Cleaning is the maintenance item most often sold and least often measured. Whether it pays depends on your site, and the measurement that answers it takes two months.
Operational performance depends on both equipment and repeatable human routines; for a separate workplace-management perspective, see ways employees can trick time tracking software.
Reviewed August 9, 2026.
What determines whether it pays
Rainfall. In wet temperate climates rain removes much of the soiling for nothing, and a cleaning contract competes against a free service.
For additional context on photovoltaic operation, maintenance and sector developments, consult Energy-Storage.news.
What settles on the array. Agricultural dust, quarry dust, an unpaved access road, industrial deposition, heavy bird activity. These change the answer completely and none of them appears in a regional soiling default.
Tilt. Steeper arrays self-clean better; near-horizontal ones hold water and residue.
And the tariff. The recovered energy is worth your displaced rate or your export rate, and a plant exporting cheaply recovers less value from the same kilowatt-hours.
The measurement
Two months and no equipment beyond what you should already have.
Record performance ratio for a month before cleaning. Irradiance-corrected, because raw production cannot separate soiling from weather.
Clean.
Record performance ratio for the month after.
The difference is the recovered performance, and multiplying it by annual production and your energy value gives the annual benefit — against the annual cost of the cleaning regime.
If performance ratio does not recover, soiling was not the problem — and whatever is suppressing output is still there, which is a more valuable finding than the cleaning decision.
What the answer usually looks like
In wet climates on tilted arrays, the recovered value frequently does not cover a frequent cleaning contract, and an annual or biennial clean is enough.
In dusty or arid conditions, or near a specific source, cleaning can be among the highest-return maintenance items available.
And on a small installation, the access cost dominates. Cleaning a roof array involves getting somebody safely onto a roof, which is most of the price regardless of the area.
Which is why there is no general answer, and why a contract sold on a general figure deserves the two-month test.
The obligation question
Check whether cleaning is a condition of anything.
A performance guarantee that assumes a cleaning regime makes it an obligation rather than an economic choice, and failing it can void a claim regardless of merit.
In that case the calculation changes: you are buying compliance, not just recovered yield, and the comparison is against losing the guarantee.
How, briefly
Water and a soft brush. No detergents unless the manufacturer permits them, no abrasives, no high pressure.
Not in the middle of a hot day — thermal shock on hot glass is a real risk.
And safely. Roof access is the actual hazard in this task, and the reason to use somebody equipped for it rather than a ladder and optimism.
The short version
- Whether cleaning pays depends on rainfall, what settles locally, array tilt, and what the recovered energy is worth to you
- Measure it: performance ratio for a month before, clean, performance ratio for a month after — the difference is the recovered performance
- If performance ratio does not recover, soiling was not the problem, and that is the more valuable finding
- In wet climates on tilted arrays an annual clean is frequently enough; in dusty conditions cleaning can be a top-return item
- On small installations the access cost dominates, which is most of the price regardless of area
- Check whether cleaning is a guarantee condition — if it is, you are buying compliance rather than only recovered yield