Editorial Article
Eight Thousand Pharmacies Closed. The Machines Sold Into the Ones That Survived.
The United States has almost 8,000 fewer retail pharmacies than it did in 2018. More than 2,000 closed in 2025 alone. Rite Aid has liquidated, Walgreens is working through a multi-year closure programme, and CVS has trimmed hundreds more.
In the same year, total US prescription dispensing revenue reached a record USD 751 billion, up 10 per cent.
Those two sentences are usually presented as a paradox. They are not. They are the mechanism that now drives the pharmacy automation market, and understanding it explains why automation revenue grows 7.76 per cent a year while unit shipments grow only 4.90 per cent.
The revenue is real, and it is not in the store
Dispensing revenue growth is overwhelmingly specialty drug revenue, and only 12 per cent of total specialty prescription revenue was dispensed through retail chains and supermarkets in 2025. Retail keeps what is left: the everyday prescription, roughly 90 per cent of it generic, on a reimbursement that deflates every year.
A pharmacy in that position has one lever, and it is not price. It is volume. Fixed costs — the pharmacist, the technician, the rent, the insurance, the software — do not move with prescription count, so the incremental cost of one more script is close to nothing and the incremental revenue is the whole margin. That economics forces consolidation: fewer stores, each dispensing more, and any store that cannot get the volume closes.
So the number of places that could buy a dispensing counter falls while the number of prescriptions rises. Anyone modelling this market by counting pharmacies will get the direction right and the magnitude wrong.
What a closure actually removes from the market
Every closed pharmacy takes a table-top counter and usually a small packaging line out of the installed base and out of the replacement cycle. It does not take the prescription with it. The prescription reappears at a surviving chain store, at an independent with capital, or in a central-fill queue.
Whether the automation reappears is a separate question, and it is the one that decides the fate of the smallest equipment family in the market. Automated table-top counting and dispensing is USD 690 million in 2025 and reaches USD 1,030 million by 2033 — a 5.14 per cent CAGR, the slowest of the five families, with its share falling from 11.17 to 9.16 per cent. That is not a category in decline. It is a category whose growth is capped by the number of independent pharmacies with the capital to replace a machine, and that number is being reduced every quarter.
Where the prescription lands in a low-capital independent, the automation does not happen. Where it lands in a chain store that is now dispensing materially more scripts than it did in 2019, it happens once and it is bigger.
The receiver of the script is the buyer of the machine
This is the single most useful analytical move in this market: track where the prescription goes, not where the pharmacy was. Three destinations matter.
Surviving chain stores. Higher script volume per site justifies capital equipment that a smaller store could never amortise. The machine gets replaced on cycle, at a larger configuration.
Independents with capital. The most price-sensitive buyer in the market, and the one most exposed to reimbursement compression. This is where the counter category lives and dies.
Central fill. One robotic line, many sites, and a hub-and-spoke architecture with automated dispensing cabinets at the spoke. This is where the growth is.
Central fill is the answer the market has chosen because it is the only configuration in which the payback is legible on one page. Harris Health in Houston is building a 60,000 square foot automated central-fill facility designed to take the system from 5,000 to 15,000 scripts a day, with Capsa Healthcare as design and installation partner, and the stated rationale is a legacy operation strained by rising volumes, outdated picking robotics and an overcrowded facility. Ballad Health took the same route across twenty-one hospitals in Northeast Tennessee and Southwest Virginia and removed a step in which six to eight technicians spent about four hours a day pulling medications from central shelves for distribution to each site.
Both projects reduce the number of installation sites and increase the automation per site. That is the arithmetic behind USD 99,038 of revenue per system in 2025 rising to USD 122,842 in 2033 while units grow at two-thirds the rate of value.
Why the machine is no longer the product
There is a version of this story in which automation simply follows script volume and everything grows in proportion. That version is wrong, and the reason is visible in what the two listed vendors did in the last twelve months.
