I’ve been handling orders for measurement and control instrumentation for about seven years now. If I remember correctly, I’ve personally made (and documented) 23 significant mistakes over that time. The total wasted budget? Roughly $15,700. Give or take a few thousand.
The most frustrating part of those mistakes: they were almost never about the product itself. The Dwyer flow meter or pressure transmitter I ordered? It worked fine. The problem was what happened before the order got placed—the specification, the communication, the assumptions.
Surface Problem: The Wrong Instrumentation Arrives
Let me start with a concrete example. In late 2022, I ordered thirty paddlewheel flow meters for a coolant monitoring system. The application was straightforward: measure flow in a closed-loop glycol loop. I’d specified the meters, the customer approved, the order went in.
When the shipment arrived, I opened the first box and immediately knew something was off. The insertion length was wrong. All thirty units.
I said, “I need a standard paddlewheel flow meter for a 2-inch pipe.” The sales engineer heard, “Standard model for a standard insertion depth.” We were using the same words but meaning different things. The result: thirty flow meters that didn’t fit. $4,200 worth.
You’d think written specs would prevent this type of confusion. But in practice, “standard” depends entirely on your frame of reference. For that vendor, the standard insertion depth was for their own fittings. For me, it was for the pipe mount we’d always used. The mismatch cost us two weeks of delay and a re-stocking fee.
Deeper Reason: The Assumption Cycle
This brings me to what I think is the real culprit: the assumption cycle.
Here’s how it works:
- An engineer or technician has a measurement need (e.g., monitor differential pressure across a filter).
- They reference a similar past application and pick a familiar product—maybe a Magnehelic differential pressure gauge or a Series 616 pressure transmitter.
- They assume that because it worked last time, it will work this time.
- The product gets installed, and something doesn’t line up—wrong output signal, incompatible process connection, insufficient accuracy for the range, you name it.
I’ve seen this pattern so many times. The most frustrating part is that nobody is trying to do a bad job. We’re all trying to be efficient. But that efficiency drive—the desire to get things done quickly—can short-circuit the specification process.
Let me rephrase that: the pressure to move fast often forces engineers to make assumptions that, in a slower environment, they would never make. Put another way: speed can be the enemy of accuracy.
The Real Cost of These Mistakes
The financial cost is obvious. My $4,200 paddlewheel blunder is a clear line item. But the hidden costs are where things get serious.
There are three categories I’ve come to track:
1. The Direct Redo Cost
This is the replacement equipment, the shipping, the restocking fees. In my 2022 case, that was $4,200 plus a restocking fee of around $420. We also had to expedite the correct units—overnight freight on thirty flow meters isn’t cheap. Total direct cost: roughly $5,300.
2. The Delay Cost
The coolant system couldn’t come online until the correct meters arrived and were installed. That delay cost the client an estimated $2,500 in lost production time. I didn’t pay that out-of-pocket, but I felt it. Delays kill team morale, and they damage your credibility with the client.
3. The Opportunity Cost
This is the one most people forget. While I was scrambling to fix the flow meter issue, I wasn’t working on other projects. I wasn’t improving other processes. I wasn’t taking on new work. The time I spent troubleshooting and reordering was time I couldn’t spend on value-adding activities. I’d estimate that cost me about $4,000 in lost billable capacity across the whole project.
So the total cost of that single mistake was closer to $11,800—not the $4,200 on the purchase order.
The Hidden Pattern: Specification Friction
What links all these mistakes together is what I now call “specification friction.” It’s the disconnect between what you think you need and what the product documentation provides.
In the old days, you’d pull out a paper catalog, look up a product, and make a call. The friction was high. You had to parse specification tables, model numbers, and option codes. That forced you to slow down.
Today, we have access to digital catalogs and online tools. The friction is lower—but only if you use them correctly. The temptation is to search for a part number, click “buy,” and move on. That’s where the assumptions creep in.
I have mixed feelings about this shift. On one hand, the speed is incredible. I can find a Dwyer Series 628 pressure transmitter spec sheet and cross-reference its options in under a minute. On the other hand, that speed sometimes makes me careless. Part of me wants to blame the tools, but another part knows the fault is mine for rushing.
The (Short) Solution: Pre-Check, Not Double-Check
After the third rejection in Q1 2024, I created a pre-check list for every instrumentation order over $500. It’s not a complex system. It’s a simple document that forces me to answer four questions before I send the order:
- What is the exact process connection (thread size, type, material)?
- What is the required output signal (4-20 mA, pulse, switch, Modbus)?
- What is the measurement range (minimum and maximum expected value)?
- What are the environmental conditions (temperature, pressure, humidity, vibration)?
That’s it. Four questions. I fill them out, send them to the supplier or verify them against the online spec sheet, and only then do I place the order.
We’ve caught 47 potential errors using this checklist in the past 18 months. The cost of those errors would have easily exceeded $30,000. The checklist cost me an hour to write. The returns are absurd.
The key is that this isn’t a double-check—it’s a pre-check. I do it before the order is generated, not after. It forces the thinking to happen upfront, where it belongs.
Why This Matters More Now
Switching to this more methodical approach cut our turnaround from an average of about 5 days to 2 days. The automated pre-check process eliminated the data entry errors we used to have. Efficiency, in this case, wasn’t about moving faster—it was about reducing the time spent on rework.
I’m not saying every traditional method is wrong. There are times when a quick call to a trusted sales engineer is faster and better than any form or checklist you can dream up. But for standard products—flow meters, pressure transmitters, temperature sensors, humidity transmitters—the spec is the spec. Getting it right the first time is the most efficient thing you can do.
The industry is moving in this direction. The best vendors now offer online specification tools that pre-fill or validate your requirements. Using them isn’t just a time-saver; it’s a risk-mitigation tool.
I still make mistakes. Last month, I almost ordered a dp transmitter with the wrong output. The checklist caught it. That alone saved me the embarrassment—to say nothing of the cost.
Roughly speaking, I’d estimate that our error rate is down about 70% since we started using this approach. That’s a number I’m pretty happy with.
