You have a health-tech idea, a rough deck, and a set of quotes that disagree by a factor of five for what you described to each agency as the same product. None of them explain the gap. That gap is not a market rate; it is a scoping difference nobody wrote down, and until somebody writes it down you cannot tell which quote is honest.
This page writes it down. Zee Palm publishes its prices — a health-tech MVP build starts at $12,000, a full custom build at $25,000, and the planning that makes either number real costs $1,500 or $1,999 depending on how much regulatory surface your product has. Below is what sits inside those numbers, what sits outside them, and the specific scope decisions that move a health-tech MVP from twelve thousand dollars to six figures.
If you only take one action from this page: buy the planning before you buy the build. A 5-day MVP Planning Sprint is $1,500 and is credited in full toward the build if you go ahead. A founder who arrives at a build with a scoped backlog spends less than one who arrives with a deck, every time, at any agency.
Why the quotes disagree so wildly
Three real variables, and they compound.
1. Whose data is it. An app that stores workout logs and a self-reported mood score is a consumer wellness product. An app that receives a lab result from a hospital system, or transmits anything a clinician wrote, is protected health information moving between covered entities. The second one needs a BAA, an audit trail, access controls, encryption at rest with key management you can explain, a breach-response runbook, and a data model where "delete this user" is a real operation rather than a hope. That is not a feature; it is a floor under every feature. It is the single largest legitimate reason one quote is triple another.
2. Whether a human is on the other end. A self-guided app is software. The moment a licensed clinician, coach, or care manager logs into an admin surface to see patient data, you have built two products: the patient app and the clinical console. The console usually needs role-based access, a caseload view, a documentation trail, and a way for a supervisor to see what a junior did. Founders routinely leave the console out of the brief and then discover it in week six.
3. How many integrations sit in version one. Each external system is a bounded project with its own auth, its own rate limits, and its own edge cases. Zee Palm prices wearable integrations at $3,500 per platform precisely because HealthKit, Health Connect, Garmin, Whoop, Fitbit and Oura are six different jobs, not one. The same logic applies to EHR work: an Epic or Cerner connection over FHIR R4 is not the same size as reading steps from a phone.
The high quote may be covering all three at once, plus a clinical console, plus a year of enterprise process. The low quote may be covering the patient app only and planning to discover the rest later. Neither is lying. They answered different questions, because you asked one question that had not been made specific yet.
The bands, with real numbers
Here is the published band structure, straight from the pricing page, with the health-tech reading of each.
| Budget | What it genuinely buys in health-tech | What it does not |
|---|---|---|
| Under $5,000 | An MVP Planning Sprint ($1,500) or Healthtech MVP Blueprint ($1,999). A clickable or coded prototype on synthetic, non-sensitive data from $3,500 — enough to test with users, run a pilot conversation, or show an investor. | A production app. Anything touching real patient data end to end. |
| $12,000 – $25,000 | A bounded version one in production: patient-facing mobile app or web platform, one or two integrations, a basic admin surface, QA and an app-store submission. Two to four wearable platforms at $3,500 each. | Every feature on the list. A platform that replaces an enterprise SaaS suite. |
| $25,000 – $100,000 | A full custom build: multi-platform, clinical console, integrations, dashboards, plus the year that follows — [maintenance from $1,000/mo](/services/app-maintenance), QA from $799/mo. | A certification. Zee Palm builds HIPAA-aware systems; it does not issue legal sign-off or decide whether your product is a regulated device. |
Two things worth saying plainly about that table.
The $3,500 prototype line is the most under-used product in health-tech. A very large number of founders raise money on a clickable prototype and synthetic data, then build the real thing with the investor's money and a validated flow. Building production software first, and discovering in user testing that the core loop is wrong, is the most expensive mistake available in this category. If you are pre-revenue and pre-raise, the prototype is usually the correct purchase and the MVP build is not.
The $25,000+ band is where most funded health-tech actually lands, and the honest reason is the clinical console plus compliance floor described above — not gold-plating. If your product has clinicians in it, plan for the upper band and be pleasantly surprised, rather than the reverse.
The scope line that decides your number
Ask yourself these six questions before you take another call with an agency. The answers are your quote.
- Does version one store PHI, or can it run on de-identified or self-reported data? If a pilot works on self-reported data, run the pilot on self-reported data. You can add the PHI-handling layer once you know people want the thing.
- Is there a clinician-facing surface in version one, or can the first ten clinicians use a spreadsheet and a weekly call? Concierge operations are not cheating. They are the cheapest way to learn what the console actually needs to do.
