of a membership test. Resolved in favour of the annual reading, and flagged in section 4 as the single input whose failure would most change this Area's conclusions.

Reconciliation note — refunds. [LEGAL] Terms and Conditions say "We do not offer refunds"; [LEGAL] Refunds & Fees Policy (updated 2025-05-05) grants full refunds in six enumerated cases. Resolved in favour of the Refunds & Fees Policy as the later, more specific and operationally instrumented document (verification_issue_refund exists in the portal [PORTAL]). The Terms are stale membership-era boilerplate.

Reconciliation note — conditions treated. [WEB] FAQ lists ADHD, anxiety, depression, insomnia, stress, chronic fatigue syndrome and OCD, and omits weight loss and ESA; the site navigation and service pages sell both weight loss and ESA letters [WEB], and the Terms list "weight loss needs" among Medical Services [LEGAL]. Resolved in favour of the service pages and Terms; the FAQ list is stale.

Reconciliation note — 35 versus 36 states. The FAQ enumerates 35 state abbreviations [WEB]; About Us says "Open states: 36" for 2025 and "Growth from 12 to 36 states" [WEB]. Unresolved and immaterial to the economics; noted so the discrepancy is not mistaken for a finding.

4. Open Questions / Parked

1. What interval is Prime billed on? The portal proves the amount ($159) and proves the coverage language ("monthly visits") but the subscription interval is not in the bundle [PORTAL]. Monthly is strongly implied and is assumed throughout section 1.4. Settled by: the Prime terms page once it publishes, a Stripe price object, or the CEO. If the interval were quarterly, the 3.28x multiple collapses to ~1.1x and Prime becomes packaging rather than repricing.
2. How many visits does Prime actually include per cycle? A "Membership Coverage Warning Viewed" event exists [PORTAL], which implies a cap or an exclusion, but its copy was not in the chunks retrieved. Settled by: the membership terms, or the coverage-warning string.
3. Is the realised 3.43 visits/patient/year driven by churn or by 90-day prescribing? The two have opposite product answers. Settled by: a cohort curve of visits-per-patient by month since first visit, and the distribution of days-supply per prescription. Both exist in MEDvidi's own data.
4. Is "35,000+ patients" annual or cumulative? Taken as annual; see section 3. Settled by: monthly active patient counts.
5. What is the true no-show and late-cancel rate? MEDvidi markets "the lowest no-show rate in the market" to clinicians [JOBS] and gives it no number. Fee revenue is modelled at 3–8% and is the softest line in section 1.6. Settled by: the appointment-status distribution.
6. What is the true dispute/chargeback rate? The BBB count is contaminated (section 3), and Trustpilot and Reddit both blocked unauthenticated retrieval on 2026-08-18. Settled by: Stripe dispute rate, or the refund-code breakdown.
7. What is blended CAC, and is there paid social spend? Paid search is ~1,400 sessions/month [TRAF]; Meta and TikTok accounts exist [WEB] and their spend is not observable here. The $133 CAC in section 1.6 is a residual, not a measurement. Settled by: the marketing P&L, or the Meta Ad Library and Google Ads Transparency Center (an A2 task).
8. Total company revenue across all funnels. ezcareclinic.io runs its own signup [WEB] and its traffic is not in the July 2026 Semrush figure for medvidi.com [TRAF]. Whether its appointments are inside the "120,000" is unknown. Settled by: revenue by brand.
9. What headcount does "300+ people" count? If it includes 117 contractors plus provider assistants, Method 4's revenue-per-employee gap largely dissolves. Settled by: an org chart, or the FTE/contractor split.
10. What would insurance actually pay? This Area deliberately does not state a reimbursement number for a 15-minute psychiatric follow-up, because a wrong one would be worse than none. Settled by: a commercial contracted rate for the relevant E/M code with a telehealth modifier in two or three of MEDvidi's largest states, plus the corresponding Medicaid fee schedule. Without that, the only defensible statements are the qualitative ones in section 1.8 and the observation that the "we work on it" sentence has stood unchanged for 33 months [ARC].
11. Did MEDvidi ever actually bill payers, or was the 2022 list an out-of-network "your plan may reimburse you" claim? The 2022 wording — "collaborates with several insurance providers by state" and "may be covered by your insurance depending on its terms" [ARC] — is ambiguous. The Terms' claims-billing authorisation [LEGAL] argues for real billing. Settled by: whether an NPI/TIN was ever contracted, or the CEO.
12. Are the 2022 subscription prices ($49/$199/$219) intro-only, and what was the ongoing price? Every capture says "first month's membership" [ARC] and none states month two. Settled by: a 2022 pricing page capture, if one exists.
13. Is the "70,000 people monthly across all 50 states" goal live or stale? It appears verbatim in the 2024-01 capture [ARC] and in the current page [WEB]. Settled by: the CEO, in one question.
14. Cost of the sponsored licence programme. MEDvidi pays for each provider's medical and DEA registrations [JOBS]; no dollar figure is stated here because the current DEA practitioner registration fee was not verified from a primary source in this pass. Parked rather than guessed.

