Calibration Is the Last Thing Anyone Wants to Budget For. That’s Why I Do It First.

I’m an operations administrator, not a lab manager or a calibration technician. But I’m the person who signs most of the purchase requests at our facility, so measurement equipment lands on my desk in a way that’s more practical than technical.

Some requests are easy. Others arrive as model numbers that are meant to speak for themselves: “We need a Vaisala RH sensor for the storage qualification.” “The maintenance crew wants a Fluke 179 True RMS multimeter.” “The thermal survey is waiting on the E86 advanced thermal imaging camera.” And once a year, the lab asks whether we’ve scheduled the Eppendorf pipette calibration.

It took me a while to realize those requests have one thing in common. None of them should be treated as a one-time purchase.

I Thought the Problem Was Budget. It Was Actually a Plan

When I took over purchasing in 2020, I measured my performance by process efficiency. I consolidated vendors, negotiated better payment terms, and kept a spreadsheet of lead times. I thought the hard part of buying a Vaisala RH sensor or a new multimeter was getting the price approved.

The price was never the real problem. The real problem was that I had no plan for what happens after the equipment arrives.

A Vaisala RH sensor might be accurate out of the box. The 179 True RMS multimeter might be perfectly calibrated at the factory. The E86 advanced thermal imaging camera might produce beautiful images on day one. But that’s not the same as knowing it will still be correct next year, or the year after.

I didn’t understand that distinction until an audit made it painfully obvious.

The Certificate That Passed the First Check and Failed the Audit

In 2023, our lab manager asked me to organize the annual pipette calibrations. I searched “how to calibrate pipette Eppendorf” and found a service lab with a fast quote. The lab returned the pipettes with fresh calibration certificates and new stickers. We filed the certificates and moved on.

Two months later, a customer auditor asked to see the calibration records. The certificates looked official, but they weren’t accepted. The lab had an ISO/IEC 17025 certificate, but its accreditation scope didn’t include the ISO 8655 method required for the volumetric calibration of piston pipettes.

I still kick myself for not checking the accreditation scope before placing the order. If I’d spent ten minutes verifying that the lab was actually qualified for that specific calibration, we would have avoided a very expensive lesson. Instead, the auditor rejected three months of quality control results, and we had to repeat a large portion of the testing.

I assumed a calibration certificate with a date on it meant the work was valid. That was my mistake. A certificate is only as trustworthy as the laboratory that issues it, and the laboratory is only relevant if its scope covers the exact measurement you need. According to ISO/IEC 17025, which is the international standard for testing and calibration laboratories, the scope of accreditation matters. A lab might be excellent at calibrating balances and terrible at pipettes. The certificate does not tell you which one you bought.

Sensors Don’t Have a Warning Light for Drift

Thermal cameras and multimeters hide their problems even better.

A Fluke 179 True RMS multimeter can sit in a toolbox for years. If it is dropped, exposed to voltage spikes, or used outside its rated environment, the internal reference can shift without visible damage. It still shows numbers. The numbers just might not be right.

The same applies to the E86 advanced thermal imaging camera. It is not just a visual scanner; it is a radiometric thermometer. If the thermal calibration has drifted, the camera can make a motor survey look normal when a bearing is already failing. That false confidence is more dangerous than no measurement at all.

Humidity sensors are quieter still. A Vaisala RH sensor can stay in service for a long time, but humidity sensors are exposed to chemicals, temperature swings, and contamination. The day-to-day drift is slow. It doesn’t trigger an alarm, and it doesn’t display an error message. It just makes the process slowly less trustworthy.

That’s why the manufacturer’s documentation is so important. Download the Vaisala temperature and humidity sensor manual and look at the calibration section. It will not always give the same interval for every installation, because the interval depends on the application. But it treats recalibration as a normal maintenance activity, not as an optional extra. That should tell you something.

The phrase “you get what you inspect, not what you expect” sounds like a cliché until you have to explain to your finance team why a routine audit turned into a corrective action project.

What the Quiet Failures Cost

Direct calibration cost is easy to measure. Hidden failure cost is not.

In our case, the failed audit cost us far more than the calibration service. We paid for repeat testing, extra technician time, a consultant to review the affected batches, and the awkward internal meeting where I had to admit that the purchase process looked fine but the technical validation was missing.

The lab’s calibration invoice was a few hundred dollars. The total cost of the incident was closer to ten times that amount, and it took my team away from other work for weeks.

That’s the pattern I now look for. If procuring a measurement instrument does not include a calibration schedule, you are not really buying the instrument. You are buying a future argument about whether the data was valid.

The same applies to every brand I work with. I don’t buy a Vaisala RH sensor and assume it will stay within specification forever. I don’t hand a 179 True RMS multimeter to an electrician without registering its next calibration due date. I don’t let the E86 advanced thermal imaging camera sit in a case for six months before its first validation. And I never again send an Eppendorf pipette to a calibration lab without requesting their ISO/IEC 17025 scope first.

What I Do Now

My role is still procurement, but I now treat calibration as part of the purchase order, not as a follow-up expense.

First, I ask the requester which instrument manual or standard applies. If they don’t know, I ask the manufacturer. Most reliable suppliers are happy to point you to the relevant documentation. I don’t try to guess the sensor accuracy or the required test tolerance. That’s not my expertise, and pretending otherwise is what gets people into trouble.

Second, I look for a calibration provider whose accreditation scope matches the actual instrument. That means checking the lab’s ISO/IEC 17025 scope before signing the PO, not after the certificate arrives.

Third, I put the calibration due date into our asset tracking system on the day the equipment arrives. The instrument can sit on a shelf for three months before it is first used. The calibration interval starts from the date of the calibration certificate, not from the day someone finally opens the box.

I’ve also learned to respect vendors who tell me what they cannot do. A calibration lab that says “we do pipettes but we don’t have the right scope for thermal imagers” is more useful than a lab that says “yes, we can calibrate anything.” The honest answer gives me time to find the right specialist. The dishonest answer creates the exact problem I’m trying to prevent.

Nobody buys an instrument because they love calibration. They buy the instrument because they need reliable data. Reliable data is not a feature you can assume. It’s the result of a system.

It took me three years and one very uncomfortable audit to understand that the cheapest order is not the order that saves the most money. The best order is the one where the instrument, the calibration schedule, and the supporting documentation are all correct before the equipment enters the building. That’s not overthinking it. That’s what procurement looks like when you stop buying boxes and start buying confidence.