Are Solar Batteries Worth It in the UK? (2026)

By Rob Hammond · Updated October 2026 · How we score

PowerGuide is reader-supported. We may earn a commission from links on this page. It never affects our rankings or what we recommend. Learn more

A home battery can be worth it, and quite often it isn’t. Which one you get depends far more on your tariff than on your solar, and the honest arithmetic is less flattering than most of the internet suggests. Here it is, with the figures sourced and dated wherever they can be.

The two ways a battery makes money

1. Off-peak arbitrage. Charge from a cheap overnight rate, discharge during the expensive day. You are paid the difference between two import rates.
2. Storing your own solar. Keep generation you’d otherwise export and use it in the evening. You are paid the difference between an import rate and an export rate.

Those two gaps are not the same size, and that single fact decides most cases. As of 20 September 2026, Octopus’s Intelligent Octopus Go charges 8p/kWh between 23:30 and 05:30 (it was 8.4p until 1 October 2026). Ofgem’s price cap for 1 October to 31 December 2026 puts the average direct-debit electricity unit rate at 26.32p/kWh. That is a gap of 18.32p per kWh.

Now the solar side. Outgoing Octopus pays a flat 12p/kWh for export. Against the same 26.32p import rate, storing a kWh of your own solar instead of exporting it is worth 14.32p, and only on the kWh you would otherwise have exported.

Arbitrage wins, and it wins on volume as well as margin: you can run an overnight charge cycle 365 days a year, whereas surplus solar is a summer phenomenon.

A worked payback, using catalogue prices

The batteries in our catalogue, priced as hardware. These three prices are indicative catalogue estimates, not sourced figures. Only the Dura16’s £3,348 traces to a named retail listing (£2,790 ex VAT at Powerland). Read them as the scale of the purchase, not as quotes, and do not read them as carrying the same weight as the Ofgem and PVGIS figures cited below:

The number that decides everything: how much electricity you use during the day. A battery only earns on the kWh it moves out of the expensive hours. Electricity you already use between 23:30 and 05:30 is billed at 8p whether you own a battery or not.

Ofgem’s own figures put a ceiling on this. The same decision that set the 2026 consumption values keeps the Economy 7 split at 58% peak / 42% off-peak (Table 2). So:

Round-trip efficiency: 89%, which is the figure on Tesla’s UK Powerwall 3 datasheet. Read it carefully, though: Tesla’s 89% is the solar to battery to home path, which is DC-coupled and involves one inversion. Grid arbitrage runs AC to DC into the battery and back from DC to AC, an extra conversion, so a real AC round trip is lower than 89%, and it will differ again on the GivEnergy and the Dura16, neither of which publishes a comparable figure. We use 89% as a best case, which means every payback below is optimistic.

The medium household, shifting 5.4kWh a day:

• To put 5.4kWh into the house you must buy 5.4 ÷ 0.89 = 6.07kWh from the grid.
• At 8p/kWh that costs 48.5p.
• The same 5.4kWh bought at 26.32p would have cost £1.42.
• Daily saving: 93.6p. Over a year: £342.

The high-use household, shifting 8kWh a day (it has 9.7kWh of daytime load, so the battery is not the binding constraint):

• Buy 8 ÷ 0.89 = 8.99kWh at 8p = 71.9p; avoid 8kWh at 26.32p = £2.11.
• Daily saving: £1.39. Over a year: £506.

Against those annual savings:

A warranty is only as good as the company behind it. GivEnergy entered administration in 2026. The brand may continue, but confirm who stands behind that 12-year warranty before you treat it as cover. On the high-use column it is the only thing that turns a 9.9-year payback into a pass.

That gap is the single most important thing on this page, and the warranty column is the test to apply to it: a payback longer than the warranty is not a payback, because the thing has to still be working when it finally breaks even.

On Ofgem’s medium consumption value, only the Dura16 passes that test, and only just, at 9.8 years against a 10-year warranty, with no margin for a fitted price above the hardware price. The GivEnergy’s 12 years falls short of 14.6, and that 12 years is itself uncertain, because GivEnergy entered administration in 2026; the Powerwall’s 10 falls short of 21.9. Move to the high-use column and the Dura16 passes comfortably at 6.6 years, the GivEnergy passes at 9.9 only if its 12-year warranty is honoured, and the Powerwall still fails at 14.8 years against 10. A home battery bought for off-peak arbitrage is a high-consumption household’s purchase, and on a medium household it is at best a close-run thing.

Three caveats that matter more than the numbers.

