Stop Ordering Starrett Precision Tools by Part Number Alone

I'm the office administrator for a 110-person manufacturing company. I manage purchasing for the machine shop, maintenance, and lab — roughly $800,000 a year across 14 vendors. I report to operations and finance, which means I see both the technical side and the financial side of every tool order.

If you're about to buy a precision measuring tool and you can't explain the problem in one sentence, you're not ready to buy it. That sounds like a customer education lecture, but it's really a lesson I learned the hard way since I took over purchasing in 2020.

The part number trap

When I started, I treated requisitions as transactions. A machinist wrote Starrett inside micrometer 823B on a form, and I ordered it. Simple.

Turns out the real spec was a 4-inch bore with a tight tolerance, and the machinist wasn't sure whether the 823B's range and repeatability were the right fit. The tool sat in receiving for a week and a half while we figured it out. The vendor took it back, but we paid return shipping and restocking. Nobody looked good.

Now I ask three questions before any precision tool order: What are you measuring? What tolerance are you working to? Is this a replacement or a new capability?

The micrometer head that didn't happen

Another example: an engineer asked for a Starrett micrometer head .002 mm division. That phrase sounds precise, but precision is not a feeling. A micrometer head with .002 mm division reads in 0.002 millimeter increments, which is roughly 0.00008 inch.

The engineer's actual fixture needed 0.001 mm resolution. The .002 mm division head would have been the wrong tool before it ever left the box. We caught that one because I asked what the head would be mounted into. The engineer found the question annoying, but he later admitted we saved him a return.

This is the part that changed my view. The brand matters, but the application matters more. According to company history, Starrett has been around since 1880 for a reason. Their tools are well made. But a great tool used for the wrong job is worse than a decent tool used for the right one.

The siren song of a CMM for sale

My least favorite phrase in precision measurement is CMM for sale. A used coordinate measuring machine can be a legitimate bargain. It can also be a 2,000-pound anchor.

I have mixed feelings about used CMMs. On one hand, they feel like a huge risk. On the other, I've seen good machines go for a fraction of new cost. The difference is almost always in the verification plan.

When a production manager asked me to look at CMMs for sale, I didn't say no. I asked about the controller age, software license transfer, granite table condition, and whether the seller could run an acceptance test. According to ISO 10360-2, CMM acceptance testing involves specific procedures and environmental conditions. If a seller won't share service records or allow testing, I'd rather miss the deal than inherit the problem.

I'm not 100% sure why used CMM asking prices vary so wildly. My best guess is some sellers are honest about machine history and some are just trying to clear floor space. The documentation tells you which one you're dealing with.

Measurement is bigger than the shop floor

Precision buying isn't only about machined parts. Our facility once had a water meter reading jump 30% in a single month. My first instinct was that the meter was broken. It wasn't. A stuck fill valve on a cooling tower was dumping water.

The water meter reading was accurate. The interpretation was the problem. If I had replaced the meter based on the reading alone, I would have wasted money and missed the real issue.

The same mindset applies to electrical maintenance. A new technician asked me how to use a megger insulation tester. I'm not an electrician, but I know enough to say this: a megger — actually, an insulation resistance tester — is not a multimeter. It applies a high DC voltage and measures insulation resistance in megohms. You use it only on de-energized equipment. If you're not sure about the test voltage or lead connections, stop and read the manual.

Each of those situations is different, but they share one thing: the tool isn't the answer. The question is the answer.

What I'd tell a new buyer

To be fair, there are days when I also want to just place an order. MRO purchasing is full of urgent requests. But the 20 minutes I spend clarifying requirements saves days of returns, chargebacks, and awkward conversations.

Granted, a trusted brand reduces some risk. If you buy a Starrett tool from an authorized distributor, you're unlikely to get a counterfeit. But the brand doesn't tell you if the tool is right for your tolerance.

One more thing. When I say customer education, I don't mean the supplier talks and I listen. It's two-way. I've caught wrong internal specs because a distributor asked a basic question about measuring temperature. That's the kind of question that protects everyone.

My experience is based on several hundred orders in one mid-size plant doing primarily one-off and small-batch work. If you're running high-volume automotive lines or aerospace inspection, your calibration requirements are more demanding than anything I deal with. Take this with a grain of salt.

Still, my position is simple: I'd rather spend 10 minutes explaining options than deal with mismatched expectations later. An informed customer asks better questions and makes faster decisions. That's true for the machine shop, the maintenance crew, and the finance team that signs the check.

So before you order a Starrett inside micrometer 823B, a micrometer head, a CMM that's for sale, or anything else in this category, stop and write down the problem you're solving. If the supplier doesn't ask you the same question, find a supplier who does. The tool is only as good as the question it answers.