How the two differ in kind
A furnace makes heat by burning something, so its output is whatever the burner is rated at, on every day of the year. A heat pump does not make heat — it MOVES it, from outdoor air that is cold into a house that needs to be warmer. That is why it can deliver several units of heat for one unit of electricity, and it is also why its capacity depends on how much heat there is outside to move.
So the two behave in opposite directions as the weather worsens. The building's demand rises as the outdoor temperature falls; the furnace's output stays flat above it; the heat pump's output falls towards it. Somewhere the two curves cross, and that crossing is the BALANCE POINT — the outdoor temperature at which the heat pump exactly meets the load. Above it the machine modulates. Below it something has to make up the difference, and sizing that something is a separate decision from sizing the heat pump.
The sizing rule differs for the same reason. Oversizing a furnace costs efficiency through short cycling and nothing else. Oversizing a heat pump is worse: it is heavily oversized for the mild weather where it does most of its annual work, and a variable-speed machine that cannot modulate low enough starts cycling too — which costs both efficiency and compressor life. A heat pump is selected against the load curve with a deliberate balance point, not against the peak.
The factors that actually differ
Show
| Air-source heat pump | Gas or oil furnace | |
|---|---|---|
| Output when it is coldest | Falls with outdoor temperature, exactly as demand rises. Manufacturers publish capacity at several outdoor temperatures for this reason, and the mid and low figures are the ones that matter. | Flat. The burner is rated at its input and delivers it at any outdoor temperature. |
| Sizing rule | Against the load CURVE with a chosen balance point, not the peak — because the peak is a few hours a year and oversizing costs efficiency for the rest of it. | Against the design heat loss, with the usual caution that safety factors multiply and an oversized furnace short cycles. |
| What happens below the balance point | Supplementary heat: electric resistance, a hybrid gas stage, or a wood stove. Sized from the GAP rather than from the whole load — and the tail below the balance point is a small share of annual energy. | Nothing. There is no balance point and no supplementary stage. |
| Delivery temperature | Lower — warm air rather than hot, or low-temperature water. Longer, gentler run times feel different, and a system tuned like a furnace feels draughty. | Hot air in short bursts. Familiar, and the reason a heat pump retrofitted onto furnace ductwork can be reported as feeling cold at the register while the room is at setpoint. |
| What the emitters have to do | Needs a large emitter surface or a large air volume, because the delivery temperature is low. Ductwork sized for a furnace is frequently too small. | Compact emitters and smaller ducts, because the temperature difference does the work. |
| Cooling | Included. The same machine reverses and is the air conditioner as well. | A separate system entirely, with its own cost, its own outdoor unit and its own sizing. |
| Defrost | In cold damp weather the outdoor coil ices and the machine periodically reverses to clear it, during which it is not heating. Real, designed for, and the reason the nominal rate and the delivered seasonal rate differ. | Not applicable. |
| Combustion safety | None to consider. No flue, no combustion air, no carbon monoxide. | A flue, a combustion air supply, and a building that must not be depressurised enough to backdraft it — which is a real constraint in an airtight house with a powerful kitchen hood. |
| Shape of the cost | Higher capital, often larger emitters or ducts too, and a running cost that depends on the ELECTRICITY-to-fuel price ratio and on the flow temperature the system runs at. | Lower capital where the gas connection exists, and a running cost tied to one fuel price. Where there is no gas main the comparison is against oil or LPG and looks very different. |
Which one, and when
Choose air-source heat pump when…
- There is no gas connection, and the alternative is oil, LPG or electric resistance.
- Cooling is wanted as well, and one machine doing both is worth more than either alone.
- The fabric is good enough that the design load is modest and the emitters can be sized for low temperatures.
- The electricity-to-fuel price ratio in your area, divided by the seasonal efficiency you can realistically expect, comes out ahead — which is a local calculation, not a general one.
Choose gas or oil furnace when…
- The house is leaky or the emitters are small, so a low-temperature system would need work before it worked.
