I don't care how good the price looks. If the specification is wrong, you haven't saved money—you have created a second purchase order, a return label, and a conversation with your facilities manager that nobody wants. Five minutes of checking a part number prevents five days of undoing the mistake. That's the rule I use for almost every order I place for our company.
I'm the office administrator for a regional facilities company. I handle the purchasing for roughly twenty maintenance techs across three locations. That's about 60-80 orders a year—valves, pumps, tools, repair kits—and in 2024, during our vendor consolidation project, I also became the person who approves most product requests before they reach finance. I'm not an engineer. I've just had enough wrong orders to know what actually saves money. But saving five minutes is not saving money.
The order that taught me about Watts regulator valves
In 2020, I ordered what I thought was a standard Watts regulator valve for a boiler feed line. I found a great price from a new supplier and placed the order in about ten minutes. Seemed efficient. The part arrived, the maintenance lead took one look at it, and said it wasn't the pressure range we needed. He asked me whether I'd checked the rating stamped on the old body. I had not.
The supplier was happy to take the return, but the restocking fee ate into the discount. The real cost was the delay: the system ran with a temporary bypass for two days while the correct valve shipped. The finance department didn't care about the discounted price. They cared about the labor hours and the fact that I'd created two invoices for one task.
Since then, I treat every Watts regulator valve order as a spec check, not a shopping task. I look at the line size, the pressure range, the connection type, and the range of adjustment. I also ask for a product data sheet before I enter the PO. If the vendor can't produce one, I move on.
People ask 'how many watts is a well pump?'—that's the wrong place to start
One question I hear from our own techs, and also see in searches, is 'how many watts is a well pump?' The question makes sense on the surface. We buy a pump, and we want to know what it will draw. But if you order a well pump based only on wattage, you're probably going to get the wrong pump.
Wattage is a starting point, not a spec. A typical 1 HP well pump might draw in the neighborhood of 2,000 watts under load, and the start-up draw can be higher for a few seconds. But the pump also has voltage, phase, service factor, discharge size, and a pressure switch setting. According to the pump nameplate, which is the final authority, you need to match all of those—not just the watts. I'm not 100% sure of every regional difference, but I've never seen a reliable order that was based only on watts.
Actually, most pump manufacturers publish a performance curve. A pump that draws the same wattage can still deliver a different flow and head. Two pumps can both be '2,000 watts,' and one might be designed for a shallow well while another is built for a deeper setup. If I order based on watts, I might end up with a pump that runs but can't deliver the pressure where we need it.
The better question is: what pump does the manual specify? When someone hands me a pump request labeled 'x water pump manual,' I understand that they don't actually want the manual—they want the right pump, and the manual is the reference that defines it. If there's no manual or model number, that's not a spec; that's a guess.
Documentation is not paperwork. It's part of the product.
Early in my procurement role, I treated manuals as afterthoughts. I'd order the part, receive it, and occasionally request the manual later if the tech asked for it. If the manual was missing, nobody wanted to chase it. That changed after I ordered a recirculating pump without checking what repair kits were available.
The pump worked, but the maintenance team needed a service kit faster than the distributor could source one. The delay cost us overtime. Now I ask about repair kits on the same PO as the pump. If a Watts product has a repair kit, I usually add it to the order unless the person who requested it says otherwise. In my experience, repair kits are cheap insurance a buyer can put on a purchase order—provided you've verified the model number first.
The same rule applies outside plumbing. I've had a maintenance lead request a 'cordless circular saw mpn:dcs577b.' That's a specific product, which helps. But before ordering it, I still check that it matches the battery platform our techs already use, and I confirm the kit contents. A good part number doesn't guarantee a match to the user's existing tools. The manual and the model number go together. I include the manual access in the order notes so nobody has to ask for it twice.
The 'how-to' gap: prevention applies to knowledge too
I've seen people search for 'how to make welding machine at home pdf.' I get the appeal of building something yourself. I'm not against DIY repair or making things on your own—that's how a lot of good technicians learn. But a homemade welding machine is not the place to work from a random PDF with no traceable source. Prevention here means checking the design first, not just the instructions.
For procurement, the same principle applies. If the knowledge isn't written down and sourced—a manual, a datasheet, a spec sheet—it isn't reliable. I would rather wait an extra day for the right documentation than trust a scrappy file that might be missing the safety section.
You might think I'm being too careful
I know there's a part of my job that looks like checking things twice for no reason. One of our techs started calling me 'the paperwork person.' That's fair. But paperwork is how we avoid the rework that nobody budgets for.
Here's where I'll admit my limitation: my experience is based on about 60-80 orders per year for a facilities operation. If you're procurement for a high-rise contractor or an industrial process plant, your pressure ranges, lead times, and documentation requirements will be different. I can't speak to those environments. But I've seen enough sideways purchasing decisions to know the pattern: the money you save on a discount disappears when you skip the spec check.
The mistake I made in 2020 wasn't buying from a new supplier. It was treating speed as the deliverable. The deliverable is a part that works the first time. A Watts regulator valve is a solid, well-known product, but only if it's the right one for the application. A well pump is the same. A cordless saw is the same. In this way, almost every purchase is the same.
I still don't like returning parts. But I dislike explaining why I didn't ask a basic question. Five minutes of checking beats five days of correcting. I still don't do it perfectly—I've lost count of the small mistakes—but I do it first now. That's the whole argument.