The 8-Point Insulation Submittal Checklist I Run Before Any Order Ships

2026-09-11 Emilia Novak

I run quality and brand compliance for a building materials distributor. Everything that leaves the warehouse gets reviewed by me first — submittals, first-article samples, pallet labels, the works. Roughly 200 items a year. Maybe 220, I'd have to check the log.

Last year I rejected about 30% of first deliveries. Almost none of them were bad products. They were the wrong product, or the right product with mismatched paperwork, or the right paperwork with a number that didn't trace back to a test report.

This checklist is for anyone ordering insulation in volume — mineral wool, rigid board, spray foam — and anyone who signs off on a submittal package before it goes to a GC or a building department. Eight steps. About 90 minutes the first time, 20 minutes once you've built the template.

Before you start, get four things on the table

The current data sheet (not a catalog page), the quote, the proposed submittal, and the code section or spec the assembly has to meet. If you don't have all four, stop. Half the rejections I write up come from comparing a quote to a data sheet that was two revisions old.

The 8-point check

1. Match the product code character-for-character

"Mineral wool batt" isn't a product. It's a family that covers R-15 at 3.5 inches, R-23 at 5.5 inches, three densities, and a couple of facings. Same story on the rigid side — if you're evaluating a rigid foam board supplier, a 15 psi and a 25 psi board often share a catalog name and use different codes.

Codes drift between documents. A rep quotes one code, the submittal gets written from the family name, the PO gets typed by someone in a hurry. By the time it reaches the plant, nobody's sure what's being made.

Checkpoint: the code on the quote, the submittal, the PO, and the pallet label match exactly. Not "close enough." Not "the newer version." Exact.

2. Find the test method behind every number

A number without a test method is marketing. Every performance claim on a data sheet should have an ASTM standard sitting next to it, and you should be able to ask for the report.

  • Thermal resistance → ASTM C518 (or C177)
  • Surface burning → ASTM E84 / UL 723. Class A means flame spread index ≤ 25 and smoke developed index ≤ 450
  • Non-combustibility → ASTM E136. Completely different claim from Class A
  • Sound absorption → ASTM C423, reported as NRC
  • Compressive strength on rigid board → ASTM C165, with grades defined in the applicable product standard

The mix-up I see most: "Class A" and "non-combustible" used as if they're the same thing. They aren't. Class A is a surface burning characteristic. Non-combustible is a material property. Stone wool passes ASTM E136. Most foam and fiberglass products don't, and that's usually fine — they're Class A and they meet code in a lot of assemblies. But if your spec calls for non-combustible and the submittal says Class A, that's a rejection, not a conversation.

One more thing: no insulation is "fireproof." If that word shows up in a submittal, whoever wrote it didn't read their own data sheet.

3. Recalculate R-value at the thickness you're actually installing

R-value per inch isn't a constant down a product line. Stone wool batts land around R-4.2 per inch at 3.5 inches and a bit less per inch as you go thicker. Rigid foam per inch is higher, and polyiso's published number is measured at a mean temperature your wall cavity isn't sitting at for most of the year.

So do the multiplication yourself. Take the per-inch figure off the data sheet, multiply by installed thickness, compare to the code requirement for your climate zone. Don't accept the catalog R-value that belongs to a different thickness — that's a two-minute mistake that turns into a failed inspection.

Then check density. Two products with identical R-value can differ a lot in density, and density is what drives sound performance, sag resistance, and how the material behaves after ten years of vibration.

If you're pricing spray foam insulation wholesale, the comparison gets harder. Spray foam quotes usually bundle material with application labor, and the material number moves with resin pricing. Ask for a line-item split — board feet of material, application rate, labor. Without that split you're comparing a material price to an installed price, and the cheaper-looking number is usually the one that's missing something.

4. Check the fire language against what the code actually asks for

Three different questions get collapsed into one, and they have three different answers:

  1. Does the material need to be non-combustible (ASTM E136)?
  2. Does the assembly need a Class A surface burning rating (ASTM E84)?
  3. Does the wall assembly need to pass NFPA 285, or hold a fire-resistance rating per ASTM E119?

A product can satisfy one and not the others. Foam insulation in certain above-grade exterior wall assemblies triggers NFPA 285 testing — that's not a mark against foam, it's a design constraint, and if you catch it late you're rebuilding the assembly. Stone wool generally has an easier path in fire-rated and non-combustible assemblies, but "generally" isn't a spec. Check the assembly, not the product category.

