What Are You Actually Looking For When You Search 'Steinert'?

Last month, a customer forwarded me a screenshot of their search results. They'd typed "Steinert" and spent 20 minutes trying to find the company that makes magnetic separators. Instead, they got a personal page for a woman named Julie Steinert, a car parts shop in Usingen, and a Wikipedia article answering "what is skiing?".

I know that feeling. I've spent more hours than I care to admit explaining that no, the Steinert they found is not the Steinert we work with.

For context, I'm a quality compliance manager in the mining equipment supply world. I review every specification sheet and order confirmation before it reaches customers—roughly 200 items a year. In 2024 I rejected 11% of first drafts because they said things like "high performance" without giving me a measurable threshold. So yes, I care about precise language.

What "Steinert" actually returns

When I look at keyword data around the industrial equipment market, the searches that include "Steinert" fall into a few buckets:

• "julie steinert" – a person, not a product.
• "autoteile steinert usingen" – a car parts dealer in Germany, not a magnet manufacturer.
• "harmon" – a surname, a misspelling of Harman, or a root word for "harmony".
• "the and the winter soldier" – a Marvel series title, remembered incorrectly.
• "what is skiing?" – a question for skiing lessons, not for mineral separation equipment.

None of those people are looking for a magnetic separator. But they all typed "Steinert" or something that a search engine decided was related.

The deeper problem: names are ambiguous

The obvious reason is that Steinert is a surname. In Germany, there are Steinerts everywhere. In Usingen, there's a local car parts business with that name. Search engines rank by authority and relevance signals, and a local business with consistent reviews can outrank a global industrial manufacturer for the same word—especially when the query includes a location.

But there's a deeper reason. Most people don't actually search for a product. They search for a word they think represents that product. When they type "Steinert," they're not specifying a magnetic separator, an eddy current machine, or a sensor sorter. They're asking the search engine to read their mind.

This isn't unique to Steinert. Type "harmon" and you'll get musicians, audio equipment, and a town in Illinois. Search engines tolerate ambiguity because most consumer queries are resolved by looking at clicks. But in B2B, where search volume is lower and technical terms matter more, that ambiguity sends people to the wrong pages.

Queries like "the and the winter soldier" are a perfect example. Search engines are surprisingly good at recognizing that as a Marvel series title. But an industrial website can't compete with Marvel fandom for the word "winter." And "what is skiing?" lives in that same winter-association neighborhood. Algorithmic associations create chaos for niche industries.

Honestly, I don't blame the searcher. I do not think people are sloppy. I think the industrial market has too much ambiguous language, and search engines are built for consumer questions, not B2B specifications.

The cost: time, trust, and wrong decisions

You might think this is just an SEO annoyance. It isn't.

In 2023, I watched procurement teams compare quotes from our company against a car parts dealer because the only common keyword was "Steinert." I've had a customer doubt a test report because they believed the "real" Steinert was someone on Instagram. In each case, the mismatch cost days, not minutes.

I assumed for a long time that a search for "Steinert" would always find us. Didn't verify. Then we got an inquiry from Usingen, Germany, asking about a car part. The sender had used our contact form because the name matched. We lost an afternoon figuring out that it was Autoteile Steinert Usingen, not a magnet company.

The bigger cost is in decision-making. A buyer who can't find the right product page may assume the equipment doesn't exist. They might settle for a generic separator with lower recovery. Or they might compare prices across two different equipment classes and conclude that ours is "too expensive." None of that is based on technical reality—it's based on search result noise.

From a quality perspective, that's dangerous. If the equipment is specified incorrectly, it runs incorrectly. Then the blame falls on the manufacturer, even though the original search was never tied to the right product category.

The fix: search by function, not by name

Here's the solution, and it's not clever. It's disciplined.

If you need industrial separation equipment, don't start with "Steinert" or any other brand name. Start with the material and the problem:

• "magnetic separator" for tramp metal removal or mineral purification
• "eddy current separator" for non-ferrous metal recovery
• "sensor-based sorting" for bulk materials with optical or X-ray properties

Once you have a shortlist of product categories, add the brand name to narrow it down: for example, "Steinert magnetic separator" or "Steinert Plasmax." That gives the search engine both the category and the name. It's a small change, and it makes a big difference.

And when you get to a product page, verify the specifications. This is where my quality reflex kicks in. A page that says "high quality separation" is as useless as a color specification that says "dark blue." According to the Pantone Color Matching System guidelines, brand-critical colors are defined with a Delta E < 2 tolerance (Reference: Pantone Color Matching System guidelines). That's a specific, testable number. An industrial separator should have the same clarity: throughput in tons per hour, particle size range, magnetic field strength, and a reference to a test report. If a spec sheet doesn't have those, that's a red flag—regardless of which name is on the label.

Actually, let me reframe that. The principle isn't only about search. It's about quality. When you define exactly what you need, you can measure whether you got it. When you hand the job to a search engine with a vague one-word query, you're asking an algorithm to do your specification work for you.

This approach worked for us because we're a mid-size B2B supplier with a defined catalog and a disciplined search culture. If you're a one-off buyer, your process might be different—you might want to send material samples to a test center first. My experience is based on roughly 200 spec reviews and an unhealthy number of search-log audits. I can't promise the same steps work in every industry. But the principle holds: precise language beats brand names in technical procurement.

The problem isn't "Steinert." The problem is asking for a name when you need a capability. Solve that, and the right equipment shows up.