Steel Buildings Suppliers Have Evolved. Your Procurement Model Hasn't.

"Steel structure" stopped being a useful spec years ago. Most buyers haven't noticed.

I've priced more than 200 rush jobs for steel frame buildings over the last few years — prefabricated aircraft hangars, exhibition halls, factory and warehouse shells. Nearly every one of them had the same origin story: the buyer went to a steel buildings supplier with an RFP written around tonnage, signed a contract, and then found out the building they were getting wasn't the building they thought they were buying.

The problem wasn't the factory. It was the RFP.

The industry has changed. The old model — design commissioned separately, drawings sent to a fabricator, a local crew hired to erect — is still baked into most RFPs. But the suppliers who can actually deliver a large-span building without a change-order storm don't work that way anymore. They design, fabricate, and erect under a single contract. That's the new baseline. The procurement model most buyers are still using was designed for a market that no longer exists.

What was best practice in 2018 is a coordination risk in 2025. That's the gap I'm writing about.

Point 1: "Light steel frame" and "steel frame residential building" are two different engineering languages

These two phrases get used interchangeably in RFQs. They shouldn't. Light steel framing — cold-formed C and Z sections, typically under 0.5-inch wall thickness — runs on a completely different structural logic than a heavy steel frame with hot-rolled H-beams and box columns. A lightweight steel structure isn't a cheaper version of a heavy frame. It's a different building type, with different spans, connection details, and foundation demands.

I still kick myself over a project in March 2024. A client sent an inquiry for "steel frame residential building, standard size." We quoted light gauge framing at 4.5m spans, which is what we'd built for comparable villa projects. Their engineer had specified hot-rolled H-beams at 6m clear span. Same words. Two different buildings. We discovered it when the first delivery arrived and the column base plates wouldn't line up with the anchor bolts. €28,000 in rework and three weeks of delay. I said "standard size." They heard "standard size." The mismatch cost both of us.

The fix isn't complicated: the RFP needs to state the structural system, the maximum span, and the load assumptions before anyone prices it. If a steel buildings supplier can't tell you which system they're quoting, that's your first red flag.

Point 2: Large spans make design, fabrication, and erection inseparable

This is where the old procurement model really breaks down. An exhibition building in Malaysia and a prefabricated aircraft hangar share one thing: they're driven by clear span. A 60m clear span hangar leaves almost no room for fabrication tolerance — if a splice plate is 40mm over spec, the rigging crew spends two days making it fit on site instead of four hours. At that scale, "who designed the connection detail, who cut the plate, and who's responsible when the erector finds a mismatch" is the whole project.

In my experience triaging emergency turnarounds, the projects that go sideways are almost never the ones where a single supplier owns design through erection. They're the ones where the fabricator says "we built to drawing" and the erector says "the drawing doesn't match the site." At 40m span, that's not a dispute. That's a stop-work order.

The industry moved toward EPC turnkey for a reason. When design, fabrication, and erection sit under one contract, the tolerances are set by the same team that has to hit them in the field. Per EN 1090-2 (the CEN standard for execution of steel structures), large-span and dynamically loaded projects typically require execution class EXC2 or above — and that's not just a certificate. It means weld traceability, documented NDT, and dimensional QC records. Those processes don't travel well across three separate contracts.

Point 3: Filter suppliers on qualification before you sort on price

Let me be blunt. If you're sorting quotes for a prefabricated aircraft hangar or a large exhibition building by lowest price, you're not buying a building — you're buying the cheapest interpretation of an incomplete document.

EN 1090 (EXC2 minimum for structural work, often EXC3 for large span or dynamic loading) and Malaysia's CIDB requirements for Construction Product Approval (CPA) and IBS certification aren't marketing badges. They're the difference between a fabricator who maintains weld traceability, calibrated NDT, and written dimensional QC — and one who cuts to the drawing, loads the truck, and wishes you luck on site. When I'm triaging a rush order, the first thing I check is cert status. Not price. A non-certified supplier can always beat a certified one on quote. They just can't deliver the same building.

The upside of the non-certified quote is typically 12–18% off the steel package. The risk is an erection-stage spec failure. I keep asking clients the same question: is 15% worth two weeks of site delay? On anything more complex than a simple shed, the math rarely works.

The pattern is consistent. The clients who call me in a panic are almost always the ones who bought on quote, not on qualification. The cost of the panic is always higher than the cost of the cert.

The pushback: isn't split procurement more flexible?

I get it. To be fair, splitting design, fabrication, and erection gives the client more control at each stage, and in markets like Indonesia and Malaysia, local contractor relationships are real assets you don't want to drop. If you have an in-house engineering team and a trusted local erector, going turnkey with an overseas supplier might not be your best fit. EPC pricing isn't always transparent, and integration carries a premium.

But here's the thing — the coordination cost of the old model is going up, not down. Ten years ago, a 5% discrepancy between design and fabrication got resolved over a phone call. Today, with multi-country supply chains and site schedules measured in days, the same discrepancy stops the rigging crew. The turnkey premium isn't a luxury. It's the price of not eating that coordination risk onsite.

Granted, not every project needs EPC. I'm saying: when you compare the flexibility of split procurement, price it with the coordination cost included. Otherwise you're comparing the wrong numbers.

Restating the point: the problem is the procurement model, not the factory

So no — the problem isn't the factory. And it isn't "prefabrication" as a concept. The industry's move toward integrated design-fabricate-erect is exactly what's made 60m spans and 24-week hangar schedules possible. What's out of date is the RFP that still treats a steel building like a pile of steel.

What used to be best practice — client-owned design, fabrication priced as a line item, erection tendered separately — is now a coordination liability. The suppliers moving to a single-contract model aren't selling a bundled service. They're the ones who can stand behind a building that gets assembled once in the field, on a schedule with no room for a second attempt.

If you're still writing RFPs by tonnage and expecting a building on the other side, that's the gap you need to close.