-
Background: A Good Project, A Deadline I Didn't Push Back On
-
The Decision I Made In Two Hours
- What Went Wrong (Three Things, Not One)
-
The Result
-
What I Do Now: Six Questions Before Any Plywood OEM Order
-
1. Ask for the certificate scope, not the certificate
-
2. Run formaldehyde classification against the destination market
-
3. Get thickness tolerance and edge squareness in writing — from the mill, not the distributor
-
4. Ask who actually makes the panel
-
5. Request an EPD when sustainability claims are on the table
-
6. Buy sample panels and measure them yourself
-
1. Ask for the certificate scope, not the certificate
-
What Changed, And What Didn't
March 14, 2023, 4:58 p.m. I was still at my desk with a cup of coffee that had gone cold two hours earlier, staring at a photo from the site. Six hundred and forty plywood panels, freshly unwrapped, stacked on a loading dock in the rain. Every single one of them had the wrong edge profile.
I'm a materials and panel sourcing lead at hks-architects. I've been handling panel specifications and plywood OEM orders for eight years now. I've personally made — and written down — six significant sourcing mistakes that together cost roughly $74,000 in wasted budget and rework. The photo from that day is the reason our team now runs a 14-point pre-order checklist.
This is what happened, and what I'd do differently.
Background: A Good Project, A Deadline I Didn't Push Back On
The project was a 96-unit modular affordable housing development — one of our hks architects sustainable architecture designs that leans hard on prefabricated panel systems. Cross-laminated timber frame, fiber-cement board on the wet walls, and about 4,200 sq ft of FSC-certified plywood sheathing and interior liner panels.
It's worth saying here that hks architects' sustainable designs tend to be panel-heavy almost by default. Prefab systems are how we hit both the carbon targets and the construction schedule. That also means a panel problem doesn't stay a panel problem for long — it becomes a crane problem, a crew problem, and a client problem.
We'd specified the plywood as an OEM private-label product: our fabricator wanted the panels delivered with their own branding and cut list, sourced through us via a building panels distributor we'd used twice before.
Everything about it looked routine. Which, honestly, is how most of my expensive mistakes start.
The Decision I Made In Two Hours
In February 2023, our regular distributor came back and said the mill they'd been using couldn't hit our April window. There'd been a fire on one of their lines in January. So we had to find a new source, and we had roughly two weeks instead of the usual eight.
Had about two hours to decide between two quotes before the tender deadline closed. Normally I'd run a full mill audit — sample panels, thickness tolerance checks, a chain-of-custody document review, the whole thing. There was no time. I went with the cheaper mill based on three things: a price sheet, a sustainability brochure with a lot of green graphics, and a phone call where the sales rep said all the right words.
In hindsight, I should have pushed the timeline out by two weeks. The client would have grumbled. Grumbling is cheaper than rework. I didn't push, and here's what that cost us.
What Went Wrong (Three Things, Not One)
1. The FSC claim didn't cover the product we bought
The brochure said the mill was FSC-certified. That part was true. What the brochure didn't say was that the certification covered their OSB line, not the plywood line we ordered. The plywood came from a sub-tier supplier, and the chain of custody broke right there.
Everything I'd read about supplier vetting said to ask for the certificate. In practice, asking for the certificate is the easy part — the hard part is matching the certificate's product scope and FSC-C code to the exact SKU on your purchase order. It takes about 30 seconds to verify a code at info.fsc.org. I didn't do it.
2. The formaldehyde class wasn't what we specified
Our spec called for panels meeting TSCA Title VI / CARB ATCM Phase 2. The mill shipped E1. For a US-market project, E1 isn't automatically equivalent — CARB ATCM Phase 2 and TSCA Title VI (40 CFR Part 770) were what our spec and the client's indoor air quality target were written against.
The rep said "E1 is basically the same thing." It isn't, and I should have known that. That one line on a spec sheet turned into a full panel-level documentation request the mill couldn't fulfil. Verify current requirements at epa.gov and arb.ca.gov before you write the PO — the laminated-product compliance dates in particular have moved around.
