Manual J or gross feeling? Why heat pump load calcs still get skipped

I looked at a quote last month for a house near Merri Creek in Melbourne's inner north — a nine-square-metre kitchen extension, single-glazed original section, and the installer's entire sizing method was "how many square metres is the house, times a number." No blower door, no window schedule, no mention of insulation. That's not a load calculation. That's a guess wearing a spreadsheet.

Here's the thing homeowners get wrong most often with heat pumps: they assume sizing is basically settled once someone tells them the floor area and the climate zone. It isn't. A proper heat pump load calculation — Manual J in the US, or the equivalent room-by-room method under AS/NZS 3251 and NatHERS-informed practice in Australia — is a room-by-room, surface-by-surface exercise. Skip it, and you end up with either an oversized unit that short-cycles and dehumidifies badly, or an undersized one that can't hold temperature on the coldest night of the year. I've written before about why oversizing quietly ruins comfort and running costs, so I won't re-litigate that here. This piece is about the calculation itself: what it actually checks, why "rule of thumb" sizing survives despite being wrong, and what you should be asking to see before you sign.

What a rule-of-thumb estimate actually misses

The most common shortcut is square-metre based: multiply floor area by a rough figure (something like 100-130 W/m² is thrown around a lot in Australia, or a BTU-per-square-foot rule in the US) and call it the design load. It's not completely useless as a sanity check, but it treats a house as a uniform box. Real houses aren't.

A load calculation has to account for the amount of glazing and its orientation, the insulation level in each wall and ceiling section, air leakage, internal heat gains from appliances and occupants, and the local design temperature — not the average winter minimum, but the specific design-day figure your climate zone uses roughly 1% or 2.5% of the time. Two houses with identical floor area can have wildly different loads if one has floor-to-ceiling west-facing glass and the other has small, well-shaded windows. A flat W/m² number can't see that difference. Only a room-by-room calc can.

Why installers skip the proper calculation anyway

Honestly, most of the time it's not laziness in the mean sense — it's commercial pressure. A full Manual J or AS/NZS 3251 assessment takes real time: measuring room dimensions, checking window specs, sometimes climbing into a roof space to look at insulation batts. On a competitive quote, that's an hour or two an installer doesn't get paid for unless the client explicitly asks and is willing to wait for it.

The other reason is that rule-of-thumb sizing rarely gets called out. If a heat pump is oversized, the failure mode is subtle — short cycling, patchy humidity control, a system that feels "fine" but costs more to run and wears out compressor components faster from constant starts and stops. Nobody rings up in month three and says "my heat pump is too big." They just notice the electricity bill is higher than the salesperson promised, and they usually blame the unit, not the sizing method.

I'd argue this is where the industry's incentives are genuinely misaligned with the homeowner's interest. A slightly oversized system is a safer bet for the installer — fewer callbacks about "it's not keeping up" on the coldest night — even though it's the worse outcome for the customer's power bill and comfort over a fifteen-year service life.

What Manual J and AS/NZS 3251 actually check

Manual J, published by the Air Conditioning Contractors of America, is the industry-standard residential load calculation method used across the US. The equivalent process in Australia sits under AS/NZS 3251 for mechanical services design, usually informed by the same building envelope data that feeds a NatHERS assessment. Both do the same fundamental job: they calculate heat loss and heat gain room by room, not for the house as a single number.

That means the assessor needs, at minimum, the dimensions and orientation of every room, the U-value or R-value of each wall, ceiling and floor assembly, the specification of every window and door (glazing type, frame material, shading), an estimate of air changes per hour from infiltration, and the design temperatures for both summer and winter for your specific location. From that, they calculate a sensible and latent load for each room, which then rolls up into a total load for the system and, critically, tells you how much airflow or capacity each zone actually needs — which matters enormously for ducted systems where duct sizing gets it wrong almost as often as the equipment does.

