Specification desk: +1-800-555-0241 · Mon–Fri 8am–6pm ET ISO 9001:2015 · LEED Compliant Products · FM Approved

10 Checks Before You Approve a Prefab Steel Warehouse Package (I Learned These the Expensive Way)

2026-10-16 · Celeste Fournier

I handle steel and envelope package orders for building products suppliers — prefab steel warehouses, pre-engineered frames, purlins, roof trusses — and I've been doing it for 11 years. In that time I've personally made, and written down, nine significant mistakes on steel packages, totaling roughly $60,000 in re-fabrication, change orders, and one air-freight shipment I still don't like talking about. Now I maintain our team's pre-release checklist. This is that checklist, minus the internal jargon.

Who this is for (and how to use it)

This is for the person who has to sign off on a prefab steel warehouse package — supplier side or buyer side — before the shop drawings get released. Roughly 1,500 to 20,000 sq ft, purlins plus some form of truss or space frame, and a deadline that someone already promised to someone else.

Work through it in order. Ten checks. Checks 2, 4, 6 and 8 are the ones people skip, and three of my nine mistakes came from skipping them.

One boundary note up front: my experience is based on about 400 packages, mostly US commercial and light-industrial. If you're working on heavy industrial, cold storage, or anything with real process loads, some of these priorities will shift.

1. Decide clear span vs. multi-span before anyone quotes

This sounds obvious, and it's still the thing that gets re-quoted most often. Clear span means no interior columns; multi-span means columns down the middle. Multi-span is almost always cheaper per square foot. It's also useless if you're installing a 30-foot racking system with a reach truck that needs a 40-foot clear aisle.

What I mean is: don't check the span against "we need about 30 feet of clear height." Check it against the tallest thing that will ever live under that frame — racking plus lift height plus sprinkler head plus the 12 inches nobody budgets for. If there's any chance of a mezzanine, a crane, or a racking change in year three, say it now. I've watched that conversation happen after deposit twice, and both times it ended with a credit request.

2. Fix the roof slope and the drainage path, including where the water ends up

Slope gets decided early and then quietly forgotten. Standing seam and through-fastened panels have different practical minimums, and the manufacturer's published minimum is a starting point, not a design. If your building is long and low, the real question isn't the slope anyway — it's where the water goes at the end of the run. A gutter, a scupper, a downspout, or the neighbor's property line. Pick one before the frame is fabricated.

Do the arithmetic on movement: steel expands and contracts at roughly 6.5 × 10⁻⁶ in/in per °F. A 300-foot roof run moving through a 100 °F swing wants to travel about 2.3 inches. Dark colors and insulated panel systems make that worse, not better. That's why long runs get expansion joints — and why the joint locations have to be on the drawing before panels are cut, not after. Ask your panel supplier for their current expansion joint guidance as of the date you order, because the numbers move with panel profile and finish.

3. Choose your purlin type before you finalize the framing

Types of purlins come in a few flavors. Cold-formed C and Z are the workhorses; hot-rolled angles and channels show up on heavier or older work. C-purlins are simple-span, easy to handle, and the sensible default for short bays, hips, valleys, and low-slope roofs where lapping is awkward. Z-purlins let you lap and nest at the supports, and a properly lapped run behaves much closer to a continuous beam than to a simple span.

It's tempting to think a purlin is a purlin and just accept whatever the frame supplier put in the quote. But gauge, depth, stiffened versus unstiffened flanges, bearing legs, and lap length all change what the roof actually carries — and they change again if you switch suppliers mid-quote, because a 10-inch Z from one manufacturer is not a drop-in for a 10-inch Z from another. Cold-formed member design here follows AISI S100, and your supplier's engineering has to run it against your loads.

AISI S100, North American Specification for the Design of Cold-Formed Steel Structural Members, is the reference document for cold-formed purlin and girt design. Your supplier should be able to tell you which edition they're designing to. Ask. If they can't answer that question, that's your answer.