- How many devices or systems must sync on day one? Almost always the answer should be one. Pick the platform your target users actually wear, not all six.
- Do you need a payment flow in version one? If you are selling B2B pilots, no — you invoice. Subscription infrastructure, paywalls and receipt validation are real work; RevenueCat removes most of it but not all.
- iOS, Android, or web? Cross-platform in Flutter or React Native makes two apps cost meaningfully less than two native builds, but "meaningfully less" is not "half". See the cost-reduction guide for where that saving is real and where it is not.
- What single outcome does the pilot have to prove? Everything not on the path to proving that outcome is a phase-two candidate. Write the sentence down. It is the most valuable artifact in your build.
Founders who can answer those six questions get quotes that cluster tightly, because every shop is pricing the same product. Founders who cannot get a wild spread, because each shop is guessing at a different set of answers.
With those six answers in hand you can build a range for your own scope in a couple of minutes. It is not a quote — the real number comes out of the planning sprint — but it prices the same variables this section describes, and it is a useful sanity check against the three quotes already sitting in your inbox.
The timeline, honestly
Most Zee Palm packages ship in 5 to 21 days. An MVP build does not, and any studio telling you a real health-tech version one ships in three weeks is describing a prototype. Here is a realistic shape.
| Phase | Duration | Output |
|---|---|---|
| Planning sprint or Blueprint | 5–7 days | Scope, prioritized backlog, user journeys, architecture options, risks, acceptance criteria, build estimate |
| Design | 1–3 weeks | Flows and screens for the bounded version one, not a full design system |
| Build | 6–12 weeks typical for a $12k–$25k scope | Working software in your repo, in increments you can see |
| Wearable or EHR integration | 2–4 weeks per platform, often parallel | One integration, edge-cased and QA'd on real devices |
| QA and release readiness | 7 days for a [$999 dedicated cycle](/services/qa-release) | Device matrix, defect report, regression checklist, go/no-go |
| App-store review | Days to weeks, outside anyone's control | Approval, or a rejection you now have to answer |
That last row is where health-tech timelines go wrong most often. Health apps get extra scrutiny in app review, and claims in your listing copy — not just your code — can trigger it. If your marketing says the app diagnoses, treats, or monitors a condition, you have invited a much harder conversation with both the store and, potentially, a regulator.
The structural point stands regardless of which guideline number is current: your intended-use statement is a scoping decision, not a marketing decision. The Healthtech MVP Blueprint at $1,999 exists mainly to force that decision early, because deciding it late invalidates work you already paid for. It covers the intended-use boundary, a sensitive-data and claims inventory, a crisis and clinical dependency map, and the points where you need a specialist rather than an engineer. It is planning, not certification.
What we have actually shipped, and what it cost in complexity
Rather than invented numbers, here is real work and what made each one expensive or cheap.
State of New Hampshire immunization platform. HL7 messaging, 50+ schools live. Government public health work: the integration surface and the data-correctness requirements dominated everything else. Nobody in this category should assume health-data interchange is a small line item.
Previa — prior-authorization automation, delivering a 75% time reduction, integrated with payer APIs including Humana and UnitedHealthcare. The expensive part was never the UI. It was every payer behaving differently and every exception path needing a human fallback.
Patient Talker — HIPAA-conscious, end-to-end encryption, and a third-party security assessment. That assessment is the honest version of "are you secure": an outside party looked, rather than a vendor asserting it about themselves.
MediBill and Physio on the Go round out the healthcare portfolio. Across the studio: 50+ apps in production, 100+ products shipped, 100k+ downloads driven, 4.8 average rating across the maintained portfolio. Full detail on the case studies page.
The compliance boundary, stated once and clearly
Zee Palm builds HIPAA-aware, compliance-conscious systems, and maps sensitive-data, claims and clinical boundaries during planning. It does not issue certifications or legal sign-off, and does not determine whether your product is a regulated medical device.
There is no such thing as HIPAA certification. Any vendor selling you one is either confused or counting on you being confused. What exists: technical and administrative safeguards you implement, a BAA you sign with each vendor touching PHI, an independent security assessment you can commission, and — separately — a regulatory determination about device status that a qualified regulatory advisor makes, not your development agency.
Budget for the specialist. A healthcare attorney or regulatory consultant is a four-figure line item that prevents six-figure mistakes, and it is genuinely outside what any HIPAA-aware development engagement covers. If a quote you are holding does not mention this split, ask about it before you sign.
Where the money leaks after launch
The MVP price is not the total cost of ownership, and pretending otherwise is how founders end up with an abandoned app eighteen months in.
