Central AC vs Window Units: The Comparison Almost Everyone Gets Wrong

Buy window units if you occupy one or two rooms most of the time, if whole-home cooling is not worth what it costs for your house or your climate, or if you rent. Buy central air if you already have usable ductwork, if you want the whole house at one temperature, or if you are cooling enough square footage that per-room appliances stop making sense. What you should not do is choose between them on a monthly running-cost figure you found online, because those figures are almost universally built on a comparison that does not hold.

They solve different problems

A window air conditioner is a complete refrigeration system in one box — compressor, both coils, fan and controls — hanging in an opening in the wall, with the hot side outdoors and the cold side indoors. It cools the air in the room it occupies. Move to another room and you are outside the machine’s reach.

Central air conditioning is one outdoor condenser and one indoor coil, cooling air at a single point and distributing it through ducts to every room, controlled by one thermostat. Its job is a whole house held at one temperature.

Framed that way, most of the usual arguments dissolve. Central air is not “better” — it is a different product doing a bigger job at a proportionally bigger cost. The real question is how much of your house you actually need cooled, and for how many hours a year.

Where the comparison usually goes wrong

Here is the error, stated plainly, because it appears everywhere including in older content on this site: a monthly cost for one window unit is compared against a monthly cost for a whole central system, and the two are presented as if they were the same job.

Of course the window unit is cheaper. It is cooling one bedroom. The central system is cooling a bedroom, three more bedrooms, a hallway, a kitchen and a living room. Comparing them tells you nothing except that cooling less space costs less money, which you already knew, and which is true of every appliance ever made.

How to compare properly

There are two honest methods and both are easy.

Method one — cool the same space. Count the rooms you would actually put window units in, price and run all of them, and set that total against the central system. A household running four window units in July is not paying one unit’s running cost, it is paying four, and the total frequently lands much closer to the central figure than the single-unit comparison suggests.

Method two — cost per conditioned square foot. Divide each option’s running cost by the floor area it actually conditions. This is the fairer comparison and it usually favours central air, because a central system’s compressor is larger and more efficient per unit of capacity than several small ones, and one big machine has less standby and cycling loss than four small machines. Where central air gives that advantage back is in duct losses, which we come to below.

Do it with your own numbers, not ours

A window unit tells you what it costs to run if you read the label. Take its cooling capacity in Btu/h, divide by the CEER rating to get watts, multiply by the hours you run it, divide by 1,000 for kilowatt-hours, and multiply by your electricity rate.

Worked example, assumptions stated: a 6,000 Btu/h unit rated at CEER 11 draws about 545 watts. Eight hours a day for ninety days is 393 kWh. At California’s average residential rate of 33.25 cents per kWh (EIA, May 2026) that is about $131 for the season — for one room. The rate is EIA’s; the arithmetic is ours, and the hours and CEER are assumptions to replace with your own.

Set that against a real baseline. EIA’s Residential Energy Consumption Survey puts average household air-conditioning spending at about $265 a year nationally, roughly $525 in the hot-humid Southeast, and about $60 in the temperate marine climate along the West Coast, where nearly half of households did not use air conditioning at all. Two caveats: that is RECS 2015 data published in 2018 and prices have risen since, and the West Coast figure describes the mild coast rather than the inland Central Valley.

Upfront cost

This is the one place where the gap is enormous and undeniable.

Window and portable units are retail purchases. You buy one, carry it home, and — with a helper and the manufacturer’s bracket — fit it yourself. Prices scale with Btu capacity and you can check them in five minutes at any retailer, so we do not publish a range here: retail pricing is not something any authority indexes, and a made-up range would be worth less to you than a two-minute search.

Central air is an installed system. Contractor cost guides put a full central AC changeout — outdoor condenser, matched indoor coil, standard labour, reusing sound existing ducts — at roughly $3,200–$6,500 for two tons, $4,900–$8,700 for three tons and $8,500–$13,000+ for five tons. If you have no ducts, add a full duct installation, which contractor guides put at roughly $3,500–$12,000 for a typical single-storey home. Our AC replacement cost guide is the canonical page for pricing by tonnage, efficiency tier and region.

Price ranges here are contractor-reported market estimates. No government body or standards organisation publishes installed residential HVAC prices, so we label these rather than present them as sourced figures.

What a window unit does not include, and central air does: ductwork and air distribution, filtration for the whole house, a thermostat controlling a whole house, a permit and inspection, refrigerant recovery and disposal of the old equipment, and — in California — Title 24 duct-leakage and charge verification where the code requires it. Those are not upsells; several are code. They are also the reason the two prices are so far apart.

Efficiency ratings: why you cannot compare the numbers

Room air conditioners are rated in CEER — the Combined Energy Efficiency Ratio, the federal metric ENERGY STAR uses for room units, which folds in standby and off-mode power as well as cooling performance. Central systems are rated in SEER2, a seasonal figure produced by an entirely different test procedure at a higher external static pressure.

Different tests, different conditions, different scales. A CEER of 11 and a SEER2 of 15 are not fifteen-versus-eleven. Any page that lines them up in a table is comparing two rulers marked in different units. Use CEER to compare window units against each other, use SEER2 to compare central systems against each other, and do not cross the streams. Our SEER2 explainer covers the central-system side.

