Six years ago, I thought the best way to prove my value as a purchasing manager was to get the lowest unit price on every order. I was wrong. Looking back at my own cost records — roughly $1.2 million a year in material buying for a commercial building contractor — the projects where we chose the cheapest first cost carried about 18% more rework and callback expense than the projects where we bought to the actual requirement. The most expensive double-sided tape I have ever bought wasn't the premium one. It was the cheap one, installed twice.
That 18% figure isn't from an industry study. It's from our own job-cost records, and that's the point. A unit price is one line on one invoice. The costs of the wrong product show up weeks later, on a service ticket, in someone else's budget.
Where that 18% number comes from
I manage procurement for a 44-person commercial building contractor. We install storefront glazing, exterior coating systems, and commercial entry doors. For the past six years, I've handled a materials budget of roughly $1.2 million a year, negotiated with more than 30 vendors, and logged every order in the same cost-tracking file.
My first two years were simpler. I collected three quotes, picked the lowest one, and reported a saving. Then, in a Q4 2021 audit, I matched purchase orders to the project manager's callback list. The pattern wasn't subtle. About six out of ten orders that had gone to the lowest quote produced some kind of follow-up cost. Not every failure was dramatic, but the service hours added up.
I don't have hard data on how other contractors buy, so I'll only speak for our shop. In our records, the cheap option wasn't cheap once the second visit was included.
The $190 saving that turned into a $1,480 repair bill
The clearest early example was an interior column cover job in spring 2023. We fabricated 24 column covers for a tenant fit-out. The drawing specified 3M double-sided tape with an acrylic adhesive. The supplier offered an alternate from another established company. It cost 38% less, and the data sheet looked reassuring.
Was the alternate product bad? I honestly don't think so. It was probably fine for many applications. What I ignored was that our application wasn't typical. The column covers saw temperature swings, and a few of them got bumped by carts. We compared the two products on single-point adhesion numbers, not on long-term behavior under real conditions.
Four months later, the edges of several covers had lifted. The tenant was about to move in, so repairs had to happen on a weekend. Lift rental, two installers, new material — the rework came to $1,480. The original material saving was $190.
Here's the part that stuck with me: the tape was a small fraction of the installed cost. Saving 38% on a small fraction was never going to be a project win, because the labor and access cost were identical no matter which tape we used. The only way to save on the small part was to risk the whole installation. That math made no sense.
The same mistake, with '3M velcro strips'
The same pattern showed up with what our field team calls '3M velcro strips.' The proper product category is reclosable fasteners, and we use them on access panels in mechanical rooms. The panels get opened by service techs, so the fasteners need to hold through dust, temperature changes, and repeated cycles.
One year, a cheaper hook-and-loop alternative looked equivalent on paper. The holding force was similar at room temperature, and the price was about 35% lower. But the real environment wasn't room temperature, and it certainly wasn't clean. The strips lost their grip faster than the specified product, and we had to replace them across about 80 panels.
The material price difference for the whole project was roughly $200. The scheduled service visit to replace the strips cost more than that before we even bought the new material. Nobody set out to make a bad decision. We simply compared the wrong numbers and skipped the environment part of the spec.
Elastomeric coating sourcing: the gallon price that lied
Elastomeric coating sourcing is where the same mistake gets even more expensive, because coatings hide cost in film thickness.
A coating is sold by the gallon, but it performs by the mil. In a Q3 2024 evaluation for a masonry wall project, we compared two elastomeric coatings side by side. Coating A was about 30% cheaper per gallon. At the specified dry film thickness, Coating A needed a second coat. Suddenly the cheaper gallon wasn't cheaper per square foot, and the extra coat added labor on top of material.
Coating B cost more per gallon, but it reached the required thickness in one coat. When we calculated cost per square foot at the specified film thickness, the more expensive product was competitive — and sometimes it came out ahead. The price per gallon had been the wrong column from the start.
That experience changed how I read architectural coating submittals. Now I ask every bidder the same questions: What is the volume solids? What dry film thickness do you get per coat? What is the recoat window? If a supplier can't answer those, their spread rate on the data sheet doesn't mean much.
There's a documentation angle too. Per FTC guidance (ftc.gov), environmental claims like 'green' or 'eco-friendly' need to be substantiated. When a coating data sheet says 'low-VOC' or 'sustainable,' we ask for the actual VOC content and the test method. If the paperwork is vague, I worry about what else wasn't verified.
Entry doors specification guide: what I ask before approving an alternate
Entry doors look simple from a distance. A door is a door — until it arrives without the factory prep the hardware schedule assumed.
In my experience, the low price on a commercial entry door usually means the quote only covers the base door, frame, and hardware. It doesn't cover the things that make the door work on a specific opening: mullions, weatherstripping, electric strikes, concealed closers, exit device prep, and the coating system on the door face.
In 2023, we compared an alternate door package on 36 openings. The alternate looked about $12,000 cheaper. By the time we added factory prep for access control, the specified hardware schedule, and the required weatherstripping, most of that saving had disappeared. We almost approved the quote based on price per door, which is exactly the wrong unit.
Three things go into every entry door RFQ now:
- Rating documentation. If the door is part of a fire-rated or egress assembly, the whole assembly has to carry the label. A substitute without the documentation isn't equivalent, no matter how good it looks in a brochure.
- Factory hardware preparation. Drilling holes and cutting prep in the field voids some warranties and always adds labor. Compare the package, not the leaf.
- Finish system. Entry doors take abuse. A weak coating leads to chalking, rust bleed, and an early replacement. We connect this review back to the same coating questions from the wall section.
If someone hands me an entry doors specification guide, the first thing I look for is whether it makes those costs visible. If it doesn't, the guide is incomplete.
Where I still buy the cheap stuff
To be fair, the answer isn't 'always buy the premium option.' Some cheap products have earned a place in our shop.
If the job is temporary, or the failure is easy to fix, paying more only buys a logo. We still buy commodity double-sided tape for masking, protective paper, and temporary mounting indoors. We buy cheaper hook-and-loop strips for presentation boards in conference rooms. Those applications don't need a wide safety margin, and their failure cost is a five-minute fix.
The rule breaks down in the other direction too. No product replaces good surface preparation. In 2024, one of our installers applied a 3M tape to a frame that looked clean but wasn't. The tape failed, and it wasn't the tape's fault. The expensive product didn't forgive our mistake. Preparation still matters more than brand.
The difference between a value buyer and a price buyer is simple: value buyers know which variable actually causes the failure. When we can name that variable, we specify to it. When we can't, substitution is guesswork.
If you don't track rework costs yet, start. Add two columns to your spreadsheet: purchase price and cost of what happened next. In six months, you'll probably see the same pattern I did. The most expensive product isn't always the one with the highest price. It's usually the one that gets installed twice.