- Peak demand is severe and sustained, and the balance point would sit high enough that supplementary heat would run for much of the winter.
- The gas connection already exists and the fuel price ratio favours it clearly at your tariffs.
- The existing ductwork is sized for high-temperature delivery and cannot be enlarged without opening the building up.
Now run your own numbers
This page holds no prices on purpose — a national average is wrong for almost every real project. Quantify both options with your dimensions and your local quotes.
Frequently asked questions
- Do heat pumps work in cold climates?
- They work, and the honest statement is that they work with less capacity and lower efficiency the colder it gets, which is a design input rather than a disqualification. Cold-climate models are specifically built to hold more of their capacity at low temperatures, and manufacturers publish the capacity at several outdoor temperatures precisely so the balance point can be found. The real questions are how many hours a year your climate spends below that balance point and what covers the gap. In many places the answer is a modest number of hours and a small electric backup stage that rarely runs; in a genuinely severe climate it may be a hybrid system with a fuel-burning stage for the tail. What does not work is sizing on the nameplate figure, which is quoted at a mild outdoor temperature and describes a condition your worst week does not contain.
- Why does my heat pump blow air that feels cool?
- Because it is. A furnace delivers air well above body temperature in short bursts, so it feels hot at the register; a heat pump delivers air that is warmer than the room but often below skin temperature, over much longer run times. The room reaches the same setpoint by a different route. Two things make it worse if they are present: airflow higher than needed, which makes the delivery cooler still, and registers positioned to blow across occupants rather than along a wall. Neither is a fault in the machine. The system is designed to run long and gently, and it is judged incorrectly by the standard a furnace set.
- Can a heat pump use my existing ductwork?
- Sometimes, and the check is airflow rather than fit. Because the delivery temperature is lower, a heat pump moves considerably more air for the same heat output than the furnace it replaces — so ductwork sized tightly for the furnace can be undersized for the heat pump. The symptom is not an obvious failure: it is high static pressure, reduced airflow, noise, and a machine that cannot reach its rated output while appearing to run normally. A duct assessment before the equipment is chosen is the difference between a retrofit that works and one that is blamed on the technology. The same logic runs the other way in a wet system, where the emitters rather than the ducts are the thing that has to grow.
- How is the backup sized, and how often does it run?
- From the GAP between the load and the heat pump's output at the design temperature, not from the whole load — which is why backup stages are so much smaller than a full furnace. How often it runs depends entirely on where the balance point sits: raise it by undersizing the heat pump and the backup runs for much of the winter at resistance-heating efficiency, which destroys the economics; lower it by oversizing and the machine cycles inefficiently for the other eleven months. That trade is the actual design decision on a heat pump installation, and it is made by looking at the hours the climate spends at each temperature rather than at the coldest night anyone remembers.
- Which is actually cheaper to run?
- It turns on two local numbers and this site will not invent either. The first is the ratio between your electricity price and your fuel price per unit of energy; the second is the seasonal efficiency the heat pump will actually achieve in your climate and at your flow temperature — not the headline figure, which is measured at mild conditions. Divide the price ratio by the seasonal efficiency and you have the comparison directly. Where gas is cheap and electricity dear the ratio is demanding; where the alternative is oil, LPG or electric resistance the heat pump usually wins comfortably. Both inputs are on your own bills, and getting them is more useful than any national average.
- What does defrost actually cost me?
- Two things, and neither is dramatic if the system was designed for it. During a defrost cycle the machine is not heating the house — it is warming its own outdoor coil, and in many systems a backup stage covers the room for those minutes. And the energy spent doing it is real, which is part of why a seasonal efficiency figure is lower than the efficiency measured at a steady mild condition. Defrost is worst in cold damp weather near freezing, where there is the most moisture available to freeze on the coil, and it is much less frequent in dry cold. A system that seems to defrost constantly usually has an airflow problem at the outdoor unit — leaves, snow drift, or too little clearance — rather than a fault in its controls.