Dirty solar panels lose between 5 and 40% of their output. That range is wide because soiling is wildly site dependent, and the honest answer for your system is somewhere specific inside it. This page is about working out where.
For a business, that percentage lands differently than it does for a homeowner. Commercial systems are sized against real consumption, so every kilowatt hour the array fails to generate is a kilowatt hour bought from the grid at commercial rates. The system keeps “working”. The shortfall just appears on the power bill, which is why soiling losses run for years on commercial sites without anyone noticing.
Where your site sits in the 5 to 40% range
The low end (5 to 10%) is a steeper roof in a clean airshed, regular rain, no bird pressure. The general dust film that builds anywhere.
The middle (10 to 20%) is the typical unserviced commercial site: low pitch roof, some local dust source, edge soiling banding along every panel, the odd bird dropping cluster.
The top end (20 to 40%) is the neglected cases we get called to. Heavy bird fouling, lichen established on the frames and creeping onto glass, sites near quarries or unsealed yards, arrays that have never been cleaned since commissioning. Losses like this are not subtle, but if nobody is comparing generation year on year, they still go unnoticed.
A worked example, purely illustrative: a 100kW system in a good NZ location generates very roughly 130,000kWh a year. A 10% soiling loss is 13,000kWh of electricity you paid to generate and bought from the grid instead. Run that over three years and the cost of never cleaning becomes very real, whatever your unit rate is.
Why droppings are worse than dust
Soiling types are not equal. An even dust film dims every cell a little. A bird dropping blocks a few cells completely, and because panels are wired in strings, a fully shaded cell can pull down the output of panels around it. A small amount of concentrated fouling can cost more than a roof full of light dust. This is also why partial DIY cleaning from an access hatch does not recover much.
How to check without climbing on the roof
Pull your monitoring data and compare the last 12 months against the previous 12, month by month. Weather explains single bad months. It does not explain steady decline. If you see drift, a clean plus inspection answers the question definitively, because the before and after inverter readings show exactly what the soiling was costing. We document those readings on every commercial clean. For larger arrays, a drone thermal inspection finds hot spots and failing panels that a visual check misses, which is worth doing at the same time as a clean.
Request a quote with your system size and location and we will price it properly. If you are working out a sensible service interval, see how often commercial solar panels should be cleaned.
Common questions
Our output looks normal. Could we still be losing money?
“Normal” compared to what? If the baseline is last quarter, soiling is invisible because it declines slowly. Compare against the same months in a previous year, or against commissioning data if you have it.
Is 40% loss realistic or scare tactics?
It is real but it is the extreme end: heavy bird fouling or years without cleaning. Most unserviced commercial sites sit in the 10 to 20% band. That is still a meaningful sum at commercial consumption.
Will cleaning restore full output?
Cleaning removes the soiling loss. If output is still down after a clean, something else is wrong (a failing panel, a string fault, inverter clipping), which is exactly what the inspection and thermal work is for.