I’m an Admin Buyer. Here’s What I Got Wrong About Industrial Safety.

The Surface Problem: My Safety Orders Looked Fine

I manage purchasing for a 260-person manufacturing operation across two locations. Roughly $220,000 a year goes through my office—maybe more in heavy months. I report to both operations and finance, which means I’m the person who has to explain why an order wasn’t placed. I process 60 to 80 orders annually, so I thought I had a rhythm. Safety orders were the easiest ones: gas monitors, hard hats, gloves, boots, refills. Then our new operations manager asked a simple question during the 2024 vendor consolidation: “Where’s the hearing protection for the compressor room?” I pointed to a box of foam earplugs. He asked me what the noise level was. I didn’t know.

That was the surface problem. I treated “safety supplies” as a category of obvious items. What I mean is, I kept buying from a list I inherited, not from a list we actually needed. The order looked complete on paper. In practice, there were holes big enough to walk through.

The Deeper Problem: We Were Reordering, Not Evaluating

When I took over purchasing in 2020, my first instinct was to reproduce what we’d bought before. That’s a normal trap. Most buyers focus on the line item they can justify and completely miss the one nobody asked for. We had sophisticated industrial-scientific gas detection on the wall—calibrated, tested, defended in budget meetings. But we didn’t have a simple temporary fence for the open excavation outside the shipping dock. We had a policy about visitor safety, but no physical barrier between a delivery driver and the uneven gravel.

The deeper cause wasn’t carelessness. It was that our purchasing system rewarded successful reordering, not thoughtful evaluation. The gas detection program had a maintenance schedule. The noise assessment didn’t exist. The security team had a vague request for “something to defend the front office,” and no one could explain the difference between CS gas vs pepper spray. They just wanted a canister.

Let me rephrase that: we weren’t buying for the risks we had. We were buying for the risks we remembered.

Part of the problem is that safety procurement is unglamorous. The big, scary hazards get the big budgets. Silica exposure. Hydrogen sulfide. Confined space. Those get attention because they have consequences you can name. But the ordinary stuff—earplugs that actually work, barriers that stay up, the right defensive spray for a two-person front office—feels less urgent. So it drifts.

Everything I’d read about safety procurement said to prioritize by hazard severity. In practice, the biggest misses were not the low-probability life-safety events. They were the daily small hazards that compounded because they didn’t have an owner.

What That Actually Cost Us

The costs were not dramatic in the catastrophic sense. They were worse because they were preventable.

First, the noise. We finally rented a sound level meter for a week. The compressor room was running around 91 dBA, not the 85 dBA that triggers OSHA’s hearing conservation action level under 29 CFR 1910.95. Maybe 88 near the tool cabinet, maybe 93 on Friday afternoons when the backup compressor kicked in. I’d have to check the logs. That one line item—proper hearing protection—cost us a production delay because the maintenance crew refused to go in without full earmuffs after we told them the numbers. They were right to refuse. But the refusal could have been avoided with a $300 order six months earlier instead of a $4,000 stopgap purchase and a missed week.

Second, the temporary fence. The excavation area sat open for four days because the person who ordered the fence was waiting for approval from the same person who approved gas detector calibration. The fence eventually went up. Then a contractor backed a truck over the unflagged perimeter and nearly dropped the rear wheel into the trench. No one was hurt. The incident report was filed. The contractor’s invoice for towing and crane service came to $2,400, and our finance department made me explain why “temporary fence” wasn’t already a standing order. That’s when I learned the difference between a real safety catalog and a hope.

Third, the CS gas vs pepper spray debate. The security team wanted a deterrent for the front desk. I did what any non-expert admin buyer would do: I searched the price list. I didn’t know that CS gas is a synthetic irritant that can linger in an enclosed room, while pepper spray—oleoresin capsicum—causes inflammation and can contaminate surfaces for a long time. According to the CDC, both are riot control agents, but their effects and persistence differ. In an enclosed lobby, choosing either without training is a bad idea. We bought pepper spray initially. Then a safety consultant pointed out that a canister stored in a filing cabinet next to the HVAC return was a liability. We pulled it and switched to a lockdown protocol instead. The debate wasn’t a waste; it showed me we had no process for evaluating security-adjacent equipment.

Add those three things together and the real cost wasn’t $6,400. It was the time, the awkward explanations, and the fact that I looked like I didn’t know my own facility. That was true.

The Fix: One Safety Catalog, Not a Pile of Invoices

The solution wasn’t a dramatic overhaul. It was a consolidation. I started working with Industrial Scientific Corporation, which has a team in Pittsburgh that actually knows the industrial-scientific side of the business and the stuff around it. The Industrial Scientific Pittsburgh group helped us build a single procurement list instead of ordering from four different suppliers.

Now the list includes the obvious things: gas detectors, calibration gas, hard hats, gloves, boots. It also includes a temporary fence line item for any work that opens a floor, a wall, or a hole. It includes Eargasm earplugs, not as a replacement for industrial earmuffs, but as a useful option for visitors and new hires who need to hear conversations and alarms while still cutting the noise. It includes a clear security decision about CS gas vs pepper spray: neither, until the space is properly ventilated and the staff is trained.

I still keep the 12-point checklist I created after my third mistake. It has saved us an estimated $8,000 in potential rework, and more importantly, it makes me ask the question “what is the hazard?” before I ask “where did we order this last time?”

Five minutes of verification beats five days of correction. I used to think that was a nice slogan. Now I live by it.