Cleaning a School Building: Coverage, Absenteeism, and Where Automation Fits

The Gap Parents Are Describing

A back-to-school survey published by the American Cleaning Institute in August 2026 found that just 39% of parents rate their school’s cleanliness as excellent. Asked where they worry, half named restrooms. Cafeterias followed at 34%, classrooms at 25%, gyms and locker rooms at 20%, common areas at 16% and buses at 15%.

The ordering is worth sitting with, because it is not the ordering a facilities budget usually reflects. Restrooms are the top concern by a wide margin and they are also the least automatable space in the building. Corridors, cafeterias and gym floors — the large open surfaces where a machine is genuinely good — sit in the middle and lower half of the list.

Anyone selling automation into K-12 should say that plainly rather than let a survey about restrooms be used as a brochure for floor scrubbers. The honest argument is indirect, and it is set out below.

The Arithmetic of a Night Shift

The gap between what parents expect and what a custodial team delivers is rarely a diligence problem. It is a coverage problem, and the numbers behind it moved in the wrong direction this year.

Across commercial cleaning generally, labor runs at roughly 65% of total cleaning cost, at fully loaded rates commonly cited between $40 and $55 an hour, with industry turnover figures reported anywhere from 100% to 400% a year. Districts do not pay commercial rates, but they compete in the same labor market for the same people, and they lose that competition more often than they win it.

Then there is the July 2026 jobs report, which showed local government education shedding roughly 50,000 jobs in a single month — the largest single category of loss in a month when the economy overall shed 23,000. Custodial and facilities roles are not broken out of that figure, but they are rarely the last positions a district protects when a budget tightens.

So the practical situation in a great many buildings is a crew smaller than it was, covering the same square footage, on the same schedule, being graded by parents on the one space that cannot be automated. That is the actual problem. A machine is one partial answer to it.

Absenteeism Is the Number With a Budget Attached

If you are building a case for a superintendent or a board, cleanliness perception is not the strongest argument available to you. Attendance is.

Research published in the American Journal of Infection Control and widely cited in this year’s coverage found that a consistent cleaning routine produced a 50% reduction in absenteeism. Note the operative word: consistent. The finding is about routine reliably executed, not about any particular product or machine.

Attendance carries direct funding consequences in most states, whether through average daily attendance formulas or through chronic-absenteeism accountability measures. That makes it the rare facilities argument that translates into a number a finance office already tracks. Before you present it, pull your own district’s attendance data and your own state’s funding formula, and build the estimate from those rather than from the study — a national research finding is a reason to investigate, not a projection you can defend at a board meeting.

The link back to automation is consistency. A night crew that is short two people does the corridors when it can. A machine that runs the corridors on a schedule does them the same way every night, which is precisely the condition the research describes.

Where a Machine Fits, and Where It Does Not

An honest allocation for a typical school building:

Good fit. Long corridors. Cafeteria and commons floors. Gymnasium floors. Large hard-floor entrances and vestibules. Media centre and open library floors. These share the characteristics that make autonomous scrubbing work: large uninterrupted hard surface, predictable layout, low obstacle density after hours, and a task that is repetitive and disliked.

Poor fit. Restrooms — fixtures, partitions, touchpoints and the specific inspection a human eye performs. Classrooms with furniture down, which are an obstacle course rather than a floor. Stairwells. Locker rooms. Anything requiring judgment about whether a surface is actually clean rather than whether it has been passed over.

So the argument is not that a machine cleans the restroom. It is that a machine takes the corridors, the cafeteria and the gym off the crew’s list, which is where the largest block of undifferentiated floor time sits, and the person is then available for the detail work that parents are actually grading. That is a smaller claim than the brochure version. It is also the one that survives a conversation with a facilities director who has been pitched before.

One further point specific to schools. A building with students in it during the day and a machine running at night is the easy case. Districts that want daytime operation should think carefully: corridors during a passing period are the single worst environment we deploy into, and the right answer is almost always to schedule around the bell rather than to ask the machine to cope.

How Districts Actually Pay for This

Facilities equipment rarely comes out of a single line. The routes we see used:

• Operating budget, against a vacancy. The most common and the most defensible: a machine funded from a custodial position the district has been unable to fill for more than a year. This is an honest trade and it presents well, provided you are candid that it is a reallocation rather than a saving.
• Capital or bond funds, where equipment above the district’s capitalization threshold qualifies. Check the threshold — it is often lower than people assume, and it governs.
• ESSER successor and state facility grants, where they still exist in your state. These are shrinking and should not be the basis of a multi-year plan.
• Subscription or Robots-as-a-Service structures, which convert a capital request into an operating line. This is often the difference between a purchase that happens this year and one that waits for a bond cycle. It is not free money and the total cost over five years is usually higher; the value is timing and the shift of maintenance risk.

What generally does not work: Perkins V. CTE funds are tied to instructional programmes and a comprehensive local needs assessment, and a floor scrubber cleaning a corridor is a facilities expense rather than an instructional one. The exception is genuine and narrow — if the machine is part of a facilities-technology or building-trades pathway that students learn on, that is a different conversation, and a legitimate one. Do not stretch it.

Running a Pilot Inside a School Year

The academic calendar constrains everything, and the districts that do this well work with it rather than against it.

Summer. Deploy, map and train while the building is empty. Mapping a corridor is straightforward with no students in it and genuinely difficult during term. If you deploy in October you will map in October.

First six weeks of term. Run and observe. Expect furniture to appear in places the map does not know about, expect a door to be propped that was closed during mapping, and expect at least one enthusiastic attempt by students to interfere with the machine. All of these are normal and all are fixable in the first month.

Winter. Review against whatever you said you would measure. Completed scheduled runs as a percentage of scheduled runs is the honest headline. Square footage per labor hour is the one facilities directors trust.

Spring. Decide on expansion in time for the next budget cycle, which in most districts means a decision by February or March, not May.

Pick one building for the pilot, and pick the one with the worst staffing rather than the best. A machine that succeeds at your easiest site has proved nothing about the sites where you need it.

The Objections You Will Hear

“Is this replacing custodial jobs?” Answer it directly, because it will be asked in a public meeting. In our experience the honest answer in K-12 is that districts are not eliminating filled positions — they are covering positions they cannot fill. Say which it is in your district. If you are in fact reducing headcount, say that too; being caught having implied otherwise costs more than the candour would have.

“What happens when it breaks?” A fair question with a specific answer: what is the response time commitment, is it on-site or remote, and what happens to the corridors in the meantime. If the answer is that the crew does them manually, then your staffing plan cannot assume the machine.

“Will students damage it?” Sometimes they will try. Machines that run overnight in a secured building are rarely a problem. Machines that run during the day in a high school need a conversation with students first, which is worth having anyway — the schools that introduce the machine to students, rather than hiding it, have markedly fewer incidents.

“We tried this before and it did not work.” Ask what specifically failed. Most of the time it is one of three things: it was never mapped properly, nobody owned it after the vendor left, or it was bought for a building whose floors are mostly carpet. All three are avoidable and none of them are about the technology.