Are solar panels worth it? The arithmetic, with the assumptions showing
Payback is generation multiplied by what a kilowatt hour is worth to you, divided by what the array cost. Every one of those three numbers is published — except the share you use at home rather than export, which is the one that decides the answer.
4 kWp, facing south
- Generates
- 3,967 kWh a year
- Used at home (35%)
- 26.3p a kWh
- Exported
- 12p a kWh
- Worth a year
- £675
- Costs to fit
- £8,000–£10,000
- Pays back in
- 12–15 years
The self-use share is an assumption, not a measurement. Everything else here is published by someone who has to stand behind it.
Pays for itself in
12–15 years
A 4 kWp south-facing array at the method's cost
Worth a year
£675
35% used at home, the rest exported
Installed cost
£8,000–£10,000
What the EPC method allows for panels — not a quote
Net over 25 years
£6,899
After the cost, with output falling half a percent a year
Generation is PVGIS for the councils it models, averaged. The import rate is Ofgem's price cap for 1 October 2026 to 31 December 2026, GB average, including VAT. The export rate is Outgoing Octopus as published on 17 September 2026. The cost is the range the middle EPC recommendation was given, to 9 September 2026: the same for every home, so a guide rather than a quote.
The answer
What an array earns, size by size
Value and cost both scale with the number of panels, so the payback column barely moves down the table. Two real effects pull it either way and neither is published, so neither is modelled here — the page says which, underneath.
| Array | Generates | Worth a year | Costs | Pays back in |
|---|---|---|---|---|
| 2 kWpAbout 5 panels | 1,984 kWh | £337 | £4,500 | 13.3 years |
| 3 kWpAbout 8 panels | 2,976 kWh | £506 | £6,750 | 13.3 years |
| 4 kWpAbout 10 panels | 3,967 kWh | £675 | £9,000 | 13.3 years |
| 5 kWpAbout 13 panels | 4,959 kWh | £844 | £11,250 | 13.3 years |
| 6 kWpAbout 15 panels | 5,951 kWh | £1,012 | £13,500 | 13.3 years |
The EPC method gives one cost range for solar panels, not a cost per kilowatt, so the cost column scales its mid point by array size. That scaling is ours; the range it scales is the method's.
Two things move the real answer in opposite directions and neither is published, so neither is in the table. Quotes come down per kilowatt as the array grows, which makes the larger rows pessimistic. And a household uses a smaller share of a larger array, because daytime demand does not grow with the roof — which makes them optimistic. The self-use section below is where to see how much that second one matters.
By roof
Which way the roof faces changes the answer, but not the verdict
A west-facing array gives up some of the year's generation against a south-facing one. It does not give up the case for fitting it, which is a distinction most quotes blur in the other direction.
What a 4 kWp array earns a year
By roof direction, at the published import and export rates.
- South12–15 years£675
- South-east13–16 years£637
- South-west13–16 years£633
- East15–19 years£536
- West15–19 years£531
The payback beside each bar uses the same installed cost for every direction, because the method gives one. A south-facing array on a simple roof is usually cheaper to fit than an east-west split across two slopes.
Why the gap is smaller than it looks
An east or west roof spreads its generation across the morning and the evening rather than concentrating it at noon. More of what it makes lands when the house is actually using electricity, so a larger share of it is worth the import rate rather than the export one — which is worth more per kilowatt hour than the kilowatt hours it lost.
Shading beats direction
None of these figures include shading, because PVGIS models a clear horizon. A chimney across half the array in the afternoon takes more off the total than the whole difference between the best and worst rows on this chart.
Self-use
The one number nobody publishes decides the whole answer
The same array, the same roof, the same rates. Only the share used at home changes, and the payback moves by 4 years.
Annual value by how much you use yourself
A 4 kWp south-facing array, at 26.3p import and 12p export.
- 20% — Out all day14–17 years£590
- 35% — Typical12–15 years£675
- 50% — Home in the day11–13 years£760
- 70% — With a battery9–11 years£874
A kilowatt hour used in the house is worth the rate you would have paid for it. A kilowatt hour exported is worth whatever your supplier offers, which is always less.
Out all day
14–17 yearsNo battery, nobody home, nothing scheduled
Typical
12–15 yearsThe assumption every other figure on this page uses
Home in the day
11–13 yearsOr a timer on the immersion and the washing
With a battery
9–11 yearsA battery big enough to hold a summer afternoon
Every other figure on this page uses 35%, which is a reasonable assumption for a household with no battery and nobody home in the day. It is an assumption, not a measurement, and it is the thing to check against your own half-hourly data before trusting any payback figure — ours or an installer's.
