Structural Steel Is Mild Steel—Until It Isn't: The Material Spec Assumptions That Cost Me $7,000

A procurement lead's account of six documented material-spec mistakes—from long steel pipe orders to aluminum rims and pipe breaks—and the two-hour checklist that now saves his team thousands. If you're still treating 'structural steel' and 'similar alloy' as enough detail, this one's for you.

In my role handling structural steel and metal material orders for 9 years, I've personally made—and documented—six significant mistakes, totaling roughly $18,000 in wasted budget. Now I maintain our team's material pre-check list, the one I should've built in 2016. This piece is the article I wish someone had handed me back then.

My position: spec assumptions are the most expensive line on the order

If you're specifying steel or aluminum by "what we used last time" instead of the actual grade and temper on the drawing, you're not saving time. You're pre-paying for rework.

That's the whole argument. Everything below is the evidence. And yes, I came to it the hard way—by being the guy who approved the wrong material because the drawing was "close enough" and I was in a hurry.

Here's the thing most procurement teams won't say out loud: the expensive mistakes in this business rarely come from bad mills or bad suppliers. They come from a two-word assumption on an order sheet. Structural steel. Similar alloy. Standard grade. Those phrases feel precise right up until the day they're not.

Argument 1: "Structural steel" is a category, not a grade

In September 2021, we ran a batch of long steel pipe for a warehouse frame. The client's spec sheet said "structural steel," and I read that as a green light. We pulled mild steel from stock—same as the previous project—cut 40 lengths, welded them, and moved on.

Then the structural engineer's load check came back. Two of the heavy-load sections were under-yielding. The material was fine for a light-span roof. It was not fine where we'd used it. We scrapped and recut the affected sections at $2,900 in material and labor, plus an 11-day schedule slip that the client was kind enough to accept without a penalty clause—that time.

So to answer the question I get at least twice a month: is structural steel mild steel? Sometimes. Often, even. But "often" is not a specification.

ASTM A36/A36M covers carbon structural steel with a minimum yield point of 36 ksi (250 MPa). ASTM A572 Gr. 50 gives you 50 ksi. They are both legitimately called structural steel, and they don't behave the same under load. Per ASTM A36/A36M-19—verify the current edition and confirm which standard your project actually requires.

The fix sounds embarrassingly simple in hindsight: the drawing has to say the grade. If it doesn't, you ask. You don't assume. I knew that in theory. I just didn't apply it because we'd done it the other way a dozen times and nothing had broken yet.

That's the trap. Familiarity is not a substitute for verification.

Argument 2: aluminum hardness is a spec, not a vibe

The aluminum side of the business taught me the same lesson in a different shape. We had an order for aluminum rims used on industrial casters and light trailer assemblies. The spec named 6061-T6. We had 6063-T5 in stock from a previous job, and I asked the client whether "similar" was acceptable. He said sure.

Here's where aluminum hardness stops being a footnote. Typical Brinell values run around 95 HB for 6061-T6 and roughly 60 HB for 6063-T5 (typical values per ASM Handbook, Vol. 2, 10th ed.; actual values vary by temper and supplier). On a rim that only ever rolls across a smooth warehouse floor, the softer alloy survives fine. On one that takes impact loads—a hit curb, a dropped pallet—it deforms, and then it doesn't.

We replaced 18 of 24 units. The client's exact words were: "You said similar." He wasn't wrong to say it. It wasn't the same, though, and the difference showed up as a pipe break at the axle sleeve on two of the failed units—the kind of failure that gets traced back to the spec sheet in a very uncomfortable meeting. A $3,200 order became a $4,100 redo.

One thing I've learned to say and mean: the mill name doesn't save you if the temper on the drawing is wrong. A premium Alcoa aluminum rims extrusion is still the wrong material for the job if the drawing called for a harder temper and someone substituted "close enough." I've had a pipe break at a weld seam on an Alcoa-spec tube, not because the mill did anything wrong, but because the service load didn't match the spec we ordered to. Good material, wrong application—same expensive result.

Argument 3: the efficiency math is lopsided, and that's the point

I know this argument sounds backward. Two extra hours reviewing material specs before an order goes out feels like slowing down. It isn't. It's the fastest thing you'll do all week.

Here's the scale we're talking about. In the past 18 months, our pre-check list has flagged 47 potential spec errors. Twelve of them were real—wrong grade, wrong temper, wrong wall thickness, wrong standard. If each had averaged even $900 to fix, that's close to $11,000 avoided from a one-page checklist. The math isn't subtle. The two-hour review consistently beats the two-week rework.

I also have the low-cost horror story, because everyone needs one. In Q2 2024, I skipped third-party material certs on a small batch of pipe fittings to save about $80 in inspection fees. One fitting failed hydro-test and took three others with it. Replacement and rush freight ran north of $600. Saved $80. Spent $600. Classic.

The pushback I always get

"For a small order, this is overkill."

I used to think so. Then a $320 order of pipe fittings went out with Sch 40 instead of Sch 80 because nobody looked twice at a batch that small. Ten pieces turned into a two-week argument with the client and a reorder. Small orders are exactly where assumptions hide, because nobody assigns anyone to watch them.

Second pushback: "You can't verify every bolt." Correct. You can't. You're not building a QA department for a freight order. You need one page, six checkboxes, and the discipline to actually look at it before the PO goes out:

  • Grade and standard named explicitly (A36 vs. A572 Gr. 50 vs. A992)
  • Alloy, temper, and hardness range on aluminum parts
  • Wall thickness and schedule confirmed against the drawing
  • Service load or application noted, even briefly
  • Cert requirement confirmed in writing
  • One person signs off who did not build the quote

That's it. No software subscription, no consultant. A page and a habit.

Restating the point, without the hedge

Spec literacy is not paperwork. It's the highest-leverage cost control you have.

Whether it's a long steel pipe order, a batch of aluminum rims, or a warehouse frame, the two hours you spend confirming is structural steel mild steel for this application and what surface hardness does this part actually need will pay for themselves every single time. I know that because I've paid the other way—six times, about $18,000, and one 11-day delay I'd rather forget.

I learned it the expensive way. You don't have to.

Lucas Tremblay

Lucas Tremblay

Lucas Tremblay is an independent aluminum products analyst covering aluminum sheet, plate, coil, extrusion, profiles, bar, tube, and heat-sink stock. He applies ASTM B209/B209M and B221 requirements while comparing alloy and temper, tensile strength, elongation, conductivity, thickness tolerance, profile geometry, straightness, surface finish, and joining constraints. His practical guides help designers, fabricators, and sourcing teams select flat or extruded products for structural, thermal, transport, architectural, and general manufacturing applications.