Search for what Epic costs and you will find a page quoting $1,200 for a small clinic, and a little further down, a range topping out at $30 million. Another site lists a median annual cost of $420.
None of these are useful, and the $420 one is not even plausible. But they keep getting published, because there is real demand for an answer and no official source to check against.
So here is the honest version: Epic does not publish pricing. Every number you find online is somebody's estimate, reconstruction, or guess. What follows is the part that can actually be verified, and the part you can reason about.
Must Read: EMR vs EHR: The Two Acronyms Quietly Confusing Your Healthcare Build
Why there is no price list
Epic sells to organisations, not off a shelf. Every deal is negotiated, scoped to the organisation's size, module mix, hosting arrangement and how much of Epic's own staff time the implementation consumes.
Contracts also tend to carry confidentiality terms, so the people who know exact figures mostly cannot publish them. The numbers that do reach daylight usually arrive by one of three routes: a public health system's board documents, a nonprofit's financial filings, or a journalist covering a cost overrun.
That last route is why the public record skews toward the enormous and the troubled. It is still the most reliable thing available.
What large health systems have actually reported
Becker's maintains a ranking of the most expensive EHR projects on record. The Epic entries:
| Health system | Reported cost | | --- | --- | | Kaiser Permanente | $4 billion | | Mayo Clinic | $1.5 billion | | Mass General Brigham | $1.2 billion | | Northwell Health | $1.2 billion | | NYC Health + Hospitals | $1 billion | | Sutter Health | $1 billion | | NewYork-Presbyterian | $964 million | | Providence | $800 million | | Trinity Health | $800 million | | Duke University Health System | $700 million | | UMass Memorial Health | $700 million | | AdventHealth | $660 million | | Memorial Hermann Health System | $500 million |
Two things worth drawing out of that table.
These are not licence fees. A billion-dollar figure covers software, hardware, Epic's implementation staff, your own project team for several years, training thousands of clinicians, and the productivity hit while everybody learns the new system. The licence is a minority of it.
And these are the giants. Kaiser's $4 billion was completed back in 2010. If you run a 200-bed community hospital, none of these numbers scale down in a way that tells you anything.
What is actually in the bill
Whatever the size, the cost lands in roughly six places. The ratios shift, the categories do not.
Software licensing. What people mean when they ask the price. Usually the smallest line in a large implementation.
Implementation services. Epic staff, your integration partners, configuration, data migration from whatever you are leaving. Typically the largest single line.
Hardware and infrastructure. Servers, network, devices at the bedside, or hosting fees if you are not running it yourself.
Training. Every clinician, every administrator, before go-live and again at every major upgrade.
Lost productivity. The line nobody budgets and everybody pays. Clinicians see fewer patients during go-live and for months afterward. In a large system this alone can run into tens of millions, and it is real money even though no invoice arrives for it.
Ongoing. Annual maintenance, upgrades, the internal Epic team you will now permanently employ. An Epic shop does not go back to having no EHR staff.
The small practice question
This is where published numbers get wildest, and there is a structural reason.
Epic's business has historically been built around large health systems. A ten-provider practice is not its natural customer, so figures that circulate for small practices tend to be extrapolations rather than quotes. One aggregator lists $1,200 one-time for a small clinic; the same page says first-year totals including implementation can reach $150,000 to $600,000. Both numbers are presented as Epic's pricing. They cannot both describe the same thing.
What a small practice should take from this: if you are under roughly 25 providers and nobody at Epic is returning your calls, that is the market working as intended, not a failure of your procurement process. athenahealth, eClinicalWorks, NextGen, DrChrono and Praxis all compete seriously at that size and will actually quote you.
The exception is joining a larger organisation's instance. If a health system in your area runs Epic and will extend it to affiliated practices, that is often the only realistic path to Epic at small scale, and the economics are entirely different from buying your own.
Epic's competitors, briefly
People searching for Epic's price are often really asking whether Epic is the right choice.
Oracle Health, formerly Cerner, is the main large-system alternative. Meditech is common in community hospitals and generally less expensive. athenahealth is cloud-based and strong in ambulatory. eClinicalWorks and NextGen compete across small and mid-size practices.
A caution on comparisons: switching EHR is one of the most disruptive things a health system can do, and the cost of the switch regularly exceeds the licence difference by a wide margin. The reported figures in the table above are mostly organisations moving to Epic, and much of that spend is the moving, not the destination.
The cost line that gets missed
Here is the one we see go wrong most often, because it is the part we get called about.
Budgets are built around licensing, implementation and training. Then, eight months in, someone asks how the lab system, the imaging platform, the patient engagement app and the three analytics tools are going to talk to Epic. Each of those is an interface. Each interface is its own scoping, mapping, testing and sign-off.
If you are a vendor rather than a hospital, the same gap shows up differently: you budget to build your product and forget that getting it connected to each customer's Epic instance is a project per customer. Our EHR integration services page sets out what that work involves.
One naming point, since most articles on this still have it wrong. Epic's developer programme used to be App Orchard. That name is retired. It is now Showroom, with tiers that differ by what you are building and whether a customer sponsors you. If a guide you are reading still says App Orchard, it has not been updated in over two years, which should tell you how carefully to treat its cost figures.
How to actually get a number
Ask Epic. That sounds glib, but it is the only route to a real figure, and the conversation is more productive if you arrive with your bed count, provider count, module list, hosting preference and current systems already settled.
Ask peers your size. Health system CIOs talk to each other, and a similar organisation that implemented in the last two years is worth more than any published estimate.
Budget the invisible lines. Lost productivity and the permanent internal team are what turn a planned number into an actual one.
And treat every figure you find online, including the table above, as a reference point rather than a quote. The large-system numbers are real because they were reported by the organisations themselves. Everything below that tier is reconstruction.
The short version
Epic does not publish prices, so there is no correct number to find. Reported implementations at large health systems run from roughly $500 million to $4 billion, and most of that is implementation, training and lost productivity rather than software. Small-practice figures online are unreliable because small practices are not Epic's core market.
If you are budgeting, the categories matter more than any headline figure, and the two that get missed are lost clinical productivity during go-live and the integration work to connect everything else you run.
Planning an Epic integration, or trying to work out what connecting to one will cost you? Tell us what you are working with and we will give you a straight read. No pitch.
Must Read: EHR Integration: What Founders Need to Know Before You Build
Sources: Becker's Hospital Review, 17 most expensive EHR projects of all time · Becker's, Partners HealthCare Epic implementation · Healthcare Dive, Partners HealthCare earnings after Epic implementation
Blogs:
Explore all blog
HL7 Integration Services: What You're Actually Paying For
A CTO told us his platform was FHIR-native, so HL7 wasn't his problem. Two weeks later the hospital sent over a spec for an ADT feed over MLLP. Here's what HL7 integration work actually involves, and what makes one interface cost three times another.
Clinical Data Integration: Why the Single Patient View Is So Hard to Build
A team demoed a beautiful unified patient dashboard. A nurse in the room asked one question: is this medication list current? Nobody could answer it. That question is the whole problem with clinical data integration.
EMR vs EHR: The Two Acronyms Quietly Confusing Your Healthcare Build
A founder once lowered his voice on a call and asked, like he was confessing something: "What's actually the difference between an EMR and an EHR?" Six months into building a healthtech product. Here's the answer, and why it changes your build.