Check the whole service cost

A robot completing a floor route does not, by itself, prove that a cleaning service costs less. The route may remove time from one task while creating other work: preparing the space, mapping, charging, emptying, checking, recovering the machine or completing the detail work it cannot do. The first question is therefore not who gets the reward. It is whether there is a reward after the whole service has been considered.

The University of Warwick’s four-cobot trial gives a useful example of the distinction. The university says the machines are intended to take on routine vacuuming and floor scrubbing so its team can focus on detailed work. It also says it will gather data over time to understand the impact. That is a stated aim and an open evaluation, not proof that the trial has already created a cash saving or a particular outcome for jobs.

Set a starting point for the trial

A fair trial should set a starting point. Record the agreed areas, cleaning frequency, paid hours, interruptions, human support time and the cleaning result expected. During the trial, record the robot’s completed work and the time spent by people around it. Compare the total service against the same agreed standard. A forecast from a supplier may be useful for planning, but it should remain a forecast until the site has checked it.

The British Cleaning Council’s 2026 research announcement places this discussion in a large and varied sector: 1.51 million people work across cleaning, hygiene and waste, and 83% of businesses are micro businesses. Its figures are not a measure of robotics savings. They do show why a single claim about efficiency cannot stand for every contract, workplace or employer.

Make the choice visible

If a genuine net benefit is established, several uses may be reasonable. A client may seek lower costs or better value. A contractor may need to recover investment and fund maintenance and support. A cleaning team may seek better pay, less pressure, more secure hours, training or more time for detailed work. The agreed service may also need more resource in areas where standards matter most.

None of those outcomes is decided by the machine. They are commercial and workforce choices. The people whose work and pay may change should see the assumptions, including what happens if the expected benefit does not appear. A proposal that names the measure, the review date and the planned destination of any released time gives everyone something real to test.

Practical takeaway

Before describing a robot as a saving, put a shared trial record in place. Include the whole service cost, human support, the agreed cleaning result and the proposed use of any benefit. Then review it with the client, contractor and cleaning team. That does not settle who should receive the largest share. It makes the decision honest enough to discuss.

Sources

  1. Campus Cleaning Services welcomes four new ‘cobots’

    University of Warwick Estates · 26 March 2026

    First-party account of a four-machine floor-care trial. It states intended benefits and planned impact data collection; it does not report proven savings, cleaning outcomes or job effects.

  2. Research shows steady growth in the cleaning, hygiene and waste industry

    British Cleaning Council · 17 February 2026

    Public summary of the BCC’s annual report. Workforce and business figures cover the combined cleaning, hygiene and waste sector and do not measure robotic savings.

Contribute to the inquiry

If you have run a cleaning-robot trial, what did you include before calling it a saving: setup, supervision, maintenance, service quality, paid hours or something else? Publicly shareable examples can help test this question.

Join the discussion on the HI company page (opens in a new tab)