3. Thickness tolerance blew out — and that's what killed it
This is the one nobody warned me about, and it's the one that broke the project.
Most buyers focus on panel grade, surface finish, and price per sheet. They completely miss thickness tolerance and edge squareness — and if your panels feed a CNC-cut prefab line, those two numbers matter more than almost anything else on the datasheet.
The spec was ±0.3 mm. What arrived was closer to ±0.8 mm on roughly 40% of the sheets. We measured 250 of them by hand over two days. The CNC line's vacuum table and tool offsets were calibrated to the tighter number. Panels shifted during the cut, edge profiles came out inconsistent, and the fabricator rejected the whole 640-panel lot.
From the outside, a plywood panel is a plywood panel. The reality is that two mills producing the "same" EN 636 grade can sit half a millimetre apart on thickness — and that half-millimetre is the difference between a working prefab line and a stopped one.
The Result
Here's the damage, in numbers:
- 640 panels rejected and returned — $9,400 in rework, return freight, and re-cut material
- 11 working days of production delay on the modular line
- One very uncomfortable call with the project architect
- About $2,100 in expedited freight to get the replacement lot to site
Total: roughly $11,500, plus a credibility hit that was worse than the money.
We caught the rest of it later. The replacement order went to a mill we'd audited back in 2021 — slower to quote, about 9% more per sheet, and they hit tolerance without a single exception across three deliveries.
What I Do Now: Six Questions Before Any Plywood OEM Order
After the third rejection in Q1 2024 — yes, there was a third, on a fiber-cement order — I built our team's pre-order checklist. Here's the short version. If you're working out how to evaluate building panels manufacturers, start here.
1. Ask for the certificate scope, not the certificate
Match the FSC-C or PEFC code to the exact product line and the exact mill. FSC's public database lets you check this in under a minute. A certificate that exists is not the same as a certificate that covers your SKU.
2. Run formaldehyde classification against the destination market
E1, E0, CARB Phase 2, TSCA Title VI, ULEF, NAUF — these are not interchangeable labels. Write the exact standard and revision into the PO, not just the class name.
3. Get thickness tolerance and edge squareness in writing — from the mill, not the distributor
Specify the grade standard (EN 636-3 for plywood, EN 300 for OSB, EN 312 for particleboard, ASTM C1186 for fiber-cement flat sheets, PS 2-18 for US structural-use panels) and the tolerance you actually need for your downstream process. Those are two separate lines on the spec sheet.
4. Ask who actually makes the panel
On OEM and private-label work, the trading company is often not the mill. Ask for the mill name, location, and production line. If they won't say, that's a red flag.
5. Request an EPD when sustainability claims are on the table
An Environmental Product Declaration carried out per EN 15804 / ISO 14025 with third-party verification is worth more than any brochure. As of early 2025, a growing share of European mills publish them — check the programme operator and the declared unit, not just the logo on page two.
6. Buy sample panels and measure them yourself
Not a photo. A physical panel. Measure thickness at six points per sheet across five sheets. It costs a couple hundred dollars and about a week. It's the cheapest insurance in this business.
What Changed, And What Didn't
Here's the thing I keep coming back to. In 2018, most of what I just described didn't exist in practice for mid-size projects. EPDs were for flagship buildings. Chain-of-custody checks were a signature on a form. Nobody asked for a digital product passport.
Now, with the EU Construction Products Regulation framework evolving and the EU Deforestation Regulation (Regulation (EU) 2023/1115) phasing in — the application date for large operators was pushed to 30 December 2025 at last check, so verify the current status at europa.eu — documentation depth has become a competitive differentiator rather than a formality.
The fundamentals haven't changed. You still need a mill that can hold tolerance and ship on time. But the execution has transformed completely, and the verification tools are better than they've ever been. Everything I needed to avoid March 2023 is publicly available and free. That's the part that annoys me most.
Bottom line: the panels weren't the problem. My process was. Six questions, about 90 minutes of work, would have caught all three failures before the PO went out.
The 14-point checklist we run now has flagged 47 potential issues in the past 18 months. Call it $61,000 in avoided rework. The coffee still goes cold. The panels don't come back wrong anymore.