I used to think the problem was price
I'm not an engineer. I'm the person who signs the purchase orders. For the past 6 years, I've managed an annual automation and lab equipment budget of about $180,000 at a mid-sized controls integrator. I've negotiated with 30+ vendors, tracked every invoice, and maintained a TCO spreadsheet that I'm slightly embarrassed to enjoy.
Last quarter, a project engineer asked why we kept specifying a SICK photoelectric sensor when another brand quoted 40% less. It's a fair question. On the surface, this is a price problem. But after auditing our 2023 spending, I found that our budget overruns almost never came from the items with the highest unit prices.
Here's the thing: they came from the cheapest items that failed at the worst possible time.
What's really behind the price gap
People think expensive vendors deliver better quality. Actually, vendors who deliver quality can charge more. The causation runs the other way. The difference is not markup—it's what they did before shipping the product.
Take the SICK encoder AFM60A. On the datasheet, it has a list of mechanical and electrical ratings. Those numbers mean something because SICK tests them under defined conditions. A generic encoder can print the same numbers on a brochure. But can they prove it? I would rather buy a component with verified performance than one with a copied spec line.
To be fair, not every application needs that level of verification. If you're running a conveyor in a clean, dry, temperature-controlled warehouse, a simpler sensor may be the rational choice. Our situation is different: our machines see vibration, washdown chemicals, and occasional operators who are not gentle. In that context, a SICK photoelectric sensor is not a luxury—it's a way to keep the line moving.
I've also learned to watch for the 'same specs' trap. Two products can list the same IP rating, same response time, same temperature range. The difference is how those specs were obtained. One was tested by a recognized lab; the other was estimated. Good luck seeing that on a comparison chart.
The cost of ignoring context
A line stop at our facility costs roughly $2,400 per hour in labor and overhead. If a sensor fails twice a year and each replacement eats three hours, that's $14,400—plus the part. The difference between a reliable sensor and a cheap one is often less than $150.
Encoders are even less forgiving. In one positioning application, a mispositioned part created scrap for half a shift before someone noticed. We estimated the loss at $4,800. Rework and paperwork added more. That's why our standard for servo-feedback and positioning is the SICK encoder AFM60A. It's not the cheapest absolute encoder I've quoted. It is the one whose published shock and vibration ratings line up with what the machine actually experiences.
Worse than a failure is a failure that looks like something else. When a cheap encoder drifts, your first instinct is to blame the controller, the wiring, or the operator. Hours get burned chasing a ghost. The cheap part becomes the most expensive part on the line. Not ideal. Workable, but not ideal.
The same logic applies in the lab
The lab budget gets the same treatment from me, and it annoys people. They ask why I care about scientific pipettes when our applications aren't pharmaceuticals. I care because bad data costs more than good equipment.
We compared two pipettes with similar claimed accuracy. One came with an ISO 17025 calibration certificate. The other had a sheet saying it was factory tested and certified. The certified pipette cost 30% more. Over three years, the cheaper one needed two extra recalibrations, and one proficiency test came back with a warning that took a week to resolve. The hidden cost was not the pipette. It was the investigation time.
According to ISO 8655, pipette performance is verified at defined volumes. That means accuracy claims are only meaningful if you know the volume, the tip, and the operator technique. A price quote alone doesn't tell you any of that. If you do quantitative work, buy the calibration history, not just the plastic.
The same principle shows up in analytical technologies. HPLC is the example I reach for most. In liquid chromatography, sample preparation errors become peak area errors. It doesn't matter if the HPLC itself is perfect—if the sample was introduced with an inaccurate pipette, the result is suspect. I've seen a $0.50 mistake trigger a $5,000 investigation.
And what is a megger insulation tester, exactly?
One more example from a different part of the budget: electrical maintenance. If you've ever asked, 'what is a megger insulation tester?'—it's a device that applies a high DC voltage, usually 250V, 500V, or 1000V, to measure insulation resistance in megohms. It tells you whether motor windings, cables, or switchgear are degrading before they short out.
We own one. We use it maybe four times a year to check motors and power cables. For our intermittent indoor use, a mid-range unit is fine. If you're doing daily predictive maintenance in a dirty plant, you need a more rugged one—and maybe one with data logging. If you never touch high-voltage equipment, you might not need one at all. Context drives the decision.
That's the honest answer. No single 'best' exists. What exists is a match between the equipment's tested capability and your actual operating environment.
What we changed, and what I'd tell you
After tracking our 2023 spending and the failures behind it, our procurement policy now requires three quotes and one simple question: under what conditions was this spec verified?
- Compare total cost, not unit price. Include downtime, calibration, rework, and disposal.
- Demand documented evidence for claims like IP ratings and accuracy. Per FTC advertising guidelines (ftc.gov), marketing claims need substantiation—I hold technical datasheets to the same standard.
- Define your real environment before comparing models. Dust, moisture, temperature, vibration, operator skill—all of it changes the right answer.
- Don't buy extra capability you don't need. A SICK photoelectric sensor or AFM60A encoder is a good default for harsh, high-stakes lines. For a low-risk application, a simpler product might be the better buy.
- For lab work, budget for calibration, not just the pipette or HPLC technologies. Ask for the certificate and the standard it was tested against.
Granted, this approach takes more upfront work. It also saves time later. I do not mean to say that every purchase should be premium. For our environment—a mid-size shop with harsh, intermittent conditions and no redundant line capacity—the math points one way. If your line is different, the math might be different. That's the honest version, and it beats pretending there's one perfect answer.