There is one real efficiency difference worth naming, and it runs in the window unit’s favour: a window unit has no ducts, so it has no duct losses. ENERGY STAR states that sealing and insulating ducts can improve heating and cooling efficiency by as much as 20 percent, which is another way of saying that a leaky, poorly insulated duct system is quietly throwing away a meaningful share of what the central system produces. Neither SEER2 nor CEER measures that, because both rate the equipment rather than the installation. If your ducts run through a hot attic and have never been sealed, the on-paper efficiency comparison overstates the central system’s real advantage. See what duct work and sealing cost.

Comfort and liveability

This is where most people actually make the decision, and it is worth being specific rather than listing adjectives.

Noise, and where the compressor sits. A central system puts its compressor outside the house. A window unit puts its compressor in the room you are trying to sleep in, four feet from your head, cycling on and off all night. This is the single most common regret with window units and it is not a manufacturing defect — it is the design.

Even temperature and humidity. A window unit produces a temperature gradient: cold near the unit, warm at the far wall, warmer still in the next room. Central air moving conditioned air to every room produces a flat profile. On humidity, both remove moisture only while the compressor is running, which is why sizing matters so much (below).

The window you lose. A unit in a window blocks light and view, prevents the window opening, and — this one matters — is a security consideration, because a window with a unit in it is not securely closed. Many units can be pushed out of a poorly braced opening from outside.

Appearance and resale. Central air is generally treated as an expected feature in a home listing in hot markets and window units are not. That is a real effect on saleability, though it is not a number anyone can put a figure on honestly.

HOA and rental rules. Many homeowners’ associations restrict window units on street-facing elevations, and many leases prohibit them outright or require landlord approval and specific mounting. Check before you buy — this catches people every summer.

Sizing a window unit, and why bigger is worse

The instinct is to buy a size up for margin. It is wrong, and the penalty is larger on a window unit than on central air.

ENERGY STAR and EPA’s right-sizing research describes the mechanism: oversized air conditioners “short-cycle,” running for shorter periods than engineered for optimum operation. “The efficiency of air conditioners is low when they first start and increases gradually, reaching peak efficiency in about 10 minutes,” and “when operating time increases from 5 to 9 minutes, efficiency improves 17 percent.” Crucially: “short operation times do not allow the system to effectively remove humidity.”

That applies with more force to a window unit, because there is no duct system and no large air volume to buffer the swing. An oversized unit blasts the room to setpoint in minutes, shuts off, and leaves the air cold and damp — the classic clammy bedroom.

Use ENERGY STAR’s own sizing chart rather than a table from a blog. It matches room square footage to Btu per hour on the room air conditioners page, and the adjustments matter: heavy direct sun needs more capacity than raw square footage suggests, a shaded room less, high ceilings raise the load because you are cooling volume rather than floor area, and rooms routinely holding several people or containing a kitchen need more.

On the central side the equivalent discipline is a Manual J load calculation with Manual S used to select equipment that delivers that capacity at your design conditions — nameplate tonnage is not the same thing. See how to size an AC properly and short-cycling.

The middle options

Portable units sit on the floor and vent through a hose into a window kit. They are the right answer when a window cannot take a mounted unit — casement windows, sliders, some HOA situations, and the wrong answer otherwise. A single-hose portable pulls its condenser air from the room it is cooling and blows it outside, putting the room under negative pressure and drawing warm outside air in through every gap. That is why a portable rarely matches a window unit of the same nameplate capacity.

A single-zone ductless mini-split is the serious version of “cool one room properly” — an installed refrigerant-line system rather than a plug-in appliance. It is quiet indoors because the compressor sits outside, far more efficient than a window unit, heats as well as cools, and does not occupy a window. Contractor cost guides put a single-zone install at roughly $1,500–$3,100 for a 9,000–12,000 Btu head, with a broad span of about $2,000–$7,000. If one room is always uncomfortable, price this before buying a third window unit; our mini-split vs central AC comparison covers the choice.

And if one part of an already-ducted house will not cool while the rest is fine, adding a machine may be the wrong move — start at why upstairs won’t cool.

Window unit safety, which most guides skip

Four things, all real and all routinely omitted:

• Support and mounting. A modern window unit is heavy enough to seriously injure someone below if it falls, and heavy enough to damage a sash it was never meant to sit on. The manufacturer’s bracket and installation instructions are not optional extras. If the unit is going above ground-floor level, brace it properly or have someone fit it.
• Electrical. A window unit belongs on a properly rated circuit. Do not run one on an extension cord or a light-duty power strip. If a unit repeatedly trips its breaker, that is a defect and not a nuisance — the Electrical Safety Foundation International states that an appliance which “repeatedly blows a fuse or trips a circuit breaker could indicate a defect that may cause a fire or electrical shock.” Reset once; if it trips again, stop.
• Egress. A unit fitted in a bedroom’s only window can block the room’s escape route in a fire. Check that the room has another way out before you fill the window.
• Condensate and biological growth. A window unit that does not drain properly holds standing water against a coil in a warm room, and what grows there gets blown into the air you breathe. EPA’s guidance on mould is the reference; the practical version is that the unit needs to be pitched so it drains outward, the drain path needs to stay clear, and a unit that smells musty needs cleaning rather than air freshener.

On the central side the equivalent line is short but absolute: before your hands go near the outdoor unit or the air handler, switch off the breaker at the panel and pull the outdoor disconnect. Both, every time.

Who should buy which

Not a scoring table — a set of situations. Find yours.

Whichever way you go, the discipline is the same: compare like with like, size it to the actual load rather than to instinct, and treat any monthly running-cost figure you did not calculate yourself as marketing until you have checked what space it describes.

FAQ