How much do solar panels make in different parts of the UK?

How much does where you live change what solar panels make?

By over a third across mainland Britain, and by more than half if you count Shetland — on paper. The paper is the MCS Irradiance Datasets workbook, the table every MCS-certified installer must use to estimate a system’s output under MIS 3002: one sheet for each of 21 postcode zones (four of them split in two, so 25 sheets), giving the kWh each kWp makes in a year for every pitch and orientation. Read directly from it (v2.0) for an unshaded south-facing roof at 35°, and multiplied by four for a 4 kWp system — ten 400 W panels, the size this guide uses throughout — the whole country ranks like this:

Your zone is decided by the letters at the start of your postcode, not by the nearest town in the list: MIS 3002 carries the full lookup on page 27 (NE and SR postcodes are zone 9E, TS and DL are zone 10, PL and TQ are zone 4, BS is 5E, M is 7E, G is 14, BT is 21 — and EH, Edinburgh’s own postcodes, is zone 15, the sheet the workbook names for Dundee, while the “Edinburgh” sheet 9S covers TD postcodes in the Borders) and notes that “these zones are the same as the SAP postcode zones” used on an EPC. The 0.8 factor built into the tables — MIS 3002 says the data “is drawn from the Climate-SAF-PVGIS dataset and multiplied by 0.8” — means these are AC figures, what would reach your sockets, not panel-label numbers.

South west against north east: what is the actual gap?

On the tables, about a sixth; on the meters, almost nothing last year. Take the two zones the question usually means: a south-facing 4 kWp system at 35° makes 4,364 kWh a year in the Plymouth zone and 3,884 in the Bristol zone, against 3,612 in the Newcastle zone and 3,656 in the Middlesbrough zone — so Newcastle is 83% of Plymouth and 93% of Bristol, and “the south west” is itself two different answers depending on which end of it you mean. Then look at what installed systems actually produced. DESNZ’s Feed-in Tariff load factor analysis, published 18 December 2025, reads the meters of the whole FiT fleet and reports that “South West England had the highest load factor, closely followed by the East of England and North East, all at around 9.5 per cent” in 2024/25. The workbook behind it gives the medians as 9.62% for the South West (34,082 systems) and 9.43% for the North East (15,262) — a 2% gap. Averaged over the five years to 2024/25 it is 10.50% against 10.01%, a 5% gap. The north east is not the far end of the table; on the meters it is the third-best region in the country.

Is it about how far north you are?

Not mainly. Read the ranking again: the Dundee zone (938 kWh per kWp) sits above Birmingham (935), Sheffield (892) and Manchester (865); the Borders zone 9S (902) is level with Newcastle (903); Aberystwyth on the Welsh coast (922) beats Sheffield and Manchester; and the four lowest mainland zones — Glasgow, Inverness, Manchester and Dumfries — are western or Highland, not simply northern. The workbook does not say why; it carries the numbers and the numbers come from the European Commission’s PVGIS climate model. But the pattern in them is plain: the east coast of Scotland outperforms the Pennine side of England, and latitude on its own is a poor guide to what a roof will make.

What did real systems make, region by region?

Here is the metered ranking for 2024/25, from the DESNZ workbook, with the median load factor turned into kWh for each kW installed (a load factor is output as a share of running flat out for all 8,760 hours in a year, so 9.62% is about 843 kWh per kW):

Two cautions from DESNZ’s own text. 2024/25 was a poor year everywhere — the national median of 9.2% was “the lowest recorded median in the time series”, with sun hours “at their lowest since 2012/13” — so the five-year averages run higher (South West 10.50%, North East 10.01%, Scotland 9.14%, London 9.13%). And London’s last place is not about its sky: the analysis says it “typically has one of the lowest regional load factors; this may be due to pollution particles settling on the panels, or because panels are shaded by tall buildings nearby”. Every region’s median fell against the previous year “except for the North East, which increased slightly”.

Why is the metered gap so much smaller than the table gap?

