The case for a home battery is usually made with a picture: sunshine going into a box during the day, coming out of it at night, and a bill going down. The picture is accurate and it explains nothing about whether the box is worth buying.
One equation does.
What a battery actually earns
Every unit a battery stores is a unit that would otherwise have been exported. Every unit it releases is a unit you would otherwise have bought. So the value of moving one kilowatt-hour through the battery is the difference between your import price and your export rate.
Annual earnings are that spread multiplied by the number of units the battery moves in a year, less what is lost in the round trip.
Both terms are knowable. The spread comes off your bill and your export tariff. The number of units moved is the one people guess at, and it is the one that decides the outcome.
Why capacity is not throughput
A 10 kWh battery does not move 10 kWh a day. It moves whatever is both available to store and needed later, which is limited at both ends.
- In December a British array generates a few kilowatt-hours across the whole day, most of which the house uses as it arrives. There is little surplus to store, so the battery sits mostly idle regardless of its size.
- In June there is plenty of surplus, but the battery can only deliver what the house consumes after generation stops. A household using 4 kWh between dusk and dawn cannot extract more than 4 kWh, whatever the battery holds.
- A battery larger than the smaller of those two limits is capacity that never gets used, in any month.
That is why the honest sizing question is evening and overnight consumption, not array size. The battery should be sized against what the house draws when the sun is down, and specifically against the summer version of that figure, because summer is when there is surplus to store.
Profiles, and what each one implies
| Profile | Surplus to store | Demand to discharge into | Case for a battery |
|---|---|---|---|
| Out all day, both working, high evening use | Large, most generation unused | Large, cooking and heating after 6pm | Strongest |
| Electric vehicle charged at home overnight | Large | Very large, but see note below | Strong, with caveats |
| Home all day, work from home, daytime laundry | Small, already self-consumed | Moderate | Weak. Self-consumption is already high |
| Low total consumption, small household | Moderate | Small | Weak. Little to discharge into |
| Large array relative to consumption | Very large | Unchanged by the array size | Limited by demand, not by surplus |
| Heat pump, electric heating | Small in winter when demand peaks | Very large, but mostly in winter | Poor seasonal match |
The last row is the one most often got wrong. A heat pump raises consumption enormously in exactly the months when a solar array has no surplus to store.
Two of those rows need expanding.
An electric vehicle looks like the perfect discharge target and frequently is not. Overnight EV tariffs price imported electricity very low for a window in the small hours, and a battery that discharges into the car is displacing electricity that was cheap to begin with. The spread it earns is against the off-peak rate rather than the standard one, which can be a fraction of the spread it appears to be earning. A battery paired with an EV tariff is often better used to store cheap off-peak grid electricity for daytime use than to store solar, which is a different product decision.
A heat pump is a seasonal mismatch. It draws heavily in December and January, when the reference array in these guides produces 134 and 144 kWh for the whole month. There is essentially no solar surplus to store in the months when the heat pump needs it.
The spread is the other half, and it moves
The wider the gap between import and export, the more each stored unit is worth. Two things narrow it.
A good export rate reduces the value of storing, because the unit you would have exported was already worth something. This is counterintuitive and it is real: negotiating a better export tariff makes a battery less attractive, not more. Households sometimes do both and are surprised that the second decision undercut the first.
A low import price does the same from the other side. And in the Netherlands, where net metering made export worth exactly as much as import until the end of 2026, the spread was zero and a battery earned nothing from this mechanism at all. What changes there on 1 January 2027 is set out in Dutch net metering ends.
What to check before deciding
- 1Your evening and overnight consumption in summer, in kWh. Half-hourly data from your smart meter gives this directly, and it is the number that sizes the battery.
- 2Your import unit rate and your export rate, both current. The difference between them is the value of each moved unit.
- 3Round-trip efficiency of the specific unit, which is the share of stored energy you get back.
- 4The installed cost, and the warranty term. A payback longer than the warranty is the definition of a bad purchase, and how to judge that is in home battery lifespan.
- 5Whether the quote models the battery hour by hour or applies an annual assumption. The second cannot capture the seasonal limits above.
Our calculator works this through on your own consumption and export rate, and reports the payback period alongside the range rather than a single figure. Where the modelled payback runs past the battery warranty it says so plainly, which is the case set out in when a home battery is not worth buying.
Monthly generation figures are modelled with PVGIS 5.2 for a 4 kWp roof-mounted array in the English Midlands. Import and export rates are household-specific and must come from your own bill and export tariff. Checked 2026. Source