Is an air-to-air heat pump cheaper than an electric heater?
Work out what a room actually loses in heat, the heater size it needs, and what holding it at a comfortable temperature costs per hour, per month and across a winter.
Using the Australian average of 48.69c/kWh — the Australian Energy Regulator, the 2026–27 Default Market Offer, from 1 July 2026.
Same room, same electricity price, one number changed. Set efficiency back to 100 to see what a plug-in heater would cost.
Common cases
What drives the cost
This is the one comparison in electric heating where the answer is not a wash. Every plug-in heater — fan, oil-filled, convector, panel, halogen, infrared — is 100% efficient at the point of use, so they all cost exactly the same to run per kilowatt hour. An air-to-air heat pump breaks that rule because it does not make heat at all: it moves heat from outside to inside, and delivers three to four units of heat for every unit of electricity it draws. That is why the figure above falls by roughly two thirds when efficiency is set to 350 rather than 100.
The catch is capital, not physics. A plug-in heater is tens; an installed air-to-air unit is thousands. The sum turns entirely on hours used per year, which is why it makes obvious sense in a living room heated all winter and no sense at all in a spare room heated at Christmas. Two details soften the headline: efficiency falls as the outside temperature drops, so the coldest days are the least efficient ones, and the same unit runs in reverse as air conditioning in summer, which is increasingly what tips the decision.
How this is worked out
A room loses heat through every surface facing somewhere colder, and through the air that leaks out and has to be warmed again. Each surface loses in proportion to its area, how well it insulates, and how much colder it is outside. Add those together and you get a heat loss coefficient — watts per degree — which belongs to the room alone and does not change with the weather. Multiply it by the temperature difference you want to hold and you have the heat the room needs. For a plug-in electric heater that is exactly the electricity it draws, because a resistive heater turns all of its electricity into heat. Nothing about this is brand-specific: a £15 fan heater and a £300 designer panel of the same wattage cost precisely the same to run.
What this assumes
- U-values for walls, roofs and windows are the published defaults from BRE's RdSAP 10 specification (February 2024) and consultation paper CONSP:16. Those are in-situ measured figures — the solid-brick default was cut from 2.1 to 1.7 W/m²K after fieldwork found real walls beat the theoretical calculation. The five options above use, in order: walls 1.7, 1.5, 0.7, 0.30 and 0.18; windows 4.8, 2.8, 2.0, 2.0 and 1.4; roofs 0.68, 0.40, 0.16, 0.16 and 0.11.
- Floor U-values (0.7, 0.7, 0.45, 0.22, 0.13) and air change rates (1.2, 1.0, 0.7, 0.55, 0.4 per hour) are representative rather than measured. A real floor depends on the ratio of exposed perimeter to area, and real air leakage depends on chimneys, flues and how sheltered the building is — neither is knowable from the questions above, and both are the largest source of uncertainty here.
- The room is assumed square. A long thin room of the same floor area has more external wall and so loses more: 2 × 8 m has a 20 m perimeter against 16 m for 4 × 4 m, and this would under-read it by roughly a quarter.
- Only surfaces facing unheated space are counted. Walls to a heated neighbour, and ceilings under heated rooms, are treated as losing nothing — close enough in practice, though a wall to a neighbour who never puts the heating on is not.
- Nothing is credited for heat you get free. People, cooking, lighting, appliances and winter sun through a south-facing window all add real warmth — often 200 to 400 W in an occupied living room — so a real bill usually lands below the figure here.
- The seasonal figure multiplies the daily cost by the number of days a year your country needs heating at all, at the average outdoor temperature for those days. It is a fair estimate of a whole winter, not a prediction of any particular one.
- The rate filled in for you is an average that spreads fixed charges across a year's usage, so it sits above what one extra unit really costs. Entering the unit rate from your own bill gives a closer answer.
Questions
How can a heat pump be more than 100% efficient?
Because it is not generating heat, it is relocating it. Electricity drives a compressor that extracts warmth from outdoor air — there is usable heat in air well below freezing — and releases it indoors. The output is heat moved plus the electricity used, which is why three to four units out per unit in breaks no laws. It is a fridge run backwards.
Do heat pumps work when it is really cold?