Omnicell reported fiscal 2025 revenue of USD 1,185 million, up 7 per cent, with product bookings and annual recurring revenue both above the midpoint of guidance, and launched Titan XT at the ASHP 2025 Midyear Clinical Meeting under a roadmap it calls autonomous medication management. BD reported fiscal 2025 revenue of USD 21.8 billion, reorganised into five segments from 1 October 2025 with Medication Management Solutions inside Connected Care, and launched the Incada Connected Care Platform alongside a next-generation Pyxis Pro. Its Medication Management Solutions unit reported USD 835 million in the quarter ended 31 December 2025, against USD 801 million a year earlier.
Neither launch is a hardware event. Both are attempts to make the fleet, rather than the machine, the unit of sale — and that is what earns the 2.73-point realisation premium in the market’s value growth. Hardware bill-of-materials deflates. Software attach and service contracts do not.
The fragility, stated plainly
That premium is the most fragile assumption in the forecast, and the adoption data says why. Barcode verification of doses runs at 73.6 per cent of US hospitals and autoverification at 73.4 per cent. Advanced analytics — the layer that justifies a subscription — runs at 5.7 per cent.
If buyers accept the platform story, realisation grows 2.73 per cent a year and the market reaches USD 11,240 million in 2033. If they resist paying a subscription for analytics they cannot integrate across three vendors’ equipment, realisation grows about 1.73 per cent, and holding units constant at 91,500 systems, 2033 value lands near USD 10,400 million instead. One assumption, USD 850 million.
What would prove the thesis wrong
The bear case for this market is not a technology failure. It is a sequencing failure, and it is observable.
If retail closures remove installation sites faster than central-fill construction adds throughput capacity, the two do not net out, because they are not substitutes. The bear case runs at 6.00 per cent and puts 2033 at USD 9,850 million. A rise in the annual US net closure count above 3,000, with no corresponding central-fill announcements, would be the signal.
The bull case runs at 9.31 per cent and reaches USD 12,600 million, and it does not require anything exotic either — just sterile compounding automation crossing 10 per cent hospital penetration by 2030 alongside faster software attach and earlier central-fill construction.
But the indicator worth watching is simpler than either scenario. It is the annual net closure count. A market whose buyer base is shrinking can grow comfortably for years, as long as the prescriptions keep landing somewhere that automates. The day they land somewhere that does not, the unit line breaks first.
This analysis draws on the 2026 analysis of the US pharmacy shakeout and the USD 751 billion dispensing revenue total, the 2023 ASHP national survey of 1,497 hospital pharmacy practice settings, Capsa Healthcare’s announcement of the Harris Health central-fill partnership (January 2025), Healthcare IT News’ account of the Ballad Health consolidation, and the Pheonix Research report Global Pharmacy Automation System Market, Size, Share and Forecast 2026–2033.
Frequently Asked Questions
Does the market shrink when pharmacies close?
No. Automation spend now follows throughput per surviving site and the number of central-fill nodes a system builds, not the number of stores that open. The market reaches USD 11,240 million by 2033 at 7.76 per cent while unit shipments grow at 4.90 per cent, because each installation is larger and carries more software and service than the one it replaces.
How many US pharmacies have closed, and why did revenue still rise?
The United States has almost 8,000 fewer retail pharmacies than in 2018, and more than 2,000 closed in 2025 alone. Rite Aid has liquidated, Walgreens is working through a multi-year closure programme, and CVS has trimmed hundreds more. Dispensing revenue still reached a record USD 751 billion in 2025, up 10 per cent, because growth is concentrated in specialty drugs that are largely dispensed outside retail chains and supermarkets.
Why is central fill the growth architecture in pharmacy automation?
Central fill is the only configuration in which the payback is legible on one page: one robotic dispensing and storage line serves many sites, with automated dispensing cabinets holding the last mile at the spoke. Harris Health is building a 60,000 square foot facility designed for 15,000 scripts a day against a legacy operation running 5,000. Ballad Health's twenty-one-hospital consolidation removed a picking step that consumed six to eight technician hours a day per site.
Which part of the market is most exposed to pharmacy closures?
Automated table-top counting and dispensing grows at 5.14 per cent, the slowest of the five equipment families, with its share of the market falling from 11.17 to 9.16 per cent. Every closure removes a counter from the replacement cycle, and whether it is replaced depends on whether the prescription lands somewhere with the capital to automate.
