Cloud PACS Pricing: What Imaging Centers Actually Pay

The short answer: almost no cloud PACS vendor publishes list pricing, so every specific figure circulating online is someone's estimate. What is public is what government buyers paid. Federal prime award records show PACS contracts running from about $253,000 for a single-site replacement at one Indian Health Service hospital to $11.1 million for a multi-site maintenance award across one VA network and $12.8 million for a national teleradiology PACS program. The spread is not mostly about the software. It is about how many sites, how many studies, and whether migration and support are inside the number or beside it.

Updated September 2026. Contract figures below are prime award records from USAspending.gov, retrieved September 2, 2026. They describe what federal buyers actually paid, which is a floor-and-ceiling reference rather than a quote for your practice.

How much does a cloud PACS cost?

For an outpatient imaging center, budget in the low tens of thousands of dollars a year for a small cloud-native deployment, and expect a multi-site group replacing an aging on-premise system to land in the six figures annually once support and storage are included. Those are planning brackets, not quotes. The reason no honest article can be more precise is that this category is priced on study volume, site count and module selection, and vendors deliberately quote against your numbers rather than publishing a rate card.

The public record is more useful than it sounds, because federal buyers publish what they signed. Here is the actual shape of PACS spending across the VA, the Department of Defense and Health and Human Services.

Two things in that table are worth carrying into your own budget. First, the $285,331 software-only award sits an order of magnitude below the network-wide awards, which tells you how much of a large PACS number is services, hardware and support rather than licences. Second, several of the biggest awards went to integrators such as Four Points Technology and Frontier Acquisitions rather than to the software vendor whose product was being bought. If you benchmark against a headline figure, you are often benchmarking against a systems integration contract, not against software.

What you are actually paying for, line by line

A cloud PACS quote usually bundles five things that are worth pricing separately, because they scale differently and because only some of them recur.

The software subscription. Normally per study, per radiologist, or in volume bands. Per-study pricing is the friendliest for a growing imaging center and the most dangerous for a group with seasonal peaks, since a busy quarter shows up directly on the invoice.

Storage. Priced per terabyte or per study, and frequently tiered so that older studies move to cheaper, slower storage. That tiering is where retrieval time for comparison priors comes from, so it is a clinical decision disguised as a line item.

Migration. A one-off project, commonly quoted at roughly $0.05 to $0.50 per study, with the upper end reflecting proprietary source formats that have to be decoded. For 200,000 studies that is $10,000 to $100,000, which is large enough that it belongs in the business case rather than in a change order later.

Interfaces. Each HL7 or DICOM connection to a RIS, EHR or modality is often billed individually. Count them before you sign, including the ones you will need when the next site opens.

Support. Ask what tier is included and what the escalation path is at 2am, because the federal record shows maintenance and support awards frequently exceeding the original software purchase.

Cloud PACS vs on premises: which one actually costs less?

Over a five-year window they usually land closer together than either vendor will admit, and the difference is in the shape of the spending rather than the total. On-premise front-loads a capital purchase and a hardware refresh every five to seven years, and assumes you employ someone to run it. Cloud converts that into a predictable operating expense and removes the refresh, then adds a dependency on your internet connection and on the vendor's data export terms.

The honest rule of thumb: if you do not employ dedicated IT staff, cloud almost always wins on total cost of ownership once you price the labor you are not paying for. If you have poor or single-path connectivity at any site, cloud can be the wrong answer regardless of the arithmetic, because a read that stalls is not cheaper at any price.

The costs that never appear on the quote

Three recur often enough to plan for. Data egress is the first: if you self-host PACS software in your own cloud tenancy, the infrastructure bill is yours, and imaging moves large objects, so egress and storage tiering can quietly become the largest line on it. Teams running that model should watch the underlying cloud spend continuously rather than at renewal, since the surprise usually arrives as a slow drift across several services rather than as one obvious charge. Connecting the account to something that tracks cloud and SaaS spend read-only is a cheap way to see the drift while it is still small.

The second is retrieval of archived priors. Cheaper storage tiers are slower, and a radiologist waiting on a decade-old comparison is an operational cost that never appears in a spreadsheet. Get the retrieval time for archived studies written into the contract alongside uptime.

The third is export. Ask, in writing, what format studies are returned in when you leave, whether the export includes presentation state and report objects or only pixel data, and what it costs per study. A vendor who will not answer that in a sales cycle will not answer it in a dispute.

What should an imaging center budget per study?

Work from your own volume rather than from a headline. Take annual study count, multiply by the per-study subscription rate you have been quoted, add storage for the retention period your state and your payers require, then add migration as a one-off. Compare that to what you spend today including staff time and the amortized cost of the last hardware purchase. If the vendor will not quote per study, ask them to convert their proposal into a per-study number, because that is the only unit that lets you compare two quotes built on different assumptions.

Five questions that decide what you pay

1. Is the price per study, per user, or per site? Model all three against your next three years of volume, because the cheapest structure today is often the most expensive at double the studies.

2. What does migration cost, quoted per study? Get it separately and in writing.

3. Which viewer is cleared for diagnostic reading? Most vendors sell a cleared diagnostic viewer and a lighter reference viewer for referrers, and they are licensed differently. The clearance question is worth checking against the FDA record rather than the datasheet, which is what our page on cloud based PACS systems and vendors sets out with verified K numbers.

4. What happens to the price when site five opens? Ask for the incremental site cost now, while you still have leverage.

5. What is the export format and fee? This is the single term that determines whether you have a supplier or an owner.

Is cloud PACS cheaper for a small imaging center?

Usually yes, and the reason is staffing rather than software. A single-site center running conventional PACS on its own servers needs someone to patch it, monitor backups, test restores and plan capacity, and for most centers that work is absorbed by a person whose actual job is something else. Cloud-native platforms also start lower because there is no capital purchase to amortize. The trade is that you are buying a dependency on connectivity and on contract terms, so the diligence effort moves from the server room to the agreement. Price both, then read the export clause twice.

Whichever way the archive decision goes, it is separable from how the reading gets done. Radiological.ai works alongside cloud, hosted and on-premise systems over standard DICOM and HL7 interfaces: it flags suspected findings for a second look, prioritizes the worklist and drafts the structured report into your template, and the radiologist reviews, edits and signs every study. For the wider picture of what imaging software costs across categories, see our radiology software pricing research, and for choosing the system itself rather than its price, PACS software compared by study volume.