First, those are hardware prices, not fitted prices. DESNZ publishes official cost statistics for small-scale solar, but that series explicitly excludes battery storage: there is no equivalent official figure for what a UK battery installation costs, so we are not going to invent one. Your fitted price will be higher than the table, and every payback above will be longer. Get quotes.

Second, the Dura16 is sold only alongside a Duracell inverter, so its £3,348 is not directly comparable with the All in One, which includes its inverter and a backup gateway.

Third, Tesla warrants the Powerwall 3 for 10 years. Even in the high-use column, a 14.8-year payback on a 10-year warranty is not a payback. That is the Powerwall earning its money on backup and 11kW of output instead, which is a real reason to buy one, just not this reason.

At this point, put your own numbers in. The calculator below starts on the medium household above (9.3kWh a day, 58% outside the cheap window, the GivEnergy’s £4,999), so you can see the 14.6-year figure and then change what is different about your home:

The solar payback calculator handles the panels-plus-battery case.

When a battery is worth it

• You have an EV. Not for the car, but for the tariff. Intelligent Octopus Go requires a compatible car or charger, and the 8p rate it unlocks applies to everything in the house, not just the charger. The EV is the key to the cheap window, and the battery is what lets you use the window for the other 18 hours of the day.
• You use a lot of electricity during the day. Working from home, an electric shower, a heat pump. The saving scales almost linearly with the kWh you shift, which is why the high-use column above earns 48% more than the medium one and turns a 14.6-year payback on the GivEnergy into a 9.9-year one.
• You have solar and export a lot of it. Not because storing solar is lucrative on its own, but because a battery lets you qualify for an import-and-export tariff like Octopus Flux, which requires solar, a battery, an export MPAN and a half-hourly smart meter.
• You need backup. If your supply drops often enough to matter, that is a reason on its own and the payback arithmetic is beside the point. Check the spec, though: of the three batteries above, only the All in One and the Powerwall 3 offer whole-home backup.

When the answer is no

This is the part the comparison sites leave out.

No EV and no time-of-use tariff. Intelligent Octopus Go needs a compatible car or charger. Without one you’re on Octopus Go, whose cheap window is five hours (00:30–05:30) rather than six, and whose rates are regional. On a single flat rate you have no spread at all, and because at best 89% of what you put in comes back out, every kWh you push through the battery costs you more than a tenth of a kWh. A battery on a flat tariff with no solar is a device for losing money slowly.

Low consumption. Ofgem’s 2026 low typical consumption value for a multi-rate household is 1,900kWh a year, the right figure to use here, because a battery on this strategy needs a time-of-use tariff and therefore a multi-rate meter. Apply the same 58% daytime split and 1,102kWh is shiftable, about 3.0kWh a day. Charging costs 1,238kWh × 8p = £99; the avoided day-rate purchase is 1,102kWh × 26.32p = £290; the saving is £191 a year. Against the cheapest battery in the table that is a 17.5-year payback on a 10-year warranty. The answer is no, and no amount of sizing will fix it: the ceiling on a battery’s earnings is the electricity you use.

Solar, but no time-of-use tariff. Take a 4kWp south-facing array at 35° in the Midlands. The European Commission’s PVGIS model puts that at 3,963kWh a year. Suppose 1,000kWh of it would otherwise be exported and the battery captures it instead. You give up export on the full 1,000kWh, at 1,000 × 12p = £120, but you only avoid importing the 890kWh that come back out of the battery: 890 × 26.32p = £234. Net, you are £114 a year better off. That is a 43.8-year payback on the GivEnergy. Solar self-consumption alone does not justify a battery at current export rates.

(Note the difference in method between the two examples. In arbitrage you buy the losses, so the loss is charged to the purchased side. In solar self-consumption you forgo the losses, because they were free generation you’d already have been paid for exporting, so the loss is charged to the delivered side. Getting these the same way round is the commonest error in battery payback maths, and it flatters the solar case by about £13 a year here.)

You haven’t insulated yet. A battery changes the price of a kWh. Insulation changes how many kWh you need. If your loft is thin, do that first: it is cheaper and the saving does not stop after ten years.

You’re moving within five years. No UK evidence establishes that a battery adds its cost back to a house price, and we’re not going to claim otherwise.

How to size one

The instinct is to buy the biggest battery you can afford. That is wrong, because a battery only earns on the kWh it actually cycles, because capacity you never discharge is capital doing nothing.

