I'd Rather Spend 5 Minutes Checking Than 5 Days Explaining: An Office Administrator's Case for Prevention Over Cure

Since January 2020, I've been the office administrator for a 180-person company. My title sounds generic, but my purchase orders are not. I manage roughly 80 orders a year across about a dozen vendors, covering everything from commercial restroom fixtures to party supplies you don't think about until the week of the party. I report to both operations and finance, which means I see every purchase twice: once when I approve it, and once when the invoice lands.

After about 150 orders, I've come to believe that prevention beats cure in every purchasing decision. A five-minute check before an order costs almost nothing. A five-day correction after an order costs budget, schedule, and credibility. If that sounds dramatic, it's because you haven't yet spent a week explaining a restocking fee to your CFO.

A part number isn't a specification

The quote line read: plumbing valves Kohler 3384, chrome finish, quantity twelve, lead time two weeks. We were refreshing two restroom floors, and the model number was never in question.

That's exactly what worried me.

In 2021, I approved a similar order based on a quote that was missing the rough-in detail. The model line was correct; the application wasn't. Nobody caught it until the wall was open, and the supplier charged us restocking plus a second freight bill. That error cost about $600 and a week of schedule, and it taught me an expensive, effective lesson: a part number is not a specification.

So last spring, before approving the Kohler valve order, I called the plumber on-site. The call took eleven minutes. He read the critical dimension from the drawings, and we compared it to the quote. The quote was missing it. We corrected the PO, sent it that afternoon, and the valves showed up on time.

I don't have hard data on how often a quote contains that kind of gap. I can tell you anecdotally that maybe one in twenty of our equipment specs needs a correction if someone actually checks. That's not a reason to distrust vendors; it's a reason to make checking routine.

A manual you can't find is a manual you don't have

In 2021, we installed a Kohler standby generator. It has the OnCUe remote monitoring system, which is genuinely useful—until the generator disagrees with you. In March 2023, after an ice storm, the controller logged a fault. Not an emergency, but not something I should ignore. The service technician asked for the Kohler OnCUe manual to confirm the code.

I said, “It's around here somewhere.” Then I discovered that “around here somewhere” was a set of email attachments from two years earlier.

The technician waited while I searched my inbox and shared drives. Twenty-five minutes later, I found the PDF; another ten minutes passed before we identified the right section. A service call that should have been forty minutes ran close to ninety. I watched the billing clock do things I will not describe here.

The prevention was embarrassingly simple. I printed the manual, put it in a labeled binder next to the generator control panel, and tucked an index card inside the cover with the OnCUe portal address and our service provider's phone number. The PDF now lives in our facility folder under a filename that doesn't require archaeology.

Preventive maintenance, I've learned, is mostly boring documentation. The machine doesn't care where its manual is. You do, at 3 p.m. on a day when something blinks.

Small purchases are not low-risk purchases

Now for the purchase that doesn't fit in the phrase “strategic sourcing”: a coworker finished her master's degree after two years of evening classes, and the office threw a small toast. My assignment: sparkling cider in glass bottles, because a toast in plastic feels wrong, and the obligatory grad cap for photos.

This was a $120 order, which is exactly why it deserves a mention. It's tempting to think the low-dollar stuff requires no checking. Honestly, I'm not sure why we treat low prices as low risk. My best guess: small prices make us believe mistakes will be small. They're not small; they're just earlier.

The cider had to arrive by a date, in glass, through the USPS network. According to USPS (usps.com), fragile items like glass bottles need enough cushioning that they can't shift in the box. So I asked the vendor a mildly annoying question: “How are the bottles packed?” The answer was “in the manufacturer's cartons,” which is not the same as “they won't shift.” We added filler. The order arrived intact.

The grad cap came separately, two days early. That was enough time to notice it was slightly squashed and fix it before the photos. If I'd ordered it with standard shipping, the event would have been the test. Lead time is a spec, too.

What I remember most about that afternoon is the wall. After the toast, someone wheeled a cart of empty bottles toward recycling and clipped the corridor wall, leaving a paint chip about the size of a quarter. We own the building, so I didn't have a landlord to call. I added “fix paint” to a mental list, and then I did the thing I criticize everyone else for: I postponed it.

Two months later, the chip had lifted at the edges, and the cleaning crew's damp cloth had spread the peeling. What should have been a 20-minute touch-up became a sand-and-prime job that took the better part of an hour one Saturday. When I finally searched how to repair chipped paint, the advice was exactly what you'd expect: clean, sand the loose edges, wipe away dust, apply primer, then paint in thin coats and feather each edge. None of it was hard. Waiting was the only part that made it worse.

That chipped wall is the best metaphor for procurement I know. It's never the original problem that kills you. It's the gap between noticing it and doing something about it.

But “checking everything” takes too long

I hear this, and I partly agree. The workaround is that you don't check everything with the same intensity. I don't call a vendor when I order copy paper. But the moment an order involves a spec, a fragile box, or a deadline, I ask one question: if this goes wrong, what does the cure cost? If the answer is “a trip to the supply closet,” I proceed. If it's “a restocking fee and a stalled project,” I spend the five minutes.

There's a second objection: some problems happen even when you check. True. Vendors ship late. Generators throw faults. Paint chips. No amount of checking prevents those moments. But when they come, I'd rather be holding the right model number, a printed manual, and a paper trail than a sincere apology.

Prevention isn't glamorous. Nobody compliments the administrator who ordered the correct valves, printed the manual, or fixed the paint while it was still a quarter-sized chip. Those are non-events, and non-events are most of this job. After five years and about 150 orders, I would still rather spend five minutes checking than five days explaining. The math was never the hard part. The discipline is.