Checking a $200,000 Hospital Robot Value Claim Line by Line

What does the $200,000 hospital robot figure actually say?

Serve Robotics says each hospital facility deploying its Moxi robots is expected to generate $200,000 to $400,000 in annual sales. That is a company expectation. It is not a measured result, and no public source we checked shows that every site reaches it.

The range is a forecast of vendor revenue, and that distinction matters more than the headline. A facility team that treats it as the hospital's annual value will build the wrong model. The claim becomes testable only after you translate it into hours, avoided costs and total payments that you can check against your own records.

The range first appears in the company's January 2026 announcement of its acquisition of Diligent Robotics, and a later PYMNTS article repeats it. Cite the January original rather than a later repost.

Where did the number come from, and who is saying it?

The range comes from a press release announcing the Diligent acquisition, and the same figures appear in Diligent's version of that announcement. The release reports more than 1.25 million deliveries by nearly 100 robots in more than 25 hospital facilities. These are vendor-reported figures.

Deal announcements serve investors as well as customers. The release also lists acquisition consideration of $29.0 million in stock, plus up to $5.3 million in earn-out, with closing expected in the first quarter of 2026, subject to customary conditions. Robotics and Automation News says the deal gives Serve access to healthcare use cases with higher revenue per robot than last-mile delivery. Read the $200,000 figure with that incentive in mind.

Why is vendor sales not the same as hospital value?

Sales to a vendor are a cost to the hospital if the hospital is the one paying. Value to the hospital is what remains after those payments are subtracted from the labor and delay the robot avoids. A site can generate high vendor revenue and still show thin net value, or the reverse.

Press coverage describes a robot-as-a-service model, in which hospitals pay a recurring fee. PYMNTS reports that NEURA leases its bed robot for a fixed monthly fee and that robot makers are renting capacity rather than selling machines. The payment stream, not the robot's list price, drives the hospital's math.

Which costs belong on the hospital's side of the test?

Include every cost the hospital carries, not only the vendor's fee. Omitted items are the most common reason a pilot looks better in a slide deck than in the operating budget.

• Recurring vendor or lease payments for the full term, plus any setup or integration charges.
• Site mapping, elevator and door integration, and network work the hospital must provide.
• Staff time for training, escorting the robot through exits or elevators, and clearing stuck or failed runs.
• Maintenance, repair dispatch, and backup units during downtime.
• Software, connectivity, and changes to nurse, pharmacy or environmental services workflows.
• Charging space, storage, and any corridor changes needed for safe routes.

Which runner and courier data should you pull first?

Pull at least three months of records for each transport type, split by urgency. Time-stamped runner logs, courier dispatch records, and pharmacy or lab request times show what the robot would actually replace. Staff estimates of how far they walk are weaker evidence.

Count runs, not vague volume. Record minutes per run from request to handoff, including waits for elevators and doors. The useful number is the staff hours freed from transport, after you subtract the time spent supervising and reassigning the robot.

Which assumptions should you challenge first?

Vendor models usually rest on a few assumptions that deserve a direct question before any numbers go into your budget.

• Utilization: a robot that can run around the clock is not the same as one that clears your real request volume.
• Time saved per run is not staff time freed unless someone reassigns those hours.
• Exceptions such as blocked corridors or elevator waits can erase savings on your slowest routes.
• Uptime claims need a fleet model that includes spare coverage while a unit is being repaired.
• Deliveries and hours must be defined the same way on both sides. Ask the vendor how each is counted.

What do outside and hospital-reported records show?

Proof News says it contacted 14 hospitals that had deployed Moxi, and five confirmed they were still running the program. That was the reporter's outreach, not a survey, so it shows the limits of public evidence rather than a success rate. A MultiCare spokesperson said the robot was brought in to 'support our nurses,' which is a staffing framing rather than a savings claim.

A hospital's chief innovation officer reported in an August news release more than 40,000 deliveries and more than 16,000 hours of work, according to PYMNTS. The hospital is not named in the excerpt we found, so treat it as a single data point. Proof News also cites a Binghamton University researcher who reviewed Reddit posts, where all 19 users describing direct experience felt negatively. That sample is small and informal, but it is one more reason to collect your own evidence before accepting either side of the debate.

How do you turn the claim into a go or no-go decision?

Build a simple annual model with three lines: freed staff hours multiplied by the loaded cost of that time, plus avoided costs such as overtime or agency coverage, minus every cost on the hospital side. Run it at your measured volume, not the vendor's average, and then run it again at the low end of your data.

Set the threshold before the pilot starts. If the result stays positive only when the robot runs at full volume, the claim is not yet proven for your site.

Service Robot Co. is a full-service commercial robot integrator that is OEM-neutral, so we select the robot for the route rather than starting from one brand. We lease robots only, on a three-year lease with one fixed payment quoted after a site assessment. We also finance, deploy, integrate, train, and service each unit through a nationwide US engineer network. A fixed payment is easier to test than a fee tied to sales, because you can compare it directly with the hours your own records show the robot can free.