With a battery
A battery does not generate anything
What it does is move kilowatt hours from the export rate to the import rate. That is worth the difference between the two, multiplied by however much it manages to move — and nothing more.
The arbitrage on its own
Raising self-use from 35% to 70% on this array is worth £199 a year at today's rates. The EPC method allows £500–£5,000 for a battery — a range wide enough to cover a single unit and a whole-house system, which is why the payback it implies runs from 3–25 years. A battery that also charges on a cheap overnight rate earns more than this, which is the case the tariff pages make.
- Panels alone
- 12–15 years
- £675 a year at 35% self-use.
- Panels and a battery
- 10–17 years
- £874 a year at 70% self-use, against both costs together.
- What the battery adds
- £199
- A year. The gap between the import rate and the export rate, moved.
- The 70% is ours
- Assumption
- Nobody publishes what share of generation a domestic battery actually shifts. Your own half-hourly data is the only honest source for it.
Where you are
The geography moves the payback by a couple of years
Generation varies by region, and so does the cap's unit rate — which means the same array is worth measurably more in some places than others. Neither difference is large enough to change the decision.
A 4 kWp south-facing array, by region
Value a year at the GB average import and export rates, with the payback beside it.
- South East11–14 years£729
- East of England11–14 years£714
- South West11–14 years£705
- London11–14 years£700
- East Midlands12–15 years£676
- West Midlands12–15 years£664
- Wales12–15 years£658
- Yorkshire and The Humber12–15 years£656
- North East12–16 years£647
- North West13–16 years£620
- Northern Ireland13–16 years£610
- Scotland14–17 years£584
The unit rate itself varies across 14 cap regions, which this chart holds constant so the generation difference shows on its own.
What moves it
Six things that change the number
A payback figure is only as good as the assumptions under it. These are the ones that move it most, in roughly that order.
The export rate is not fixed
One supplier's own export products currently run from 4.1p to 15.9p a kilowatt hour across 6 tariffs. Picking the wrong one costs more than most of the differences on this page.
The import rate moves quarterly
Payback is quoted against today's cap. If electricity gets dearer the array pays back faster, and vice versa — which makes solar a hedge against the price of electricity as much as an investment in it.
Panels degrade, inverters fail
These figures assume output falls about half a percent a year, which is a normal warranty term. They do not include replacing the inverter, which most installations need once in 25 years.
The cost is a method figure
£8,000–£10,000 is what the EPC method allows for panels — the same figure for every home in the country. A real quote depends on scaffolding, roof type, the DNO's answer and how many other jobs the installer has that month.
Nothing here is a return on capital
The money is not invested, it is spent. What comes back is a bill that is smaller and an export payment that is new, and neither is guaranteed in the way an interest rate is.
Selling the house does not end it
Panels transfer with the property and usually add nothing measurable to the price. If you expect to move within a few years, the payback figure on this page is not your payback figure.
Questions
Common questions about solar payback
How long do solar panels take to pay for themselves in the UK?
On these figures, 12–15 years for a 4 kWp south-facing array: £675 a year against £8,000–£10,000 to fit. That assumes 35% of what it generates is used in the house. Use more and it comes down; export more and it goes up.
Are solar panels worth it in the UK?
Financially, on a clear south, east or west roof, yes: the array pays for itself in 12–15 years and keeps generating for a good while after. The cases where it is not are a heavily shaded roof, a house you expect to sell within a few years, and a household with no way to use anything during daylight and a poor export tariff.
Is solar worth it with a battery?
A battery does not generate anything — it moves kilowatt hours from the export rate to the import rate. At today's rates that is worth real money, but the EPC method allows £500–£5,000 for one, a range wide enough that it cannot settle the question on its own. A battery earns its keep faster when it also charges on a cheap overnight tariff, which is the case the battery guide makes.
Do I still get paid for exporting?
Yes, through the Smart Export Guarantee. Every large supplier has to offer a rate above zero, and the obligation says nothing about it being a good one — which is why the rates on offer differ by several times over. You can take your export tariff from a different supplier than your import one.
What about the feed-in tariff?
It closed to new applicants in 2019. It paid for generation as well as export at much higher rates, which is why payback figures quoted from that era are not comparable with anything on this page.
Why do installers quote a shorter payback than this?
Usually a higher self-use assumption, sometimes an energy price rise built into the projection, occasionally both. Ask which self-use share the quote assumes. If the answer is over half and you are out all day, the figure is optimistic.
The generation behind these sums is on the output page, the export rates are compared on the export rates page, and whether a roof suits panels at all is the solar guide. If you are weighing storage, the battery guide takes the other half of the question.
Work it out for your roof
Enter a postcode and we'll use the modelled generation for that council, the cap rate for that region and the export rates available there.