5. What this does NOT cover

• The funnel and the acquisition channel. Traffic, keywords, programmatic state pages, lead magnets, ad libraries and conversion-rate mechanics belong to A2. This Area uses a single July 2026 traffic figure and a paid/organic split, and only as a consistency check.
• The regulatory cliff. The DEA telemedicine flexibilities expiring 2026-12-31 [WEB] are the largest single input to any forecast in this document and are A4's subject. Section 1.5's 2026 run-rate assumes the current regime continues; it should be read as conditional.
• Clinical supply and operations in depth. Who the 117 providers are, their licensure spread, where the three shipped AI features bite, and the Provider Assistant model are A5's. Section 1.5's capacity test is deliberately crude and labelled as an upper bound.
• The competitive landscape. A6 owns it. The four price points in 1.9 are the minimum needed to interrogate the cash-pay decision; three of them are review-site secondary.
• Ownership, cap table, funding and the Smart IT relationship. A7's. No funding history, valuation or investor is asserted here; the revenue-per-employee argument in 1.5 leans on A7's territory and is labelled as inference.
• Anything behind authentication. No account was created, no login attempted, no consult purchased, no host scanned. The portal evidence is entirely from publicly served static JavaScript and one SVG asset. There is therefore no observed price for Prime as a patient would see it at checkout, no observed membership terms page, and no observed portal service catalogue.
• A P&L. Everything below the contribution line in 1.6 is a modelled range, not a statement about MEDvidi's actual spending. Profitability is not asserted in either direction.
• Per-service prices that are not published. Prescription refill, medication management and prescription management have no published price; the FAQ's "the majority of our services cost $195 … and $159" [WEB] is treated as covering them, and the residual possibility of service-specific pricing inside the portal is parked.

6. What this means for a VP of Product

The first thing to do with this Area is to stop treating MEDvidi as a consultation business. It is a subscription business that has not been charging like one, and its own engineering has already reached that conclusion. The clinical mechanism forces a patient on a Schedule II stimulant back through the door roughly every 30 days — a new prescription is legally required every time [WEB], refills unlock at 25 days and pharmacies dispense at 30 [WEB] — while the commercial system collects 3.43 visits a year [WEB]. That is a ~30% capture rate against the cadence the medicine itself dictates. Everything a product organisation could usefully do here lives inside that gap. The single number to walk into the room with is not $195 and not $20M; it is 3.43, and the follow-on question is whether the gap is churn or 90-day prescribing, because the answer determines whether the fix is retention design or repricing. That question is answerable from MEDvidi's own event stream in days, and asking it is a better opening move than any opinion.

The second thing is to be extremely clear-eyed about MEDvidi Prime, because it is both the right move and the one most likely to be run badly. Right, because it converts $581 of realised annual value into $1,908 [REASON], needs only ~11% of the patient base to add $5M [REASON], and is accretive whether members use twelve visits or three. Likely to be run badly for three reasons. First, the launch wedge — offering the membership at the instant a patient owes a $50 late-reschedule fee [PORTAL] — is exceptional behavioural design and terrible cohort selection: the population most likely to accept a save-offer at a penalty moment is the population most likely to churn, dispute and complain, so the test will produce a flattering conversion rate and an unflattering retention curve, and someone must insist that the success metric is 3-month retained revenue rather than join rate. Second, the membership improves margin as members consume less care [REASON], which puts the product incentive against the clinical one; the person owning PM, Design, Data and Engineering together [JD] is the only person who can put a guardrail metric on that, and it should be in the launch definition, not added later. Third, the Terms and Conditions currently say both "we do not offer refunds" and that fees will be "automatically charged … until you cancel your registration" [LEGAL] — legacy language that was harmless while nothing recurred and becomes a live consumer-protection exposure the day Prime ships to everyone. Fixing the legal surface is not a legal task, it is a launch dependency.