If you're comparing a ducted system to a multi-head split, the load calc is also where you find out whether ducted actually makes sense for your floor plan, something I go into more in ducted vs split-system heat pumps.

The blower door and envelope data problem

Here's a wrinkle that catches a lot of people out: a load calculation is only as good as the envelope data going into it. If nobody has measured your home's actual air leakage — via a blower door test — the assessor is estimating infiltration from a table based on age and construction type, which can be off by a wide margin, especially on older weatherboard or brick-veneer homes that have had gaps sealed inconsistently over decades of renovations.

This is one of the reasons I keep pointing people toward getting air sealing assessed before or alongside heat pump sizing, not after. If you've had a proper air sealing job done, or you're planning one, that changes your design load meaningfully — sometimes by 15-20%. We've covered why air sealing has to come before insulation in the order of operations, and the same logic applies here: get your envelope numbers as accurate as you can before anyone sizes equipment against them. An assessor doing a rushed quote-stage estimate usually doesn't have blower door data, and won't ask for it unless you bring it up.

What running-cost mistakes actually look like downstream

The running-cost consequences of bad sizing aren't dramatic, which is part of why they're missed. An oversized unit reaches its setpoint fast, then shuts off, then starts again — each start is where a compressor draws the most current and works least efficiently, so the coefficient of performance you see on the spec sheet (tested at steady-state conditions, per AS/NZS or the equivalent ENERGY STAR test procedure) never actually shows up in your bill. You bought a machine rated at, say, a COP of 4, and you're living somewhere closer to 2.5-3 in practice because it never runs long enough to hit its efficient operating band.

For hot water heat pumps specifically, undersizing shows up differently — as a system that constantly runs its backup electric resistance element to make up shortfall, which quietly wrecks the running-cost advantage that made you choose a heat pump water heater in the first place. I go through that math in detail in the heat pump hot water running-cost piece, but the short version is: a wrong-sized tank or unit undermines the whole premise of the purchase.

What to actually ask for before you sign

Ask the installer directly whether they perform a Manual J (US) or an AS/NZS 3251-based room-by-room load calculation, or whether their quote is based on a square-metre rule of thumb. Ask to see the room-by-room output, not just a single final number — a legitimate calculation produces a report with individual room loads, and a reluctance to show that is itself a signal. Ask what design temperature they used and where it came from; it should be a recognised climate-zone figure for your postcode or county, not a rough guess.

If you're getting multiple quotes and the capacities differ by more than about 20%, that's not normal variation — it means at least one installer isn't doing a real calculation. This is exactly the kind of red flag I set out in how to vet a heat pump installer, and it's worth reading alongside this piece because sizing method and installer competence are basically the same question asked two ways.

One more thing worth checking, particularly in the US: confirm whether the load calc used is genuinely Manual J compliant (ACCA's method) rather than a manufacturer's simplified "quick load" tool that skips steps. Some of those shortcuts are fine for a rough budget estimate, but shouldn't be the basis for the equipment you actually buy. The ENERGY STAR heating and cooling equipment guidance is a reasonable independent reference point if you want to sanity-check what your installer is telling you against a non-sales source.

Where the extra cost of a proper calc actually pays off

A full load calculation from an independent assessor or a diligent installer typically adds a modest fee to the process, sometimes bundled into the quote and sometimes billed separately. On a system that will run for a decade or more and represents a meaningful chunk of your household electricity use — heating, cooling and often hot water combined — that's a small amount to spend for the certainty that you're not paying to run an oversized compressor for the next fifteen years, or living with an undersized one that can't cope on the one week a year it actually matters.

My own bias, for what it's worth: I think the industry undersells this step because it's genuinely inconvenient to sell. It's slower, it costs the installer time upfront, and it sometimes tells the customer their preferred brand or model doesn't actually fit their load — which isn't a fun conversation on either side of the table. But it's the difference between a heat pump that performs the way its spec sheet claims, and one that just sits there working harder than it should, quietly, for years.