Size it to the electricity you use outside your cheap window, on an average day, not a worst day. For the medium 3,400kWh household above that’s 5.4kWh, so a 13.5kWh battery is more than twice the size the arithmetic supports, and you are paying for 8kWh of capacity that will sit idle most nights. Even the high-use 6,100kWh household only needs 9.7kWh. The extra capacity earns its keep only in winter, or if you add a heat pump or an EV that charges during the day, or if you want a reserve held back for power cuts.

Two structural choices follow:

• All-in-one or modular? An all-in-one like the GivEnergy or the Powerwall 3 includes the inverter and is a simpler install. Modular packs like the Pylontech US5000 at 4.56kWh usable are the cheaper route if you already have a compatible hybrid inverter, and a bad deal if you don’t, because you’re then buying an inverter too.
• Chemistry. All four batteries here are LFP, which is now the default for home storage. It is the chemistry you want; treat anything else as needing a specific justification.

Compare the full set on our home battery storage page.

Two things that change the maths in 2027

VAT on the battery. HMRC zero-rates the installation of battery storage in residential accommodation, including a standalone battery charged only from the grid, until 31 March 2027, after which it returns to the 5% reduced rate (Notice 708/6). On a £5,000 install that is a £250 difference, which is real but is not on its own a reason to rush.

VAT on the electricity. The government removed VAT from domestic electricity bills from 1 October 2026 to 31 March 2027 (Ofgem). That is why the 26.32p cap rate is what it is. When VAT returns, the day rate rises, the spread against an 8p overnight rate widens, and every payback on this page gets shorter. Worth knowing; not worth buying a battery over.

All prices and rates were checked on 20 September 2026. Tariff rates are regional and change with wholesale prices, and the price cap is reset quarterly, so re-check before you commit.

What’s changed (changelog)

• 3 October 2026. Corrected the Powerwall 3’s medium-household payback from 22.0 to 21.9 years (£7,500 ÷ £342 is 21.93). Also corrected the high-use summary, which said all three batteries pass the warranty test: the Powerwall’s 14.8 years against a 10-year warranty does not, as the caveats further down already said.

• 20 September 2026 (third revision, same day). Qualified the GivEnergy warranty: the company entered administration in 2026, so a bare “12 years” was doing consumer-protection work it cannot support, and the product page already said so. Also labelled the three catalogue prices as indicative estimates, and only the Dura16’s traces to a named retail listing, so they are not read as carrying the same weight as the Ofgem, PVGIS and DESNZ figures cited here.

• 20 September 2026 (second revision, same day). Added the warranty column to the payback table after review. An earlier version of this page claimed no battery paid back inside its warranty on the medium household; the Duracell Dura16’s 10-year warranty does cover its 9.8-year payback, and the guide now says so. The low-consumption case also applied the multi-rate day/night split to Ofgem’s single-rate consumption value; it now uses the multi-rate low value of 1,900kWh, giving £191 a year and a 17.5-year payback rather than £161 and 20.8.

• 20 September 2026 (revised same day). Corrected the worked payback after review. The first version of this rewrite assumed 8kWh a day of shiftable load on Ofgem’s medium 3,400kWh consumption value; Ofgem’s own Economy 7 split (58% peak) caps that household at 5.4kWh, so the saving is £342 a year, not £506, and the paybacks are years longer. The solar self-consumption example also charged round-trip losses to the wrong side of the calculation (£127 a year, now correctly £114). Added the caveat that Tesla’s 89% is a DC-coupled figure, not a grid AC round trip.

• 20 September 2026. Rewritten around a fully sourced worked payback using the October 2026 price cap, the Intelligent Octopus Go 8p rate and real catalogue prices, plus the cases where the answer is no. The previous version asserted a “5–13kWh” sizing range and a 10-year warranty norm without sources; both have been replaced with cited figures.

Related products

GivEnergy · 13.5kWh

GivEnergy All in One

Integrated battery, inverter and gateway in one cabinet, with whole-home backup. A strong fit for storage on a single unit, but check warranty support given GivEnergy's recent administration.

B4.2 ~£4,999

Specs & review

Tesla · PW3

Tesla Powerwall 3

13.5kWh battery with a built-in solar inverter and class-leading continuous output, including whole-home backup.

B4.4 ~£7,500

Specs & review

Duracell · PD-16KWH-50V-1G

Duracell Dura16

A 16kWh low-voltage battery built by Puredrive Energy under the Duracell name, with 14.4kWh usable and an 8,000-cycle rating, sold to pair with Duracell's own inverter.

B4.1 ~£3,348

Specs & review