What actually drives the cost of a steel building

By Johnny · 23 October 2026 · 7 min read

Owners usually ask for a price per square metre, expect one number, and receive five. It looks like suppliers are hiding something. Most of the time they are not: the five numbers answer five different questions, because the brief was never specific enough to fix one answer.

If you understand what moves the number, you can control four of the five drivers yourself — and you can tell whether a quote is genuinely cheaper or just built on different assumptions.

The five drivers, in order of effect

1. Span

For a simply supported member, bending demand rises with the square of the span. Double the span and the load effect is roughly four times greater, not two. The member therefore does not just get longer — it gets deeper and heavier per metre, so the increase compounds.

Span also reaches into the foundation. A wider frame pushes harder at the column base, and overturning and uplift are resisted by concrete and ground, not by steel. On weak ground this is where an apparently modest span change becomes a large number.

This is the one driver worth arguing about with your own layout drawing. A single internal column on a rack line often costs nothing in usable area and removes a lot of steel.

2. Height

Eave height changes two things at once. It lengthens the column, which reduces its resistance to buckling, and it increases the wall area that wind acts on. Both push section sizes up. Height is also what people revise late, because it is set by machines, trucks and stack heights that arrive after the building has been ordered.

3. The loads you declare

This is the least supplied information in most enquiries, and it is the one that makes quotes incomparable. It covers:

• The design wind speed for the site, in km/h, taken from the local code.
• Whether a crane is coming — now or in five years — and what it must lift.
• What the roof carries: nothing, solar, a suspended conveyor, services, maintenance access.
• Mezzanine or future floor loads, even if the floor is a later phase.
• Seismic requirement, where the region has one.

Two suppliers given different wind assumptions will produce two correct designs at two different prices. The cheaper one is not a bargain; it is a building designed for a gentler site.

4. The envelope

Cladding, insulation, roof build-up and the number of openings. A single-skin roof and an insulated sandwich panel roof are different products at different prices, and neither is wrong — they serve different uses. Cold stores, food handling, and anything with humidity control cannot use the cheap option, and pretending otherwise only moves the cost to the operation.

5. Getting it there and standing it up

On export work this is frequently the decisive line, and it is the one owners never see in a per-square-metre comparison. What matters:

• Member segmentation. Parts must fit the container and be liftable by the crane that will actually be on site. Design decisions made for factory efficiency can be wrong for a site with a small crane and a narrow access road.
• Distance and port. Inland haulage from a landlocked site can exceed the sea freight.
• Local erection capability. A design that assumes a large crane and a skilled crew is not cheap if neither exists locally.
• Ground and foundations. Often built by someone else, and often the source of the delays that cost more than the steel.

Where cost can come out without weakening the building

There are legitimate savings and illegitimate ones. Legitimate:

• Re-examine the span against the real layout, not the drawing habit.
• Fix the height from the machine or the truck, then stop changing it.
• Declare the loads properly once, so nobody prices a guess.
• Match the envelope to the use — do not insulate a store that never needed it, and never economise on a cold store.
• Design the segmentation for the site, not only for the container.
• Build the crane bay now if a crane is likely later. Retrofitting a crane changes columns, rail beam and usually foundations.

Illegitimate savings are the ones that quietly transfer risk to you: thinner cladding that dents during erection, omitted bracing, drainage sized for a gentler climate, an assumption of ground conditions nobody tested, and a scope that excludes freight, foundations or erection without saying so plainly.

The mistake I see most

Comparing quotes by total number instead of by scope. Two figures that differ by a wide margin usually differ because one of them does not include the same things — not because one supplier is generous. Before you compare anything, put the same five drivers in front of every supplier and ask each one to state what is excluded.

If you have a span, a height and a wind speed, you already have enough for a comparable conversation. Without them, you are collecting prices for five different buildings.