Because the tables compare one ideal roof moved around the country, and the meters compare every real roof where it stands. The MCS figures above are like-for-like: the same south-facing, unshaded, 35° array in each zone, which is the right comparison for your decision. The FiT fleet is 282,047 systems facing every direction, on every pitch, behind every tree and chimney, installed at every point since 2010 — and that mix is broadly the same in every region, so the differences between regions shrink towards the underlying climate gap and the differences within a region swamp it. The DESNZ text says as much: “the median load factor varies across regions, but the distributions are similar from region to region.” The ranking still holds at the ends — the South West is top on both the tables and the meters, and Scotland and the North West are low on both — but the size of the gap depends entirely on which question you are asking.

Does the roof matter more than the region?

Yes, and by more than the map does. Turn the Brighton-zone system to face east or west and it makes 3,532 kWh — only a couple of percent more than a south-facing system in the Manchester zone makes (3,460), so the best zone in the country with the wrong face is level with one of the worst zones with the right one. Face it north and the effect is larger again: a north-facing roof at 35° makes 51–59% of a south-facing one in every zone, so a north-facing 4 kWp system in London (about 2,160 kWh) makes less than a south-facing one in Lerwick (2,944). The move from Brighton to Glasgow costs a quarter of the output; the move from south-facing to north-facing costs nearly half of it wherever you are. Does a north-facing roof work at all? goes through the orientation and pitch tables in full.

What counts as a standard system on a semi?

Something between 4 and 4.5 kWp, and the exact figure is a fact about your roof. The Energy Saving Trust’s solar advice, last updated 27 August 2026, puts it as “a 4.5kWp system typically covers between 20 to 30m² of roof surface area, typically using around 12 panels”. This guide uses 4 kWp — ten panels at 400 W, the panel size Google’s roof model assumes — because it is what the tables multiply by most easily; for any other size, multiply your zone’s kWh per kWp by the kWp on your quote. The number of panels that actually fit on a semi depends on the length of the roof, the chimney and the roof windows, and a gabled semi has one long south-ish face or none, which is why the count on the roof card is measured from the imagery rather than assumed from the house type. If it helps to see the arithmetic worked from the label on the panels through to a year’s kWh, how do installers turn kWp into kWh? does it step by step.

What about Scotland and Northern Ireland?

Scotland’s central belt and east coast sit with the north of England, not below it. From the table above, a south-facing 4 kWp system at 35° makes 3,752 kWh a year in zone 15 — the Dundee sheet, which is also Edinburgh’s and Fife’s zone — 3,608 in the Borders zone, 3,488 in Aberdeen, 3,332 in Glasgow and Inverness, 3,164 in Stornoway, 3,028 in Kirkwall and 2,944 in Lerwick — and Belfast’s 3,380 sits between Manchester and Glasgow. The Lerwick figure, the lowest in the workbook, is still 65% of Brighton’s; there is no zone in which the same panels make nothing worth having. On the meters, Scotland’s median was 8.50% in 2024/25 (about 745 kWh per kW installed) and averaged 9.14% over the five years to then — the lowest of the nations and a whisker above London, from a fleet of 18,119 metered systems.

How does this site allow for where you live?

Automatically, because the model it uses carries the local sky. The roof card is built on Google’s Solar API, whose methodology states the estimate takes into account “historical cloud and temperature patterns” for the location as well as the shadows “cast by nearby structures and trees”, so the figure for a Newcastle roof already has Newcastle’s cloud in it and you do not need to apply a zone factor yourself. It will not match the MCS table exactly, and should not: Google models your actual roof face by face rather than an ideal 35° south-facing one, and reduces its DC estimate to AC with a documented 0.85 derate where the MCS tables carry a 0.8 factor. The workings panel names both the estimate and the derate. What neither model can see is a tree that has grown since the photograph or a chimney’s afternoon shadow on a particular panel — that is the surveyor’s, and MIS 3002 says so.

What should I take from this if I am in the north east?

That the postcode is the smallest of the three things that decide your number. On the tables, a south-facing 4 kWp system in the Newcastle zone makes about 17% less than the same system in Plymouth and about 7% less than in Bristol; on the meters, the region came within 2% of the South West last year and within 5% over five years. Set that against the roof: facing east or west instead of south costs about 20% in every zone, facing north costs 41–49%, and heavy shading can cost more than either. So the useful question for a north-east roof is not “is it far enough south?” but “which way does it face, how steep is it, and what is in front of it?” — which are the three things the roof card measures before it shows you a figure.