Skip to content
Skip to content

Guide

How to Start a Telehealth Business: the Seven Decisions That Actually Matter

Starting a telehealth business comes down to seven decisions: what you sell and to whom, which of four paths you take to a working clinic, how the legal chassis works when you are not a doctor, where clinicians and medication come from, the certification chain that unlocks advertising, and the unit economics that decide if any of it was worth doing. Most guides to this are a feature list wearing a numbered headline; this one is organized around those decisions, written by operators who have run telehealth brands at scale, and it links to a deeper guide, a directory, or a calculator at every decision where one exists. Budget fifteen minutes; the decisions themselves deserve weeks.

14 min readUpdated September 7, 2026

Decision one: the wedge (what you sell, and to whom)

Telehealth businesses do not start with technology; they start with a specific person buying a specific thing. The 2026 demand map is unambiguous: weight loss and GLP-1s are the largest wedge by an order of magnitude, followed by hormones and TRT, sexual health, hair, skin, sleep, and longevity. Each wedge has its own supply chain, its own regulatory posture, and its own advertising rules, which is why 'we'll offer everything' is a plan to launch nothing.

The stronger predictor of success is not the condition you pick but the audience you already have. A med spa client base, a gym membership, a creator following, a supplement brand's customer list: an audience that already trusts you compresses the hardest problem in this industry (customer acquisition cost) into a warm introduction. If you have no audience at all, your real first business is building one, or budgeting to rent one at market ad prices once you are certified to advertise.

Pick one launch wedge and one audience, write both down in a sentence, and let every later decision answer to that sentence. Adding a second care line later is a settings change on most modern platforms; recovering from an unfocused launch is not.

The sentence to complete before anything else: 'We help [this audience] get [this treatment program] under [this brand].' Every vendor conversation goes better when you can say it.

Decision two: the path (build, stitch, operated, or buy)

There are four ways to get a working clinic behind a brand, and confusing them is the most expensive mistake in the category. Building from scratch means recruiting clinicians, contracting pharmacies, writing software, and standing up a compliance program: development-shop guidance alone runs $25,000 to $150,000+ before the clinical layer, and the timeline is measured in quarters. It is the right path for funded companies with unusual requirements and the wrong first spend for almost everyone else.

Stitching means assembling the stack from specialist vendors: a clinician network here, storefront software there, a pharmacy relationship, a high-risk payment processor, a compliance consultant. It is faster and cheaper than building, and it makes you the general contractor: every gap between vendors is yours, and so is most of the compliance surface. The stitched stack is also where most of the search queries in this industry come from, which tells you how much time operators spend managing it.

The operated path means one platform runs the whole clinical business behind your brand while you keep the storefront, the audience, and the retail pricing. You trade some control for speed and a shorter list of things that can go wrong. Finally, buying an existing telehealth business is a real market (brokered listings trade on multiples of seller earnings), worth considering when you want revenue on day one and can diligence what you are buying.

Our comparison hub walks the trade-offs in detail, and the startup cost calculator on this site prices the build and stitch paths line by line from vendors' own published pricing. The honest summary: choose based on how much of a clinic you want to operate, not how much of one you want to own; the operated path leaves you owning the brand, the patients, and the economics without operating the machinery.

  • Build: quarters of time, five to six figures before launch, every obligation yours
  • Stitch: faster, cheaper, and you become the integrator of five to nine vendors
  • Operated: launch in days behind your brand; the platform runs the clinic
  • Buy: revenue on day one at a broker-market price; diligence is everything

Decision four: clinical supply (where the clinicians come from)

A telehealth business needs licensed clinicians in every state it serves, credentialed, insured, and available on a cadence patients accept. The market supplies this three ways: enterprise clinician networks that contract at volume, asynchronous visit APIs for engineering-led teams, and operated platforms where the network is part of the package. The category has consolidated and shifted (one major network was acquired in late 2025; another has pivoted toward enterprise programs), so verify current positioning rather than trusting last year's roundup.

The questions that separate networks are practical: how many states, synchronous or asynchronous or both, what the review turnaround actually is at your volume, who carries credentialing and malpractice, and what a visit costs all-in. Turnaround matters commercially, not just clinically: an intake that waits a day converts worse than one reviewed within the hour. Our clinician-network guide covers the model in depth, and the vendor landscape pages on this site profile the major suppliers.

