8 Years of Buying Building Products, 23 Mistakes, and the $186K That Made Me Build a Better Sourcing Checklist

I spent eight years on the buying side of the building products industry before moving into operations at Cornerstone Building Brands. During that time, I signed off on more than 600 purchase orders for windows, exterior doors, siding, rainscreen cladding, and cement board. I also documented 23 of my own significant mistakes. Total waste: roughly $186,000 in materials, expedite fees, and reinstallation labor.

I mention that number because it taught me something that no supplier brochure ever did: most product sourcing advice is conditional, but people hand it out like it's universal. A low-priced aluminum window wholesale deal can be smart for one buyer and a disaster for another. A premium cladding system can be exactly right for one project and overkill for the next.

So if you ask me, “which building products supplier should I use?” my honest answer is: it depends on which of the three buyers you are.

There is no “best” supplier until you know which buyer you are

After years of making mistakes, I've narrowed most of my buyers into three scenarios:

1. The project buyer. You're a contractor or installer purchasing windows, doors, and siding for a specific job.
2. The inventory buyer. You're a distributor or dealer sourcing aluminum windows wholesale, cladding lines, or panel products to stock and resell.
3. The specification buyer. You're an architect, consultant, or owner's rep who writes the requirements somebody else will purchase against.

I've sat in all three chairs. The most expensive mistakes happen when I used the wrong checklist for the wrong chair.

Scenario 1: You're the contractor buying for a specific project

In the fall of 2019, my company placed a $48,000 order for 340 aluminum windows on a mid-rise retrofit. The unit price was good. The performance rating matched the spec. And the product number was correct.

What I didn't verify was the field condition. We ordered based on a drawing set that the general contractor had changed ten days earlier. On one elevation, 26 openings ended up 1.5 inches narrower than the windows we ordered. We could wait six weeks for a replacement run, or we could pay to expedite. We paid the expedite fee — $6,800 — plus $4,900 in labor adjustments. The installation schedule slipped by ten days.

My mistake was assuming that an approved drawing equals a measured opening. It doesn't. I've since added three rules to our project-buying checklist:

• Get field measurements or build in a contingency. Especially on retrofit work, verify actual rough openings before you commit to a production run. If the product has a rough opening schedule, read it.
• Ask for the performance label, not the brochure line. A window schedule that reads “Cornerstone Building Brands windows, series 4000” doesn't tell you the rated performance. I now ask to see the actual certification label under AAMA/WDMA/CSA 101/I.S.2/A440 for the exact model and glazing configuration, not a similar model.
• Check what's included. Does the quoted unit include panning, brickmold, sill pan, or installation clips? In aluminum window pricing, those are often separate line items. “The window was only $180” usually means $180 for the frame and sash — nothing else.

This level of checking feels tedious on a two-day bid, but it's cheaper than any expedite fee I've ever paid.

Scenario 2: You're the distributor sourcing products to resell

For most of my career, I was the inventory buyer. That's where I made my most frustrating mistake.

In 2022, we added a new imported rainscreen line. The supplier sent an impressive 80-page rainscreen cladding catalog. The panel colors were beautiful, the lead times were acceptable, and the per-square-foot price beat our existing line by $2.40. What we didn't do was verify that the catalog's performance data matched the product we were actually buying.

The catalog claimed a tested wall assembly with results from Europe. When the first warranty issue arrived, we discovered the tested assembly used different cavity depths, different fasteners, and a different substrate than what was sold here. We spent three weeks and about 40 staff hours figuring out whose responsibility it was. The profit on that first order disappeared completely.

