I coordinate field service for an instrumentation company that sells and supports industrial measurement equipment. When a plant loses a critical sensor at midnight, I'm the one who arranges the replacement, the courier, the calibration, and sometimes the therapy session afterward (not that we charge for that last one). Over the last five years, I've handled more than 200 urgent jobs. Last year alone we processed 47 rush orders, same-day turnarounds included, and still managed 95% on-time delivery.
Here's the opinion I've formed from those calls: the sticker price isn't the price — it's just the deposit. More often than not, the failure that triggered the emergency began months earlier, in a purchasing decision. Someone compared quotes and selected the cheapest number. That number was real. So was the cost that came later.
I'm talking about total cost of ownership — TCO, if you like acronyms — and I don't mean it as a buzzword. TCO includes the purchase price plus installation, training, calibration, maintenance, downtime, repairs, support, and disposal. Sticker price shows up in the purchase order. TCO shows up in the incident report.
Here are three examples from my world. See if they sound familiar.
The $990 Saving That Became a $10,000 Night
In March 2024, a beverage plant that we support called our emergency number at 1:40 a.m. Their labeling line kept stopping — false triggers, jammed bottles, the works. The maintenance manager was exhausted and, honestly, so was I.
When I asked what sensor was in the line, he laughed. 'The ones we bought online,' he said. 'What did you think?'
We found 18 photoelectric sensors on that machine. Fifteen of them were no-name units from an online marketplace, installed the previous summer. They cost around $30 each. A proper sensor for that application — in this case, an ifm photoelectric sensor — would have run around $95. That's a savings of about $990 for the set. I'm sure it looked responsible in somebody's cost review.
I'll be honest: ifm isn't the only maker of durable sensors. But when you search for discussions about durability of ifm photoelectric sensors vs others, most of what comes back is engineering talk — housing ratings, optical reliability, datasheets that match reality. After years of looking at failed parts, I understand why that matters.
The line sat still for roughly five hours. The maintenance manager estimated the cost at something like $2,000 an hour (don't hold me to the arithmetic; all numbers are from him, but the point stands with any reasonable estimate). That's $10,000 of lost production. Then came our emergency callout, new sensors, and the labor to install them — about $1,900.
I hesitated before recommending they replace all 15 suspect sensors rather than just the one that died that night. It was their budget, not mine. But the line had already faulted twice that week, and I couldn't look at the data and recommend doing the same thing again. The 'cheapest repair' would have made the third call inevitable.
The plant paid about $1,900 for parts and labor; the line ultimately lost about $10,000. A $990 savings turned into a five-figure night.
The Thermal Camera That Wasn't Actually Cheaper
Another angle, same trap.
In autumn 2024, an electrical contractor told me he had bought a budget thermal imager for electrical panel inspections. The on-screen image looked good. But exporting data for reports was painful — half the time the software crashed, and the 'radiometric' files were anything but easy to use. Twenty minutes to scan a panel, two hours to write up the report. And there was no calibration documentation that would convince his larger clients.
He asked about a FLIR E76 advanced thermal imaging camera, the model we stock for exactly this type of work. It costs more — in the thousands, depending on configuration and current pricing — so on a purchase order it looks like a luxury. But look at the TCO line: the E76 gives him reliable radiometric data, report-ready software, support from people who answer the phone, and calibration traceability. That's not brand tax. That's paying for things that otherwise come back as labor hours, missed deadlines, or — worse — a missed hotspot that later becomes a catastrophic failure.
He bought it in the end, but that's not the point of the story. The point is that the 'cheaper' camera was never cheaper; people just don't add two hours of unpaid engineering time to the price tag. The cost is invisible until it lands in someone's timesheet.
What About the Login You Never Think About?
Here's a part of TCO that doesn't appear on any quote: access to the manufacturer's support infrastructure.
Earlier that year, a quality manager called us in a panic. His company had bought a used coordinate measuring machine. The machine itself was in good shape. But the software license had lapsed that morning, in the middle of an urgent inspection. Nobody in the building knew how to reach the previous owner. Nobody had a Hexagon portal login. The asset was registered to an engineer who had left the company a year ago.
We spent an afternoon sorting it out — setting up a new Hexagon support login, transferring the registration, and downloading the current software. It sounds like paperwork. But while we were doing it, their quality lab was idle, and a customer audit was three days away. That idle lab was a real cost, and it had been baked into the 'great deal' they got on the used machine.
If you buy measurement equipment, make sure the support handover — portal access, registration, calibration records, logins — is part of the negotiation. Whether it's a Hexagon support login or something else, that access has value. You only notice it after you need it.
When the Cheap Option Is Actually the Right One
I can hear the objection coming: 'Not every purchase needs a premium product. We have budget limits.'
Fair. In my opinion, cheap isn't the problem. Unconscious cheap is.
There are plenty of situations where a low-cost instrument is the rational choice. If a multimeter sits in a drawer and gets used once a month for basic checks, buy the modestly priced one. If the failure of a component doesn't stop production or create a safety risk, save the money. TCO can justify cheap purchases — if the loss, when the device fails, is small enough.
What drives me crazy is buying cheap for critical equipment without computing what a failure costs. I once watched an operations manager tell procurement, 'We need to keep costs down.' He meant the total cost of running the plant. What procurement heard was 'find the lowest quotation.' Both sides were using the same words and meaning different things. It took an emergency to reveal that gap. By then the cost wasn't a line item; it was lost production.
A policy like 'always buy the premium supplier' would be just as lazy. A policy of 'calculate the cost of failure before you compare prices' is harder to argue with.
So Before You Pick the Lowest Quote
I'm not telling anyone to buy the most expensive instrument on the market. I'm telling you to ask the question I ask every client when they are comparing two quotes: if this fails at the worst possible moment, what does that cost? Once you have that number, you'll know which of the two quotes is actually cheaper.
If you'd rather not do that math at 1:40 a.m., I understand. I'll be there anyway. But the easier call is to do it now — before the sensor dies, before the camera misses the hotspot, before you discover nobody has the support login. That's what TCO is for.