- Maintenance. OS versions ship annually and break things. Dependencies deprecate. Certificates expire. App maintenance starts at $1,000/mo and it is not optional for a product with users.
- QA on a cadence. Ongoing QA and release management from $799/mo, or a $999 seven-day readiness cycle before each significant release.
- The integration you added. Every wearable or payer API you connect is a surface that can change under you. Vendors deprecate endpoints. Budget attention, not just money.
- The clinical console you deferred. If you ran your pilot on spreadsheets, the console is a real project and it arrives right when you are busiest with customers.
A useful mental model: for a $12,000–$25,000 build, assume roughly 15–25% of the build cost per year to keep it healthy and modestly improving. That is a planning heuristic derived from the published retainer prices above, not a measured industry figure — treat it as a budgeting starting point, not a benchmark.
How to spend less with us
The unprofitable advice, because it is true.
Buy the sprint, not the build, if you are still deciding. $1,500, five days, credited toward the build. If the sprint concludes your idea needs a different shape, you have spent $1,500 instead of $40,000. Take the backlog to any agency you like.
Buy the prototype, not the MVP, if you are pre-raise. $3,500 on synthetic data. Investors respond to a thing they can tap through. So do the first ten users whose reaction determines whether the MVP is worth building.
Do not buy six wearable integrations. Buy one. Look at what your actual users wear, then add a second only when support tickets ask for it.
Do not build the admin console in version one if fewer than twenty people will use it. Run it manually. Learn what it needs. Build it in version two with a specification instead of a guess.
Use off-the-shelf where it fits. Subscription infrastructure, analytics, transactional email, scheduling — these are solved. Custom-building any of them in an MVP is a decision you should have to justify, not a default.
If you already have an app that isn't working, do not rebuild by reflex. The App Rescue package is $2,500 over ten days: a full technical assessment plus a fix plan and a quote. Frequently the foundation is sound and the failure is three specific things. Paying for the repair rather than a rewrite has saved founders more money than any other item on the price list.
What to do next
If you have a defined product and a budget, tell us the scope and we will tell you the number — including if the number is "you do not need us yet."
If you do not have a defined product, start with the MVP Planning Sprint at $1,500. If your product touches clinical data, crisis paths, or health claims, start with the Healthtech MVP Blueprint at $1,999 instead, because the extra $499 buys the intended-use and claims work that determines everything downstream.
If you are a funded health-tech team weighing agencies, the health-tech startups page lays out how the engagement runs, and /trust covers security posture, BAAs and what we will and will not sign.
Frequently asked questions
How much does a health-tech MVP cost?
At Zee Palm, an MVP build starts at $12,000 and a full custom build at $25,000, with planning at $1,500–$1,999 credited toward the build. Most funded health-tech products with a clinician-facing surface land in the $25,000–$100,000 band. Enterprise development shops commonly quote six figures for the same brief; the difference is usually scope, process overhead, and team size rather than output quality.
Why do agency quotes for the same app vary so much?
Because "the same app" is not the same app until scope is written down. The three variables that move a quote most are whether version one stores protected health information, whether a clinician-facing console is included, and how many external systems must integrate on day one. Answer those three and competing quotes converge quickly.
How long does a health-tech MVP take to build?
Plan on 5–7 days of planning, 1–3 weeks of design, and 6–12 weeks of build for a $12,000–$25,000 scope, plus 2–4 weeks per wearable or EHR integration, which often runs in parallel. Add a QA cycle and app-store review, which is outside anyone's control and slower for health apps.
Can I build a HIPAA-compliant app for under $25,000?
You can build a HIPAA-aware system in that band if the scope is genuinely bounded — a limited data set, one integration, no clinical console. What you cannot buy at any price is a certification, because HIPAA certification does not exist. You will also need a healthcare attorney or regulatory advisor, which sits outside the development budget.
Should I build a prototype or go straight to an MVP?
If you have not validated the core loop with real users, or you are raising on the strength of the idea, build the $3,500 prototype on synthetic data first. If you have paying pilot customers waiting and a validated flow, go to the MVP. The prototype is the cheaper mistake by an order of magnitude.
What ongoing costs should I budget after launch?
Maintenance from $1,000/mo, optional ongoing QA from $799/mo, plus platform fees, infrastructure and any third-party API licensing your product depends on. As a planning heuristic, set aside roughly 15–25% of the build cost per year to keep a live product healthy — then adjust once you can see your real infrastructure and support load.
Do you sign a BAA?
Discuss it with us directly on /contact; what we will and will not sign, and the security posture behind it, is set out on /trust. What we are clear about up front: we build to the compliance boundary your privacy and regulatory specialists define, and we do not substitute for them.