Every procurement guide starts with “know your requirements.” That’s true, but it’s not the first question that matters. When I get a purchase request, the first thing I ask is: how much pain will be caused if this tool arrives late?
I’m a procurement manager at a 180-person precision manufacturing company. I manage a tooling and calibration budget of roughly $340,000 a year (maybe $330,000; I’d have to check the current plan). Over the past seven years, I’ve approved something like $180,000 in measurement and inspection equipment purchases, and I’ve made more than a few purchases I’d like to take back. The pattern I see is not about brand quality. It’s about matching the purchase decision to the timing of the need.
This framework applies whether you’re buying a Cognex vision sensor, a handheld oscilloscope, a multimeter for the maintenance bench, or a set of Starrett calipers.
There Is No One Right Answer
People often ask me which machine vision brand is best, whether a Fluke 289 true RMS multimeter is worth the money, or “where are Starrett calipers made?” These questions make sense, but the answer changes depending on the situation. The right purchase in an emergency is often the wrong purchase in a planned project.
The Three Scenarios I Use Before Buying Anything
- A line is down or a deadline is close. Late delivery is not an option.
- A planned upgrade. You have enough lead time to compare total cost.
- A small lab or new capability. You’re testing a process, not scaling it yet.
Situation A: The Line Is Down
In March 2024, our Cognex vision sensor on the packaging line started sending false rejects to the PLC. The part was still passable, but the sensor’s communication board was intermittent. We set up a manual check with two people, and that lasted about three hours before I started getting phone calls. The line was losing about $900 per hour in productive time.
I pulled up the Cognex product catalog, found the exact model we had installed, and called three distributors. The cheapest quote was 13% below list with a 5-day lead time. The second distributor had the part in stock and could get it to us by 10 a.m. the next morning for $380 extra. I signed the expedited PO (not that I loved it).
Was that a poor use of budget? No. The $380 charge bought certainty. A 5-day lead time from the cheaper supplier would have meant more hours of manual inspection, a delayed customer order, and a much larger cost than the air freight fee.
Uncertain cheap is more expensive than certain expensive.
In this situation, I don’t optimize for unit price. I optimize for the delivery date. I also ask the supplier to commit to a time in writing. If a salesperson says “probably by Friday,” that’s not a commitment. If the line is down, “probably” is the biggest risk.
Situation B: A Planned Upgrade
When there is no emergency, the decision changes. A planned project gives you time to build a total cost of ownership comparison. We did this in Q4 2024 for a new inspection station at the end of our assembly line. I specified a Cognex vision sensor in every quote. That may sound like I was favoring one brand, but the reason is practical: if you don’t specify the sensor, each integrator quotes a different hardware platform with different cables, software, and maintenance costs. You end up comparing apples to parts bins.
One quote was 19% lower than the second. It still looked like the winner until I added the missing pieces: no I/O cable, no lens bracket, no software license transfer, and no training day. After those items, the “cheap” quote was only 2% better, and it came with less support. We took the second quote.
Use the same logic for test equipment. When our maintenance group asked for a handheld oscilloscope and a higher-quality meter, the low quote did not include probes or software. The Fluke 289 true RMS multimeter had a higher sticker price, but it replaced two older meters and gave us data logging that saved at least an hour of troubleshooting each week. In our cost model, it paid back in about 14 months.
For planned purchases, I ask one question for every line item: what does this cost over three years, including spare parts, training, and the time to fix it when it fails? If a supplier can’t answer that with numbers, I keep looking.
Situation C: A Small Lab or a New Capability
The third situation is different. If you’re a machine shop, an R&D lab, or a startup that’s adding measurement for the first time, the cheapest capable tool is usually the right one. You don’t need a full machine vision system to check a hundred parts a week. You need a simple sensor, a good caliper, and maybe a handheld oscilloscope.
A common question in this situation is “where are Starrett calipers made?” It’s a sensible question because country of origin can affect lead time, repair options, and confidence in quality. The honest answer: it depends on the model. Some Starrett calipers are made in the U.S., some in Brazil, and some models are sourced from overseas. Model lines change. In our last order in January 2025, the exact caliper we bought was not made in the U.S., even though the distributor’s website said “assembled in USA.” Country-of-origin statements matter.
If you care about origin, ask for the model number and a written country-of-origin declaration before you issue the purchase order. If you don’t care about origin, buy a good mid-tier caliper with a calibration certificate and spend the savings on training.
The same “minimum credible tool” approach applies to a handheld oscilloscope. A $99 scope can show you a signal exists. If your team needs to measure noise and decode protocols, spend more. But for a small lab, first prove the need. Once you win a contract that requires in-line inspection, upgrade to a Cognex vision sensor and let the revenue cover it.
How to Decide Which Situation You’re In
I use three triggering questions before every purchase:
- What happens if this part is late? If the answer includes “line stops” or “customer order delayed,” you are in situation A.
- Do I have at least two weeks of lead time? If yes, you’re probably in situation B. Take the time to compare total cost, not sticker price.
- Is this purchase for a process that hasn’t generated revenue yet? If yes, you’re in situation C. Buy the minimum reliable tool and upgrade after demand is proven.
Here’s where my own bias shows up: I have been burned by “cheap” options more than once. The numbers said take the low-priced vendor. My gut said the supplier wasn’t ready. The cheap option had quality failures that cost $1,200 in rework. Since then, my procurement policy requires at least three quotes, but the tie-breaker is always the consequence of being late.
And before I forget: ask for delivery commitments in writing. Per FTC advertising guidelines (ftc.gov), claims should be truthful and substantiated. I don’t need a legal case—I just need an email that says “guaranteed on-site by Thursday.” If a supplier is willing to write it down, I trust the date more.
This was accurate as of Q1 2025. Product catalogs and prices change fast. Verify current model numbers, country-of-origin, and lead times before you budget.
The Bottom Line
Don’t ask “which sensor is best?” Ask which purchase matches the cost of being late. For a downed line, buy certainty. For a planned project, compare TCO. For a new lab, buy the smallest credible tool. That approach has saved us from several expensive mistakes, and it will work for your shop too.
Leave a technical question