Now I tell every distributor I work with to treat a product line decision like a hiring decision. Here's my checklist:

• Read the rainscreen cladding catalog the way you'd read an insurance policy. Look for assembly-level test results: structural, air leakage, water penetration. If the catalog only lists material properties of the panel itself, that's not a system — that's a skin.
• Request a written company profile. Not a one-page “About Us” page. I want manufacturing locations, third-party audits, warranty claim procedures, and whether they support OEM or private label. For example, I can read the Cornerstone Building Brands company profile and quickly learn what product categories they actually manufacture versus what they distribute. That transparency saves time.
• Do a mock warranty claim before signing an annual agreement. Ask: “If a bad batch arrives and half is already installed, what happens?” Time how long it takes to get a real technical answer. A supplier who takes three weeks to respond to a pretend claim will take three months to respond to a real one.

For anyone buying aluminum windows wholesale, I'd add one more: require quotes in line-item format. Frame, glass, hardware, coating, packaging, and freight should all be separate. Two numbers that look similar on the bottom line can be very different in what they actually include. AAMA 2603 vs. 2604 coating requirements alone can explain a “great deal” that isn't.

Scenario 3: You're the specifier writing the requirements

My costliest single mistake happened on the specification side. In early 2023, I was helping a team review a cement board rainscreen installation — about 8,600 square feet of cladding on a commercial building.

We approved a substitute “or equal” to the specified cement board without thoroughly comparing the two technical documents. The panels looked identical. Both were fiber-cement. But the original manufacturer's cement board specification guide required a ventilated cavity and open joints at specific widths. The substitute used a different joint detail that called for sealant-backer rod and a continuous drainage plane.

The installer followed the drawings, not the guide. After the first heavy rain, water wicked into the joints and stained the panel surface on one full elevation. We removed and reinstalled about 1,400 square feet. That cost $31,000 and added three weeks to the schedule. The “or equal” substitution saved maybe $3,000 in material.

If you write specifications, here's what I'd ask for:

• Ask for the full cement board specification guide, not the one-page cut sheet. The important information — fastener spacing, joint width, flashing requirements, allowable substrate — is usually buried in the back of the document.
• Specify by system, not by product photo. A rainscreen cladding catalog makes panels look like the entire solution. The real solution includes the air gap, flashings, joints, and water-resistive barrier. Specify all of it or none of it.
• Treat “or equal” as a red flag unless you compare test data. If the substitute doesn't have the same classification under ASTM standards and the same installation instructions, it isn't equal. ASTM C1186 covers flat fiber-cement sheets, but equivalent classification doesn't mean equivalent installation.
• Don't forget commercial fire-rated doors. A door leaf label doesn't automatically mean the frame and hardware are part of the listed assembly. I've seen a 45-minute rated door installed in a non-rated frame because someone read the label on the leaf. That's a code issue, not a product issue.

Which scenario are you actually in?

The hardest part is that some buyers fit more than one category. A contractor who also stocks a small inventory is both Scenario 1 and Scenario 2. An architect who helps a client buy products is both Scenario 3 and Scenario 1.

Here's how I tell people to decide where to focus:

• If your name is on the schedule and your crew is doing the installation, you're primarily Scenario 1. Measure, verify, and get the installation guide before you issue the PO.
• If you expect to sell the same products across many projects, you're Scenario 2. Don't choose a line on the strength of one project; evaluate the manufacturer's support, claims process, and documentation.
• If your deliverable is the drawing and the written spec, you're Scenario 3. Your best protection is a complete specification that names the exact system and the data required for approval.

And if you genuinely don't know which scenario you belong to? Start with Scenario 3. Write down the performance requirements first, then check the manufacturer's documentation against reality before you start negotiating price.

An honest boundary to all of this

This approach works well for mid-size B2B purchasing with repeatable orders. It's not the right amount of process for every situation. If you're a builder replacing twelve standard windows on one house, you don't need a 60-page submittal package. You need accurate measurements and a warranty that transfers to the homeowner.

I have mixed feelings about the level of scrutiny I now recommend. On one hand, I've seen it prevent real losses. On the other hand, too much documentation can slow down a good project. The balance is simpler than people think: read the technical documents that come with the product, verify that the supplier stands behind what they sell, and never assume that what worked for someone else's project will work on yours.

That last mistake is the one I made 23 times. $186,000 later, I finally stopped assuming.