The "always use Z-purlins" thinking comes from an era when lapped systems were the only economical way to reach long spans. That's changed. With modern standing seam profiles and shorter bays, a well-detailed C-purlin job can come in cheaper and erect faster, especially on roofs with a lot of penetrations.

4. Check purlin spacing against the actual panel, and against collateral loads

Purlin spacing usually lands between 4 and 8 feet on center, with 5 feet being the number most of us reach for out of habit. The number that actually matters is what your specific roof panel allows at your specific wind and snow loads. That's the panel manufacturer's span table, not the frame supplier's default.

Then add the loads nobody puts on the first drawing: rooftop units, curb and duct supports, sprinkler mains, walkway pads, conduit runs, and — increasingly — solar. On a 9,400 sq ft package in 2023, we froze purlin sizes and then received a mechanical drawing showing eleven RTUs and a full sprinkler grid. Change order: $16,800. Fabrication pushed back two weeks. Solar is the one catching people now, because a ballasted array adds dead load and it changes your wind uplift profile at the same time.

If you don't have the mechanical drawing yet, put a written placeholder in the scope: "purlin sizing to be verified against final collateral loads, allowance of X psf." Get it signed. A verbal "we'll figure that out later" is worth exactly nothing at the fabrication release meeting.

5. Space frame, prefabricated roof truss, or rigid frame — decide on the column grid, not the price

I went back and forth between a space frame and shop-welded prefabricated steel roof trusses for about three weeks on a distribution center. The space frame handled the irregular column grid and the long clear span cleanly. The trusses were cheaper to fabricate and far easier to ship. I chose the space frame because the column grid was fixed by the site, and I'd rather pay for the right geometry than fight the wrong one for six months.

The rough rule: a rigid frame is the cheapest answer up to moderate spans with a regular grid. Prefabricated steel roof trusses make sense when you need longer spans and fewer columns and the pieces fit on a trailer. Space frames — spaceframe systems with ball or plate nodes and bolted tube members — come into their own for very long column-free spans, irregular column layouts, and roofs where the structure is meant to be seen.

What the space frame brochures don't tell you is that the tolerance stack is unforgiving. Node positioning errors accumulate across a bay, and reaming connections at height is miserable work. In 2022 we hit a node mismatch of about 14 mm on a corner bay that cost two days and a specialist crew to resolve. Not a design failure — an erection sequence and tolerance conversation we hadn't had early enough. Ask in writing what the fabrication and erection tolerance is, and who bears the cost when it's exceeded.

MBMA member manufacturers are required to hold IAS AC472 accreditation, which covers the quality control system for metal building system fabrication. If you're buying a pre-engineered package, that's a reasonable minimum filter and it's publicly verifiable — just check the current status directly with IAS, since accreditation lapses.

6. Walk the knee bracing clearance with the door schedule in front of you

Knee bracing construction is a diagonal member tying the column to the rafter near the eave. It's usually there to stiffen the frame, often paired with pinned bases, where the frame takes its lateral stiffness from the bracing rather than from a fully rigid knee. It's efficient. It also occupies the top corner of your clear space, and it is almost always discovered too late.

The check: open the door schedule and the elevations side by side. Door tracks, overhead door springs, mezzanine decks, and anything that needs full height at the sidewall can all foul a knee brace. In September 2021 we found six roll-up doors on a package that couldn't open fully because the knee braces sat exactly where the tracks needed to run. Field rework was $4,100, and the customer watched all of it. Now the door schedule is check six, every time.

If the brace conflicts, the options are usually to move the door, change the frame to a rigid moment connection, or add a wind column at the opening. Each carries a cost and a lead time. Get the answer before fabrication.

7. Freeze the anchor bolt layout and the column reactions early

The foundation contractor will start pouring before your building data is final unless someone stops them. Anchor bolt layout, projection, base plate size, and design reactions all have to be frozen and stamped — and frozen against the final loads, including whatever you added in check 4.

