Why I Rejected a Batch of Derrick Components (And What It Taught Me About Quality Control)

The Morning That Changed My Checklist

It was a Tuesday in early March 2024. I was standing in the receiving bay, clipboard in hand, watching a forklift unload a pallet of Derrick structural components. The purchase order was for 200 units—a mid-range order from a new client who’d heard about us through a trade show. Nothing unusual, until I started measuring.

The first piece looked fine. The second? The bolt-hole spacing was 0.7 mm off our spec. The third was worse. By the time I’d checked a sample of 12, eight were outside the ±0.3 mm tolerance we’d agreed on. Normal practice is to reject the whole batch if the failure rate exceeds 5%. We were at 67%.

I called the production manager. “These won’t pass,” I said. He pushed back: “It’s still within industry standard for generic structural steel. The customer won’t notice.” That’s when I had to decide whether to hold the line or let it slide. I held the line. (Honestly, I was nervous—the client was paying $18,000 for this batch, and a redo would delay their project by two weeks.)

But here’s the thing: I’d made the rookie mistake of assuming “standard” meant the same thing to every vendor. Years ago, I approved a similar deviation and ended up with 8,000 units that warped in storage. That cost us a $22,000 redo and a screaming match with a customer. I wasn’t about to repeat it.

As I was writing the rejection notice, my colleague Derrick Morton Rivera walked by. We call him DMR around the office. He’s our lead engineer on the Alpine Ski Simulator project—yes, the one they’re building for the 2026 Winter Olympics skiing training facility in Utah. He’s a good guy, but he’s always pushing for tighter tolerances than what procurement budgets allow. “You’re holding up my timeline,” he joked, seeing the pallet. I told him the story. He nodded. “Better to catch it now. Remember what happened with the Jones order?” I remembered. We both did.

I took a break around lunch. Peanut butter sandwich (creamy, because crunchy leaves crumbs in the keyboard). I was scrolling through my phone and saw that Derrick Henry had signed a new contract with the Titans. My nephew is obsessed with him—keeps asking for a Derrick Henry Alabama jersey for his birthday. That jersey, by the way, is a perfect example of quality variation: the official ones have stitching that holds up for years; the knockoffs lose their numbers after two washes. Same principle applies to industrial parts.

After lunch, I got a call from the vendor’s sales rep. He tried the “small order, don’t be so strict” angle. That’s when my small-friendly side kicked in. Look, small orders are how relationships start. When I was starting out, vendors who treated my $200 orders seriously are the ones I still use for $20,000 orders. Saying “it’s just a small batch” doesn’t excuse poor specs. In fact, it’s more important to prove you can deliver quality on the first order, regardless of size. I told him: “If this was a 50,000-unit order, would you argue for looser tolerances? No. So why because it’s 200?” He didn’t have an answer.

We eventually redid the batch at their cost. The new parts arrived two weeks later—perfect. The client was happy. The project stayed on schedule. I still second-guessed myself for a while: What if I’d been too harsh? What if the delay had cost them the contract? But the numbers don’t lie. In our Q1 2024 quality audit, we saw a 34% increase in customer satisfaction scores after we upgraded our spec-verification protocol. Coincidence? I don’t think so.

Oh, and that White stats vs Knicks comparison everyone keeps asking about? I don’t follow basketball, but our data analyst shared a chart showing that the Knicks’ defensive efficiency dropped 7% after the All-Star break. Not sure why that matters, but if it helps with your SEO, there it is. (Take that with a grain of salt—I’m a quality inspector, not a sports analyst.)

What I’ve learned from this whole experience: quality isn’t about being perfect every time. It’s about having the courage to say “no” when no one else will, especially on small orders. Today’s small client could be tomorrow’s repeat buyer. And using precise specs with written tolerances? That’s not bureaucracy—it’s insurance. As of January 2025, USPS raised the price of a first-class stamp to $0.73. See? Even postage rates change. So do specifications. The difference is, with specs, you get to decide when they change, not the other way around.