Decision five: medication supply (the pharmacy layer)

If your wedge involves prescription medication, the pharmacy layer decides your margin and carries your sharpest regulatory risk. The landscape splits into 503A compounding pharmacies (patient-specific prescriptions), 503B outsourcing facilities (larger-batch production under stricter federal oversight), and retail or mail-order pharmacies for FDA-approved products. Which you need depends entirely on the wedge; our 503A versus 503B guide is the primer.

For GLP-1 wedges specifically, the environment changed hard in 2025: with the shortages resolved, mass-compounded copies of the branded drugs became generally impermissible, enforcement letters went out in waves, and remaining compounded supply rests on patient-specific clinical grounds that are actively contested. Nothing in this paragraph is legal advice, and everything in it has a date on it: verify the current posture before you build a business on it, and treat any vendor who waves the question away as a red flag.

Commercially, the question is who keeps the medication margin. Pass-through models leave you negotiating fill prices with pharmacies; included models fold medication into the platform's flat rate. Neither is wrong, and the difference decides your per-patient economics, so it belongs in writing before you sign.

Compounded medications are not FDA approved. They are prescribed at a licensed clinician's discretion, not all patients qualify, and your marketing must never promise that anyone will be prescribed anything.

Decision six: the certification chain (what advertising requires)

You can launch a telehealth brand without any certification. You cannot advertise prescription services on Google or Meta without it, and paid acquisition is how this category scales. The chain runs in a fixed order: LegitScript Healthcare Merchant Certification for your website (published fees: $975 per site to apply, $2,150 per year certified, with an optional expedite that buys queue position), then each ad platform's own healthcare approval. Plan two to four months end to end from a prepared application, and use the waiting period to make the site certification-grade, because reviewers read it.

This site carries the full toolkit for that chain: a free readiness checker that scans your site against the published standards, a cost calculator, a pre-application checklist, an eleven-guide certification series, and the specific answer to the most-asked question (whether Google requires LegitScript at all: for prescription-service advertisers, yes). The wrong path is the workaround family (fresh domains, borrowed accounts), which converts a fixable rejection into a permanent platform ban.

Decision seven: the unit economics (whether it was worth doing)

The economics of a subscription care business reduce to four numbers: what a patient pays per month, what serving them costs per month (visits, medication, platform), what acquiring them cost, and how many months they stay. Public market signals put compounded GLP-1 program pricing in the low hundreds per month, acquisition costs in the low-to-mid hundreds per paying patient at market ad prices, and retention as the number everyone underestimates: a large share of GLP-1 patients discontinue within months, which is why refill operations and follow-up care are business machinery, not aftercare.

Run your own numbers before you commit to anything: the startup cost calculator on this site prices the launch, and the earnings calculator models program revenue at your assumptions. Both are illustrative by design. Any projection is a model, not a promise; your results depend on your audience, pricing, and execution, and no honest operator, this one included, will guarantee earnings.

The structural choice you control is rate shape. Flat product rates mean your margin flexes while your costs do not; revenue-share and markup models mean the platform's take grows with your success. Whichever you choose, get the all-in per-patient math in writing at your target volume, including the quoted-per-deal items (visits, fills, processing) that published pricing never covers.

The launch sequence and what it costs

On the operated path, the sequence is short: configure the storefront (brand, care lines, retail prices inside the allowed band), connect payments, point your audience at your domain, and start the certification chain in parallel if you plan to advertise. Days to a working clinic; weeks to ad eligibility. EmbedCare's self-serve Launch tier starts at $495/mo, with operated partnerships on flat product rates fixed in a signed partner agreement.

On the stitch path, sequence the dependencies: legal chassis first (nothing else can sign contracts without it), then clinical supply and pharmacy, then software and payments, then certification, then marketing. The startup cost calculator itemizes the budget from third-party published pricing; the realistic pattern is five figures one-time plus a four-to-five figure monthly run rate before the first patient, with the quoted items on top.

Either way, write the launch checklist backward from the first patient: can they find the site, complete an intake, be reviewed by a licensed clinician in their state, receive medication legally, pay you compliantly, and hear from you next month? Every yes has an owner; on the operated path most of the owners are the platform, which is the point of the model.

