Every order passed inspection. That was the problem.
I review incoming product before it reaches our customers—roughly 200 SKUs a year, and on one annual contract around 50,000 units. For three years running, the same failure repeated: we ordered Fluke multimeters with exact model numbers, exact quantities, exact delivery windows. Everything matched the PO. And something was still wrong.
Not wrong in a way that shows up on a receiving dock. Wrong in a way that shows up six weeks later, in a customer email, usually phrased as this doesn't match what we asked for. If you've ever had a lot rejected for a reason nobody can quite name, you know how expensive that ambiguity gets.
Here's what I eventually worked out: we weren't ordering instruments. We were ordering catalog numbers and calling it a specification.
The mistake I made first
In my first year, I made the classic specification error: I assumed the word "standard" meant the same thing to every supplier. We had a spec sheet that said standard test leads. In my head, that meant leads rated to the voltage of the application. One supplier read it as whatever leads happen to ship in the box. Both readings were defensible. Only one of them was safe.
That batch never got rejected for a defect. It got flagged because a technician looked at the rating printed on a connector and asked a question nobody in our chain could answer. That was the first time I understood that a PO is not a spec, no matter how detailed the PO looks.
The model number is a product, not a specification
When a procurement specialist writes Fluke 117 on a purchase order, they believe they've specified a measurement instrument. What they've actually specified is a part number. And a part number carries less information than most buyers assume.
The Fluke 117 is a rugged, general-purpose true-RMS unit with non-contact voltage detection, and for a lot of commercial electrician work it is genuinely the right answer. But "the 117" tells you nothing about the measurement category your application demands, the calibration state at receipt, the traceability chain behind that calibration, or whether the test leads in the box are rated for the environment where they'll actually be used. Those are four separate specifications, and only one of them ever makes it onto the PO.
What I mean is that the gap isn't between brands—it's between what a buyer thinks a model number communicates and what a model number actually communicates, which is basically shape, product family, and a nominal feature set, and nothing at all about the conditions the instrument will be used under.
Three places the gap opens
1. The measurement category nobody reads
Measurement categories come out of IEC 61010-1, the safety standard for electrical test and measurement equipment. CAT II, CAT III, and CAT IV describe where in an installation an instrument is rated to be used, and those ratings are not interchangeable. An instrument used in the wrong category isn't just out of spec—it's a hazard.
The problem is that CAT ratings are printed on the face of the meter in small type and appear on almost no purchase orders. When someone asks me what the best Fluke multimeter for HVAC work is, the honest answer is that it depends on where the technician is standing. A CAT III rated unit at a distribution panel and a CAT II rated unit at a control circuit are not the same tool, even when the catalog photos look similar.
2. Accessories are part of the measurement
I used to treat test leads, probes, and sensors as consumables—things you order after the real decision has been made. That was backwards.
A meter's stated accuracy assumes an intact, correctly rated measurement path. Test leads carry their own voltage rating and their own CAT rating. Sensors carry their own accuracy class, their own temperature range, and their own response time. If sensor sourcing is treated as an afterthought, you can end up with a perfectly calibrated meter reading through a lead that was never rated for the circuit it's touching.
The same logic applies to clamp meter sourcing. Jaw size, current range, and true-RMS capability matter far more than whether the catalog photo looks right. Ordering by photo is how you end up with a clamp that won't fit around the conductor you actually needed to measure.
3. Calibration state and traceability
Here's the part that took me longest to internalize: an instrument's accuracy is a claim, and a claim is only as good as its evidence. That evidence is a calibration record with traceability back to a recognized reference, plus a stated calibration interval.
I want to say we had 14 units flagged in one audit for missing calibration records, but don't quote me on the exact number—it was somewhere in that range. What I remember clearly is the auditor's question. Not is this meter accurate, but how do you know. We didn't have a good answer, because we had never written calibration state into the order.
Per FTC guidance (ftc.gov), advertising and product claims must be truthful and not misleading, substantiated with evidence, and clear about endorsements. That standard applies to the accuracy claims your business repeats downstream—not only to the manufacturer's marketing.
If you're repeating an accuracy figure to your own customer, you own the substantiation for it. That isn't a legal technicality. It's a procurement requirement.
What the gap actually costs
Never expected the expensive failure to be the boring one. Turns out the counterfeit part everyone warns you about is easier to catch than a genuine unit that's correct in every respect and documented in none of them.
The obvious cost is a rejected lot. The real cost is everything orbiting it:
- Rework time—the hours spent documenting why something was wrong, plus the replacement shipment and the expedited freight to cover it.
- Audit exposure—missing calibration records turn a routine review into a corrective action plan.
- Customer trust—one wrong spec reaching a contractor becomes a conversation about whether you understand their work at all.
- Safety—the one line item with no dollar figure attached, and the one that matters most.
We once had a batch where the paperwork said one thing and the physical rating said another. It didn't injure anyone. It cost us a redo anyway, and it changed how we write specifications permanently.
How to evaluate a tester supplier without trusting the datasheet alone
I'm not going to hand you a twelve-point checklist. The fix is smaller than that, and honestly the checklist is what got us into trouble in the first place—it looked thorough enough that we stopped asking questions.
When we started putting these four questions in writing, our rejection rate dropped and our lead times actually got shorter, because we stopped having to redo orders:
- What is the measurement category rating for this instrument in this application, and which standard defines it? If the answer is a model number, you're talking to a reseller, not a supplier.
- What is the calibration state at receipt, and what is the traceability chain? Ask for the record, not the promise.
- Are the included leads, probes, and sensors rated for the same environment as the meter? Get the rating in writing rather than implied.
- Can you confirm the accessory ecosystem—replacement leads, sensor compatibility, calibration service intervals? A meter you can't maintain is a meter with a fixed end date.
Notice that none of those questions are about the manufacturer's brand. A non-branded unit isn't automatically a bad unit. It's an undocumented unit until somebody answers the four questions above. The documentation is the differentiator, not the label.
What I'd tell a new buyer
We didn't have a quality problem. We had a specification problem that looked exactly like a quality problem, which is why it took three years and a lot of unnecessary freight to fix.
The shift was moving from ordering part numbers to ordering documented requirements, and then verifying the documentation at receipt instead of trusting it at purchase. Not glamorous work. It's basically writing down what you already assumed, and then making the supplier confirm it before the pallet moves.
If you're placing a Fluke multimeter order this quarter, the model number is the easy part—you already have it. The part worth spending an afternoon on is the four lines above it on the spec sheet. That said, I'll admit we still get it wrong occasionally. We just catch it at the dock now instead of at the customer.
Sarah Okonkwo
Sarah Okonkwo is an electrical test and measurement analyst specializing in multimeters, clamp meters, installation testers, sensors, and power-quality monitors. She uses IEC 61010-1 and IEC 61010-2-032 safety requirements for test and current-clamp equipment, then applies IEC 61000-4-30 methods to power-quality parameters while examining category ratings, accuracy, resolution, bandwidth, harmonics, voltage events, and uncertainty. She helps technicians and buyers select instruments, set up safe measurements, interpret readings, and diagnose faults within the intended circuit environment.