Third, the pricing conversation cannot be avoided and should not be conflated with Prime. $195/$159 has not moved in 33 months [ARC] while a direct competitor for the same medication-intent search now advertises $51 self-pay and accepts 400+ insurance plans plus HSA/FSA [COMP]. MEDvidi's follow-up sits at 82% of its initial price, where Circle Medical's is 56% and Done's recurring fee is 26% [REASON]. MEDvidi is alone in the set in refusing to discount the leg that recurs — which is precisely the leg its whole business depends on. The tradeoff to name out loud: cutting the follow-up price is the highest-confidence lever on cadence and the highest-risk lever on revenue, because at 74.9% contribution per follow-up [REASON] a price cut needs a big volume response to pay for itself. Prime is the elegant way to have it both ways — hold the $159 nominal price, change what it buys — and that is presumably why it exists. Someone should still run the price test, because if $159 is what suppresses cadence, Prime monetises the symptom rather than curing the disease.

Fourth, know where the leverage genuinely is not. Clinical capacity is not the constraint: at 1,026 visits per provider per year against a ~15-hour weekly minimum, the panel is running at roughly 37% of its contracted floor, with something like 2.7x headroom [REASON]. Acquisition cost is probably not the constraint either: paid search is 1,400 sessions a month against 153,850 organic [TRAF], and the residual CAC estimate of ~$133 sits just under the $136.54 contribution of a single initial visit [REASON] — meaning the first visit plausibly pays for the patient. That combination is what makes the whole structure work, and it is also what makes the structure fragile in exactly one place: the business is a levered bet on an organic search position. A new VP of Product should therefore be extremely reluctant to fund anything that competes for engineering capacity with (a) patient lifetime and (b) whatever protects the SEO asset, and should say so as a cut list rather than a priority list.

Fifth, the two side verticals want opposite decisions, and both decisions are cheap. ESA letters cost a patient $354 and five weeks at MEDvidi [REASON] in a market that transacts at $99–$250 [COMP]. Either that is a deliberate compliance moat, in which case the page should say so loudly and stop competing on a query it cannot win, or it is a leak, in which case the honest move is to stop ranking for it. Weight loss is the mirror image: consult-only at ~$249–$567 per patient-year [REASON] with no drug on the balance sheet, competing for evaluation intent against medication-inclusive players charging $179–$399 a month all-in [COMP] — one of whom, MEDVi, has a name so close that complaints about its tirzepatide auto-refills are landing on MEDvidi's BBB profile [BBB]. The weight-loss line is a legitimate high-margin use of spare prescriber capacity, and it is not a business MEDvidi can win at the offer level without taking on drug supply, which is a different company. Say that explicitly rather than letting it drift.

Sixth, and this is the one that should be asked of the CEO in the first ten minutes: the product and engineering organisation is very large for this revenue. Fifty-five-plus people in PM, design, data and engineering [JD] against roughly $20M [REASON] puts revenue per head somewhere between $68k and $102k, against ~$499k at LifeMD [MKT]. That ratio is only survivable with a non-US cost base — which the Russian-fluency requirement and the smart-it.io recruiters [JD] both point at. So the real question is not "can we afford this org" but "what is this org for". A clinic doing 120,000 visits a year does not need thirty engineers to run a booking flow. The two-track structure — Core and AI Clinic [JD] — plus three shipped AI features that all point at internal cost (AI Scribe on notes, Chart Review AI on protocol deviations, Agentic AI Receptionist owning internal support processes) [JD] and a recruiting pitch built around removing administrative burden from clinicians [JOBS] together suggest the answer is that MEDvidi is building an operating system for a clinic and using its own clinic as customer zero. If that is right, the economics of this Area are the wrong denominator for judging the product org, and the VP of Product's first strategic act is to establish which business the engineering is being spent on. If it is only the clinic, the org is over-built and the cut list writes itself. If it is a platform, then someone needs to say who the second customer is, and neither the website nor the job posting mentions one.

Finally, the five questions this Area earns. What interval and what visit allowance does Prime carry, and what is the retention curve on the late-reschedule cohort versus everyone else? Is the 3.43 visits per patient a churn number or a days-supply number? Why did we stop taking insurance in 2023 in six states, and what specifically has to be true for us to start again — because the FAQ has promised it for thirty-three months? What is the engineering organisation actually building — our clinic, or a product other clinics will buy? And which is the bigger 2027 revenue line: converting the existing 35,000 patients to Prime, or adding the next 35,000 patients — and does the answer change if the DEA flexibilities are not extended past 31 December?