What you need

  • Your postcode. Used only to find the weather for your spot.
  • The system size in kWp. On the MCS certificate, the handover pack or the quote; often a number like 3.6 or 4.2.
  • Roughly which way the panels face and how steep the roof is. A guess is fine.
  • The year it was installed.
  • How many kWh it generated over the last twelve months. Two generation-meter readings a year apart, or your monitoring app's last-twelve-months total. Monthly figures, if you have them, show where any gap sits.

What it cannot see. Shading from trees and chimneys, a dirty array, a string that is wired but dead, or a meter that reads export rather than generation. That is why the result is a range and a verdict, not a diagnosis. If it says "below", the Health Check is the step that finds the cause.

Privacy. No name, no email. The postcode locates the weather and is not kept by the site beyond the day's spam limit. The system details and the resulting location are stored as a run in my engineering software, with nothing that identifies you attached. Privacy notice.

Your system

Two roofs? Enter the bigger one, or ask me.
From two meter readings a year apart, or the app's twelve-month total.
I have monthly figures too (optional)

kWh generated in each of the last twelve months. Leave blank any you do not have.

How it works

Where the expectation comes from

  1. Your postcode becomes a location and the location becomes a typical weather year: hourly sunshine and temperature from PVGIS, the European Commission's solar database, built from satellite records.
  2. A model of a system like yours is run through that year, hour by hour, with the losses every real system has: temperature, wiring, the inverter, dirt, and the horizon around you.
  3. Two numbers come out. The P50 is the central expectation, the figure a typical year should give. The P90 is the lower end of normal: nine years in ten should beat it. Then I take off half a per cent for every year since installation, which is what good panels lose with age.
  4. Your figure is placed against them. Above the P90 is normal; below it is a shortfall that weather does not explain, and the gap in kWh and pounds is the start of a conversation about why.

The engine is ApexYield, the simulation software I use for solar farms and commercial roofs, run on a residential setting. It is the same arithmetic a lender's engineer uses, pointed at a house.