Bolts set in the wrong place are one of the few errors a change order can't fix. They get fixed with a drill and a welder, and depending on who's inspecting, sometimes not even that. Get the reactions in a signed, dated document and hand it to the foundation contractor in writing, with a read receipt.

8. Confirm the bracing scheme, and confirm who supplies it

Rod bracing, cable bracing, tension-only versus compression-capable — and then the parts that always go missing: sag rods, bridging, and the turnbuckles that let crews take up slack. On cold-formed purlins, bridging or sag rods are typically required at intervals to keep the compression flange from rolling, and the purlin supplier's design will specify how many rows.

This is more of a scope-trap check than a technical one. "Bracing by others" on a drawings package means it's yours until someone proves otherwise. Budget $2,000 to $6,000 and a week of confusion for every line item you can't attribute to a named supplier.

Also confirm the bracing is compatible with the roof panel system. A brace rod that lands where a standing seam clip has to go is a small clash that turns into a large argument.

9. Match the shipping and erection sequence to the crane you actually have

The quote assumes a crane and a crew that nobody has priced. Check bundle weights, maximum piece length, and trailer configuration against the site. Over-length permits take time. Long space frame members and long trusses both need checking against route restrictions, not just against whatever trailer you can rent.

Then walk the erection sequence. Where does the crane sit? How does it move? Can it reach the far bay without a second setup? Can the site take the outrigger loads? A fabricator told me once that the trusses would ship in six weeks. Did I believe it? Roughly 70 percent, which is why we built a buffer into the schedule. Space frames in particular usually need temporary support towers and a defined erection order, and none of that appears on the sales drawing.

10. Write down who owns the risk — and buy delivery certainty if the date is real

Last check, and the one that has saved me the most money. Before you release, put in writing who pays for: re-fabrication due to a design change you made, re-fabrication due to a supplier error, field modification, freight delays, and erection standby time. That paragraph is worth more than any discount you'll negotiate.

And then there's the date itself. In March 2024 we paid $1,900 for expedited fabrication and committed freight on a package where the alternative was missing a tenant's move-in date. On paper that looked expensive. The alternative was a delay claim several times that size, plus a customer who would have re-bid the next building without calling us. (I should add: that expedite fee was on top of normal freight, not instead of it. Budget for both.)

Rush fees buy certainty, not just speed. "Probably on time" isn't a schedule — it's a wish with a purchase order stapled to it. At least, that's been my experience with deadline-critical packages. If your schedule is genuinely soft for a month, take the cheaper route and sleep fine. If it isn't, pay for the date.

What I check twice

Five things, roughly in order of how often they bite:

  • Door and opening heights, checked against knee brace locations and purlin depths.
  • Any load that arrived after the framing was sized — mechanical, sprinkler, solar, a future mezzanine.
  • Bracing and bridging scope, in writing, with a named supplier.
  • Anchor bolt layout against the final, frozen reactions.
  • Who pays when something is wrong. That one is actually first.

One more thing. None of these checks requires special software or a bigger team. It requires about three hours — or rather, three hours on a simple job and most of a day on a complicated one — plus the willingness to call the mechanical engineer before the shop drawings release instead of after. Three hours, across forty packages. That's what would have saved me most of my $60,000.

Celeste Fournier
Celeste Fournier

Celeste Fournier is an interior building-products analyst covering wall panels, gypsum board, acoustic ceilings, decorative moulding, wallpaper systems, architectural coatings, sealants, and finishing accessories. She applies ASTM C423 sound-absorption results and ASTM E84 surface-burning data while assessing NRC, flame-spread index, smoke-developed index, dry-film thickness, adhesion, joint movement, cleanability, and finish consistency. Her technical articles help designers, contractors, and facility teams coordinate appearance, acoustics, fire-performance documentation, application conditions, and long-term maintenance.