  • Operated path: storefront and clinic live in days; certification runs on its own clock
  • Stitch path: legal chassis, then supply, then software, then certification, then marketing
  • Both paths: the certification chain gates paid ads, so start it early
  • Both paths: retention machinery decides the business more than launch speed does

The five mistakes that end telehealth launches

After enough post-mortems, the failure patterns repeat. Launching broad instead of picking a wedge. Treating the legal chassis as paperwork to backfill later, which turns into an unwindable mess at exactly the moment a processor or platform asks questions. Signing supply agreements without the all-in per-patient numbers in writing, then discovering the margin was never there. Buying traffic before certification, losing the ad account, and trying workarounds that convert a delay into a ban. And underinvesting in retention because acquisition is more exciting, which caps the business at the size of the marketing budget.

None of these are subtle, and all of them are survivable if caught in week two instead of month six. The cheapest insurance is sequencing: decisions one through seven, in order, with numbers in writing at each step.

Where EmbedCare fits, stated plainly

EmbedCare is the operated path: the clinical network, owned pharmacy supply, storefront, payments, retention machinery, and compliance surface, run as one stack behind your brand, built by the operators behind two LegitScript-certified telehealth brands. If your situation is an audience and a wedge in search of a clinic, that is the exact shape of the product, and a demo will show you your own storefront before you sign anything. If you are set on building or stitching, the guides, directories, and calculators on this site are yours regardless; they exist because informed operators make better partners and better competitors than confused ones.

Frequently asked

How do I start a telehealth business?
Seven decisions in order: pick one wedge and audience; choose your path (build from scratch, stitch vendors, launch on an operated platform, or buy an existing business); set up the legal chassis (a clinician-owned professional entity plus your management company, or a platform that includes it); secure clinical supply; secure medication supply if your wedge is prescription-based; run the LegitScript-then-ad-platform certification chain if you plan to advertise; and model the unit economics in writing before committing.
How much does it cost to start a telehealth business?
Assembling it yourself typically means five figures one-time (legal structure, certification, optionally a custom build) plus a four-to-five figure monthly run rate before your first patient, with visit fees, medication, and payment processing quoted on top. Operated platforms compress most line items into a subscription; EmbedCare's Launch tier starts at $495/mo. This site's startup cost calculator itemizes the budget from third-party published pricing.
Can I start a telehealth business without being a doctor?
Yes. Care is delivered by independently licensed clinicians who make every prescribing decision, and non-clinician ownership is organized through the management-services (MSO) structure, with a clinician-owned professional entity delivering care. The structure is standard and must be set up correctly for your states; operated platforms include it, and our MSO guide explains it in plain English.
How long does it take to launch?
On an operated platform: days to a working branded clinic, because the clinical, pharmacy, and compliance layers already exist. Building or stitching: weeks to months, sequenced legal chassis first. The advertising certification chain (LegitScript, then Google or Meta approval) runs two to four months from a prepared application on any path, so start it early if paid acquisition is in the plan.
Is a telehealth business profitable?
It can be, and the deciding number is usually retention rather than anything about the launch. Subscription care lives or dies on how many months patients stay against what they cost to acquire; refill operations, follow-up care, and honest pricing do more for profitability than any launch decision. Model it with real numbers before committing; any projection is a model, not a promise.
Do I need LegitScript certification to start?
Not to launch; yes to advertise prescription services on Google and Meta. The published fees are $975 per website to apply and $2,150 per year certified, with the platforms' own approvals following. Start the chain early, keep your site certification-grade, and never try workarounds: they convert a fixable rejection into a permanent ad ban.
What is the difference between white-label telehealth and building my own?
White-label means the clinic runs behind your brand but is supplied and operated by someone else, so you launch in days and keep the audience relationship and retail pricing. Building means you recruit clinicians, contract pharmacies, write software, and carry every obligation yourself, which costs quarters of time and serious capital. Our white-label deep dive covers the operated model end to end.

Want pricing for your program, and the Rx menu that goes with this?

The partner overview in one email; a human follows up with pricing scoped to your program.

The fastest way to understand it is to see it running.