About five years ago, I took over purchasing for a 45-person company that runs a mix of office, workshop, and outdoor storage space. I manage maybe $180,000 a year in orders across roughly 30 vendors, and I report to both operations and finance. I am not a structural engineer. I am the person who buys the stuff engineers and contractors say they need. That distinction matters more than I once thought.
The Fence Bracket That Started It
In early 2024, a windstorm pulled a 40-foot section of our perimeter fence loose. It didn't snap in half; it tilted, because a few of the brackets holding the posts to the concrete curb had pulled out of the concrete. Someone sent me a photo. The brackets looked like typical L-shaped metal plates, maybe 3/16 inch thick, with rust-colored coating. My first thought was to find a replacement. One supplier quoted a Simpson Strong-Tie fence bracket at $8.40 each. Another offered a similar-looking bracket at $4.20. I went back and forth for two days. Same shape, same paint, half the price. But I had also learned just enough to ask one awkward question: Where is your load table?
The $4.20 bracket didn't have one. The $8.40 bracket had a published load value, a code listing reference, and installation instructions. That didn't automatically mean the cheap one was poor. It meant that its performance was unknown, and 'unknown' is not a structural specification.
The Deeper Cause: We Buy by Look, Not by Engineering
I think the root problem is not deceptive pricing or shady suppliers. The root problem is that many of us in procurement are trained to compare SKUs by appearance, dimensions, and price. We have a spreadsheet with columns for unit cost, lead time, and quantity. We do not have a column for 'tested to what?' or 'under what conditions?' So when a part looks identical to another part, we assume it's equivalent.
For structural hardware, that assumption is dangerous. A connector made of the same gauge steel can be nothing like its branded counterpart if the steel grade is different, the bend radius is off, or the fastener holes are not where a load model was designed. It might hold a shelf. It might not hold a fence in a wind gust.
What I mean is this: the visible part is not the product. The product is the combination of geometry, material, finish, and the test data that says this exact geometry carries this exact load. The data is not a marketing add-on. It's the actual deliverable.
This became clearer when I started buying Simpson Strong-Tie Titen HD concrete screws. The name sounds like a specialty item. It is. Titen HD anchors are listed concrete screws with published pullout and shear values, but those values are tied to installation details: proper hole diameter, correct embedment, edge distances, and torque. If you skip any of those, you aren't installing an anchor. You're installing a screw that looks like an anchor. The difference shows in a pull test, or during wind, or when an inspector asks for the report.
What a Wrong Anchor Actually Costs
The expensive part, in my experience, is rarely the more costly fastener. The expensive part is when the cheap fastener fails. In 2022, we had a metal shelf system collapse in a storage bay. Nothing hit it; the shelf simply pulled out of a concrete wall. The contractor had not used a code-listed anchor. He used a generic concrete screw. The quoted bid was $1,300 less than a competitor because it included 'supply of concrete anchors' as a line without a part number.
That decision ended up costing us roughly $9,500 in cleanup, reinstall, re-proofing, and a weekend of down time in that bay. The hardware savings was maybe $350. I wish I could say the problem was malevolent. It wasn't. The problem was that nobody in the chain specified the product by an evaluated number. 'Concrete screw' is not a spec.
Then there are the quieter costs: the time your facilities person spends returning to site, the awkward email to the managing director, the inspector who loses confidence in your next submittal. Those costs do not show up in a purchase order, but they are real.
This is why I've learned to ask what is NOT included before I ask what the price is. It sounds strange for a purchasing person to admit, because we're supposed to be price-first. But a quoted price for a connector is misleading if it does not include the load documentation, the coating spec, or the fastener's compatibility with the base material. The vendor who answers that question honestly is the one I trust.
The cheapest quote is only cheap if it answers the same question as the expensive quote.
A Coupling Nut and a Torque Wrench Are Both Specs
I used to think of items like a zinc-plated steel regular coupling nut as buy-the-cheapest hardware. Then a maintenance project required joining two threaded rods that were holding a light fixture grid. The rod was under tension. A coupling nut has to engage both rods properly, with the right thread fit and enough material for the load. The grade of steel matters. The plating matters if there is moisture. The manufacturer's part number matters if you need to trace it later.
I now buy those from a manufacturer that publishes the spec instead of an unwrapped bin that may or may not be grade 2, grade 8, or something from an unlabeled import lot. Does that sound like overkill for a coupling nut? Maybe. But the process is the same: I want to know the part is what it says it is.
Same with tooling. I read a handful of 24330 drive click torque wrench reviews before buying one. Some reviews said the click is subtle, which is true (unfortunately). But a torque wrench is not decorative. Titen HD installation instructions tell you to tighten the anchor to a specific torque, because over-tightening can degrade holding strength. A click wrench removes my 'that feels tight enough' brain from the equation. I bought it because I needed repeatability, not because I wanted a drawer of nice tools.
Even Placement Is Part of the System
Once, someone asked me for advice on where to place kitchen cabinet hardware. It seemed unrelated to structural connectors. But the same logic applies: the location is a performance decision, not a cosmetic one. A drawer pull placed too high or too low changes the wrist angle. A handle on a cabinet door that is installed over a concealed hinge screw can compromise the fixing. The right hardware in the wrong location fails in its own way.
The principle sticks with me: placement is not decoration. A structural bracket placed where the load path doesn't follow its axis will not develop its rated capacity. A cabinet pull placed where your hand can't comfortably use it is a daily irritation. Both are failures of specification, not failure of the part itself.
The Boring, Practical Fix
After five years of managing purchasing, I've come to believe that good buying is not about choosing the expensive brand. It's about making the unknown visible. Here's what I now do, and it takes less time than the rework it avoids:
- Specify the application, not just the object. For a fence bracket, that means post size, material, and the kind of load it might see.
- Use a manufacturer part number wherever possible. If someone substitutes, ask for the same type of published data, not just a matching photo.
- Buy the installation device. If the anchor needs a specific torque, buy the wrench. If the post base needs a specific bolt size, put the bolt in the same order.
- Ask what is not included. Load table? Code report? Installation video? If the answer is mostly blank, treat it as a red flag.
- Treat placement as a spec. Whether it's a bracket on a concrete curb or a pull on a kitchen cabinet, location changes how the part performs.
According to Simpson Strong-Tie's published specifications (strongtie.com), structural connectors and anchors are tested to recognized standards and listed with load values under specific installation conditions. Those conditions are always on the detail sheet. I keep a folder of them.
I don't think everyone needs to become a materials engineer. Honestly, I'm not one, and I still get a little lost in load tables. But I stopped pretending that a fastener is just a fastener. The part number is a promise. The data tells you what that promise includes.
Looking back, I should have standardized our fastener purchases years earlier. At the time, I thought keeping more suppliers meant more options and better prices. It meant more unknown variables. If I could redo that first fence bracket order, I would ask the $4.20 vendor to send a data sheet before I even looked at the price. That one question would have saved us time, stress, and a repair bill that still makes me wince.
