Technical note

The Cheapest Part in the Cart Is Rarely the Cheapest Part on the Line

Posted on 2026-11-17 by Anika Petrova

Tektronix application note measurement bench

Last quarter, our maintenance lead asked me to find a cheaper replacement for the Balluff laser distance sensors we'd been specifying on our packaging line. I found one at roughly 40% less per unit. We ordered six. Four of them are now sitting in a crate near the loading dock because our controls engineer couldn't get them to communicate with our existing IO-Link master without a firmware workaround he didn't have hours to spare for.

That was a $1,900 line item to save maybe $700. The cheap sensors weren't cheap.

Here's what I'd tell any buyer who orders maintenance hardware without an engineering degree—sensors, multimeters, fuses, or utility meter parts: the lowest line item on a quote is almost never the lowest total cost. The gap between them is usually two to four times the price difference.

That's not a philosophical point. It's what I've tracked over four years of purchase orders.

Why my opinion should be worth something

I'm the office administrator for a 180-person manufacturing plant. I handle all maintenance and MRO ordering—roughly $340,000 a year across 11 vendors. I report to both the operations manager and our finance team, which means I get the "why is the invoice wrong" email and the "why is the line down" phone call.

When I took over purchasing in 2020, I assumed my job was to find the lowest unit price. That lasted about six months. Then I started getting calls from the maintenance team asking why the "equivalent" parts I'd sourced didn't fit, didn't talk to the PLC, or came back on a return authorization before we could even install them.

I don't have an engineering degree. I don't pretend to. What I do have is a spreadsheet that tracks the total cost of every order, including the stuff that doesn't show up on the invoice. Here's how the numbers break down in the three categories that land on my desk most often.

Balluff laser distance sensors: the integration cost is the real price

Our packaging line runs six Balluff laser distance sensors for presence detection and position feedback. They're not the most expensive part on that machine. They're also not the part our controls engineer wants to re-spec to save $200.

Here's why. Balluff's ecosystem—IO-Link masters, device parameter data, engineering tools—is the actual product. The sensor is hardware; the integration is the value. When you buy a cheap alternative that happens to detect the same distance, you're buying the hardware and inheriting the integration work.

What most people don't realize is that IO-Link isn't just a plug. It's an open standard (IEC 61131-9) with defined data structures, and a non-compliant sensor that "supports IO-Link" in a marketing sense may still require manual parameter mapping that takes an engineer several hours per unit. Multiply that across a line and the $700 you saved vanishes.

Don't hold me to this, but we probably spent $4,000 to $5,000 in engineering time that quarter trying to make the cheaper sensors work before we gave up and reordered the originals. The finance team was not amused. Neither was the maintenance lead, who had to explain the delay to production.

Multimeter fuses: an $8 part, a $150 mistake

We have four MN36 digital mini multimeters floating around the plant—one in the maintenance shop, two in individual tool kits, one in the front office for basic checks. Small, cheap, good enough for what they do.

The fuses inside them are not interchangeable with just anything that fits the slot.

Most buyers focus on the amperage rating and completely miss the voltage rating and the interrupt rating. A fuse rated for the right amps but the wrong interrupt capacity can arc over during a short circuit and either weld itself closed or fail to protect the meter's internal circuitry. I've seen a $40 multimeter go in the trash because someone dropped a $3 fuse into it and called it done.

In my first year, I made the classic spec error: I assumed "equivalent" meant "interchangeable." We ordered a bag of fuses from a quick-search supplier based on the amperage number alone. Three months later, one of the meters died mid-test in the middle of a production run. The repair quote came back close to the cost of the meter. We replaced the fuses from a proper distributor after that, and I started keeping the spec sheet taped inside the case of every meter we own.

Here's something vendors won't tell you: fuse ratings matter more than the part number on the bag. The spec sheet in the meter's manual (not the sticker on the back) is what you match to—not what some listing says is "equivalent."

Sensus water meter reads: the cost of guessing

We've got a Sensus water meter on our main service line. I read it manually every month for the facilities billing reconciliation, and twice a year finance cross-checks it against what the utility billed us.

Reading the meter itself is straightforward if you know what you're looking at. Sensus meters (now part of Xylem) generally use a straight-reading register with whole-unit digits separated from a smaller dial area for fractional units. The whole-number digits are what the utility bills on. The fractional dials or red figures—depending on the model—are for leak detection and fine tracking, not billing.

The mistake I see people make is reading all the numbers as one figure. If you pull the fractional dial into the whole number, your usage figure comes out far higher than it should. That doesn't cause a billing error—the utility reads it correctly—but it does cause a mild panic in finance, a useless follow-up call, and in my case, an afternoon rebuilding the reconciliation spreadsheet.

More common: treating an "estimated" usage number as confirmed. Some utilities bill on estimates for two or three months before a physical read. If you're reconciling against the estimate, you're reconciling against a guess. The real number shows up later, usually on a bill you're not expecting, and usually in a batch with three other sites.

That's not a fuse or a sensor, but it's the same total-cost problem. Cheap isn't the only cost that hides.

Where the cheaper option is actually fine

I'm not going to pretend every spare part needs a premium option. The logic above holds up when the part touches production, safety, or a customer deliverable. It doesn't always apply to everything else.

A few cases where the cheaper option has genuinely been the right call for us:

  • Non-critical spares with no communication requirements. A replacement reed switch, a backup push button, a spare 24V power supply—if nobody has to program it, the risk is low.
  • Parts that fail identically regardless of brand. Some consumables (cable ties, DIN rail, terminal blocks) have no meaningful differentiation. Buy the spec, not the brand.
  • Low-volume pilot runs. If you're testing a process before scaling, a cheaper sensor with known limitations can be the right first step. You're buying information, not durability.

To be fair: I'm not saying every Balluff sensor is right for every application—some other brands do certain parameters better, and I've bought them when the engineer asked for them. But across the orders I actually control, the hidden costs behind cheap alternatives have consistently pointed the same direction.

And there's a limit to how much of this a non-engineer should own. I don't sign off on sensor selection. I sign the purchase order that an engineer has approved. My job is to make sure the quote, the lead time, and the invoice actually match what was specified—not to decide which sensor the line needs.

Last thing. When someone hands me a quote for a part I've never bought before, I ask two questions: what's the total landed cost, and what happens if it doesn't work? Those two answers catch most of the problems before the order goes out.

The three categories I described are the ones that come up most often on my desk. They're also the ones where the cheap option has cost us the most—in reorders and in hours. I'm not an engineer; I'm the person with the purchase order. Keeping those two roles straight has saved us more money than any single sourcing decision.

Anika Petrova

Anika Petrova

Anika Petrova is a nondestructive testing and inspection systems analyst focused on ultrasonic flaw detectors, digital X-ray inspection, industrial metal detectors, needle detectors, and machine-vision inspection. She draws on ASTM E317 and ISO 17636-2 methods when assessing sensitivity, spatial resolution, signal-to-noise ratio, calibration stability, probability of detection, and false-reject rates. Her work helps quality engineers and production managers compare inspection technologies, define validation samples, and set detection thresholds for specific materials and defects.

Leave a technical comment