5. Verify environmental claims against the FTC Green Guides

Per FTC Green Guides (16 CFR Part 260), a "recyclable" claim requires that at least 60% of consumers have access to recycling facilities for that material. Recycled content claims have to be specific and substantiated. And per FTC advertising guidance, every claim has to be truthful, not misleading, and backed by evidence.

So when a thermal insulation manufacturer tells you their product contains recycled content — and plenty of them do — ask for the percentage and the verification behind it. If they can't hand you the report, the claim isn't real yet. It might be true. But it isn't enforceable, and it's not something you want to repeat in a spec package.

6. Require lot code and plant ID on every pallet

We didn't have a formal lot-tracking requirement until 2022. We do now because of a batch of rigid board where compressive strength was visibly off — about 12% under the grade on the label, measured per ASTM C165. Tolerance for that grade runs ±10%. The vendor said it was "within industry standard." We rejected the batch, they refabricated at their cost, and a project slipped nine days.

Without lot codes I couldn't tell you which of their three plants it came from, or which other orders pulled off the same run. That's the part that hurt. Not the nine days — the not-knowing.

Checkpoint: lot code and plant ID printed on the pallet label and captured on the receiving document. If a supplier won't do it, that tells you something.

7. Run a blind sample test before you switch suppliers

I ran one with our field crew: two boards, same spec, same thickness. One from the incumbent, one from a candidate. Nobody knew which was which. Only 22% could guess correctly. That's not the interesting part.

The interesting part was that when I asked which one they'd rather install, most of them picked the same board — and it wasn't the one we'd been buying. That's a real signal about handling, cutting, and dust. I'd been reading data sheets for two weeks and the field crew settled it in ten minutes.

Caveat: easier to install doesn't automatically mean better for your project. The blind test tells you whether a real difference exists. It doesn't tell you whether that difference matters for your assembly. Two separate questions.

8. Put the substitution clause in the PO

Last step, most commonly skipped. It's one paragraph.

No substitution of product code, density, thickness, or manufacturer is permitted without written approval from the buyer at least 10 business days before shipment. Unapproved substitutions are grounds for rejection at the supplier's expense.

Saved $800 by skipping this on a first run once, figuring "it's a standard product, what could change." What changed was the facing. We spent about $6,400 on rework and a nine-day delay. Haven't skipped it since.

Common mistakes, in order of how often I see them

  • Accepting a submittal without the test reports attached. The data sheet is a summary. The report is the evidence.
  • Approving the sample instead of the production lot. Samples get hand-picked. Nobody hand-picks pallet 14.
  • Comparing quotes by price per square foot without matching R-value and thickness. That's not a comparison, that's a coin flip.
  • Assuming a one-stop shop is automatically better. The vendor who told me "this isn't our strength, here's who does it better" got the next three orders without a bid. The one who said yes to everything we asked has been wrong on specs twice.

Is rockwool insulation good? It depends on the assembly

I get asked this a lot, so here's the honest version.

For fire-rated assemblies, acoustic separation, and any cavity where non-combustibility is written into the code, stone wool is usually the simplest answer. It's non-combustible per ASTM E136, it doesn't lose performance when it gets damp, and it cuts clean. For sound, the density does work that lightweight products can't.

But if you're trying to hit R-21 in a 2x4 cavity, mineral wool won't get you there. Rigid board or spray foam will. Same for a tight rim-joist cavity where R-per-inch is the binding constraint. That's not a knock on stone wool — it's arithmetic.

A supplier who tells you where their product stops being the right answer is a supplier you can trust on the orders where it is. At least, that's been my experience on the distribution side.

Emilia Novak

Emilia Novak

Emilia Novak is a flooring and architectural-surfaces analyst covering ceramic and porcelain tile, natural stone, resilient flooring, underlayments, countertops, adhesives, grout, and installation accessories. She uses ASTM C373 and ASTM C648 test evidence while comparing water absorption, breaking strength, slab flatness, substrate moisture, joint width, slip resistance, and installed tolerances. Her specification guides help architects, contractors, and buyers match surface systems to traffic, wet-area exposure, maintenance demands, and substrate conditions.