Yes, though less efficiently. Modern cold-climate units produce useful heat well below −15 °C, but the ratio falls as the gap between outside and inside widens, so a unit delivering four units of heat per unit of electricity in mild weather might deliver two and a half in a hard frost. That means the worst efficiency lands on the days you need it most, which is why seasonal averages matter more than headline figures.
Is an air-to-air heat pump the same as air conditioning?
Mechanically, yes — most air conditioners sold today are reversible and will heat as well as cool. If you already have air conditioning with a heat setting, you may already own a heat pump, and running it in winter is likely to be cheaper than any plug-in heater in the house.
Which plug-in electric heater is cheapest to run?
None of them, and any advertisement claiming otherwise is selling something. Identical wattage means identical cost. What varies is how quickly the heat is felt and where it goes: infrared warms people and objects directly rather than the air, which can mean comfort at a lower setting in a draughty or intermittently used space. That is a real advantage, but it is about how the heat is delivered, not how much you pay per unit.
Are some electric heaters cheaper to run than others?
No. Every plug-in electric heater is 100% efficient at the point of use — fan, oil-filled, convector, panel, halogen and infrared alike. A 2 kW heater of any type uses 2 kWh an hour and costs the same. What differs is how the heat feels and how quickly it arrives: infrared warms objects and people directly so it can feel effective sooner, and an oil-filled radiator keeps releasing heat after it switches off. Neither changes the meter reading. Anything advertised as a cheap-to-run electric heater is either a heat pump or a claim worth ignoring.
Is it cheaper to leave the heating on low all day?
No, and the physics above is why. A room loses heat in proportion to how much warmer it is than outside, so a room kept warm all day spends every one of those hours losing heat. Letting it cool means a smaller temperature difference for most of the day and therefore less total loss. Reheating costs something, but less than the standing loss it avoids. The exception is a home so damp that letting it go cold invites condensation, which is a different problem with a different fix.
Why is my conservatory so expensive to heat?
Because almost all of it faces outside. Glass loses heat several times faster than an insulated wall, a conservatory usually has three glazed walls plus a glazed or thin roof, and its floor sits over cold ground. Put a conservatory's numbers in above and the heat loss often comes out three or four times a normal room of the same size. The honest answer is usually to heat it only when it is in use, and to accept it as a spring and autumn room.
Should I heat one room instead of the whole house?
Often yes, if the rest of the house genuinely stays unheated and you accept it being cold. Heating one room costs roughly what this calculator says; heating a whole house costs the same arithmetic repeated for every room, plus the corridors between them. The catch is that unheated rooms get cold surfaces, and cold surfaces attract condensation and then mould. If you go down this route, ventilate the cold rooms rather than sealing them up.
How much does turning the thermostat down by one degree save?
Roughly 7 to 10% of the heating bill for that room, and you can see it in the maths: cost is proportional to the gap between inside and outside. Holding 20 °C against 8 °C is a 12-degree gap; holding 19 °C is 11 degrees, which is 8% less. The saving is bigger in a mild climate, because one degree is a larger share of a smaller gap — which is also why the same advice is worth more in Portugal than in Finland.
Is an air-to-air heat pump worth it for one room?
On running cost, usually yes and by a lot. A heat pump does not make heat, it moves it, so it delivers three to four units of heat per unit of electricity — set efficiency to 350 above and the bill falls by about two thirds. Against that sits the installation cost, which is thousands rather than tens, so the sum turns on how many hours a year you actually heat that room. A room heated all winter pays it back; a spare bedroom heated at Christmas does not.
Related tools
- How much does it cost to run an appliance? Work out what any appliance costs to run per use, per month and per year on your own electricity rate — including what it draws on standby.
- Why do my windows get wet, and will mould grow? Work out the dew point in a room, whether a cold window or wall will run with water, and the humidity you need to stay below to keep it dry.
Sources
- BRE — RdSAP 10 specification (roof, floor and window U-value tables) — checked 2026-08-27
- BRE CONSP:16 — review of default wall U-values for existing buildings — checked 2026-08-27
- BRE — in-situ measurements of wall U-values in English housing — checked 2026-08-27
- ERA5 reanalysis via the Open-Meteo archive — heating climate, 2016–2025 — checked 2026-08-27
- Eurostat — heating degree days by country (cross-check on the climate figures) — checked 2026-08-27
Rates from AER for the 2026–27 Default Market Offer, from 1 July 2026. Last reviewed 27 August 2026.