NHS cleaning bill reaches £1.6 billion as cost per cleaner rises 5%

• £1,645mspent on cleaning in 2025/26, up 5.9%
• +0.6%cleaning staff, 42,631 to 42,884 full-time equivalents
• £45.80a square metre a year at the typical site (median of 2,055)
• 73.6%of the bill goes on 215 general acute hospitals

What we found

1.The bill rose 5.9%; the workforce barely moved

The NHS spent £1,645 million on cleaning in 2025/26, against £1,554 million in 2024/25. Cleaning staff rose from 42,631 to 42,884 whole-time equivalents (WTE), so the cost per cleaning WTE rose 5.2%, from £36,453 to £38,360. These are national totals for every trust, from the return’s own summary table; the site file adds up to the same £1,645,066,424.

2.Three-quarters of the money goes on acute hospitals

The 215 general acute hospitals in the return account for £1,210 million, 73.6% of all NHS cleaning spend, from 7% of its sites. The largest single figure is £21.7 million, at one teaching hospital.

3.The typical site spends about £46 a square metre, but the range is wide

Across 2,055 sites the median is £45.80 per square metre of gross internal floor area a year. A quarter spend under £27.20 and a quarter over £71.60; the 10th percentile is £15.10 and the 90th £101.50. Most of that spread is the kind of building: support facilities have a median of £22.00 and community hospitals with beds the most, £79.30.

4.Like for like, acute hospitals spend between £42 and £91 a square metre

Among the 215 general acute hospitals, the 10th to 90th percentile range is £42.10 to £91.30 (median £63.20): a ratio of 2.2, against 6.7 for all sites. Bigger hospitals are not cheaper to clean per square metre: there is no relationship between floor area and cost per square metre (rank correlation 0.07, 95% interval –0.06 to 0.21).

5.One full-time cleaner looks after about 600 square metres of an acute hospital

The median general acute hospital reports 607 m² of floor area per cleaning WTE (10th to 90th percentile 424 to 870 m²), and £37,580 of cleaning cost per WTE. Across all 2,055 sites the median is 782 m² per WTE.

6.In acute hospitals, in-house and outsourced cleaning cost about the same per square metre

In-house services account for £1,073 million (65.2% of spend, 1,231 site returns) and outsourced contracts £529 million (32.2%, 898 returns); hybrid and mixed arrangements make up the rest. Taken across all sites, outsourced sites look cheaper (median £40.60 per m² against £50.70), but they are smaller and fewer of them are hospitals. Within general acute hospitals the outsourced median is £65.60 and the in-house median £62.10 (62 and 149 hospitals). Split by size, the two are level in smaller and middle-sized hospitals (£61.80 and £61.80; £65.00 and £64.30) and £67.20 outsourced against £61.10 in-house in the largest third. Floor area per cleaner is the same either way: 606 and 611 m² per WTE.

Earlier studies of trust-level data from 2005 to 2014 found outsourced cleaning cheaper. This is a different measure (cost per square metre, within acute hospitals) in a different decade, so the results are not directly comparable.

7.Where a hospital is goes a long way to deciding who cleans it

Outsourcing is the norm in the East of England (75.8% of sites, 61.7% of spend) and London (71.0% of sites, 50.3% of spend), and the exception in the North East and Yorkshire (12.1% of sites) and the Midlands (25.8%). London’s acute hospitals spend the most per square metre (median £79.50, 37 hospitals), the North West’s the least (£54.60, 33 hospitals).

8.Spending more per square metre does not go with a higher cleanliness score

Across 732 NHS sites found in both this return and the 2025 Patient-Led Assessments of the Care Environment (PLACE), the rank correlation between cleaning cost per square metre and the PLACE Cleanliness score is 0.05 (95% interval –0.03 to 0.13): no relationship. The average Cleanliness score is 98.9%, 99.0%, 98.9% and 99.1% from the lowest-spending quarter of sites to the highest. Within general acute hospitals alone (203 sites) the correlation is –0.10 (–0.23 to 0.03), again no relationship.

The scores leave little room to tell sites apart: the median NHS site scores 99.6%, and a third (32.4%) score 100%. An absence of correlation is not proof that spending makes no difference: some sites may spend more because they are harder to keep clean.

What the figures do not show

• The estates return is completed by each trust. NHS England validates it, but data quality remains the submitting trust’s responsibility, and corrections are sometimes published later.
• Cleaning cost has a set definition: pay up to supervisor level and consumables, including contract staff. It leaves out toilet rolls, paper towels and hand hygiene products, cleaning done by nurses, management costs and contract management fees. Trusts apportion shared housekeeping time themselves.
• Floor area is gross internal area, which includes plant rooms and multi-storey car parks. Cost per square metre is a benchmark, not a price for cleaning a square metre.
• In-house or outsourced is each site’s main method only, and the two groups differ in type, size and region, so compare like with like.
• PLACE is a patient-led assessment of the care environment, carried out on one day by a team that includes patient assessors, at a sample of wards and areas. It is not a microbiological measure.
• Correlation is not cause.

Method

• Sources. NHS England, Estates Returns Information Collection (ERIC) 2025/26, site data and data report (version 2), published 15 October 2026; NHS England, Patient-Led Assessments of the Care Environment (PLACE) 2025, site scores, published 26 February 2026.
• The sites. The site file has 2,923 rows from 205 trusts. The per-site figures leave out 168 “Other Reportable Site” rows (each a trust’s total for its small sites, not one site), 30 unoccupied sites, 665 rows where cleaning is “Not Applicable” (416 of them leased and fully serviced by the landlord) and 5 rows with a cost of zero. That leaves 2,055 sites carrying £1,620.7 million, 98.5% of national spend. National totals and the in-house and outsourced shares of spend use every row with a cost.
• Cost per square metre is each site’s cleaning cost over its gross internal floor area. Using occupied floor area instead gives almost the same result (median £49.10). Seven sites spend under £2 per m² and none more than £400; unusual values are kept, and the medians barely move without them.
• Types of site and in-house or outsourced are the return’s own categories, as each trust reported them. Hybrid and mixed arrangements are shown but not compared.
• PLACE. 821 of PLACE 2025’s 860 NHS sites with a Cleanliness score match an ERIC site code, and 732 of those are in the study (the rest have no cleaning cost of their own). Independent sector sites are left out.
• Statistics. Percentiles by linear interpolation (as Excel’s PERCENTILE.INC). Spearman rank correlation with a two-sided permutation test (10,000 shuffles) and a bootstrap 95% interval (2,000 resamples), seeded so the analysis repeats exactly.
• Field codes. The published site file and the data definitions workbook give the cleaning fields different codes (S14_01, S14_04 and S14_25 in the file; S11_02, S11_03 and S11_01 in the workbook). We read the file by its codes and checked each column’s label. Anyone citing a field should cite its label too.

We found no published analysis of the 2025/26 cleaning fields at site level. Earlier work used trust-level ERIC data for 2005 to 2014: Toffolutti, Reeves, McKee and Stuckler (Social Science & Medicine, 2017) and Elkomy, Cookson and Jones (Public Administration Review, 2019).

Using these figures

You are welcome to quote or chart the figures. Please credit “SiteClara analysis of NHS England ERIC 2025/26” and link to this page. Questions about the method, or the full results by type of site, region and trust: help@siteclara.com.

Contains information from NHS England, licenced under the current version of the Open Government Licence. NHS England has not reviewed or endorsed this study.

This study was put together by SiteClara, which helps cleaning and facilities teams keep a record of the work done at each location.