Same Title, $33,000 Apart: Why Engineering Searches Stall Before They Start

An empty engineering or skilled-trades seat doesn't sit quietly. A line runs short-handed, a project slips, overtime lands on the people who stayed, and a few of them start taking recruiter calls. Most of these seats stay open for reasons that were knowable before anyone posted the job.

I've watched this happen for close to thirty years. The search doesn't fall apart in week ten. It was already lost in week one, and nobody looked.

Four things cause it, over and over.

The pay was benchmarked against the wrong place, and the wrong industry

An employer pulls a national median, or whatever a salary site returned, and sets a band against it.

The national median for mechanical engineers was $104,110 in May 2025, according to the Bureau of Labor Statistics. But nobody hires at the national median. Massachusetts ran about $119,300. New Hampshire sat near $103,660 and Maine near $102,430. A band that looks fine against one of those numbers is a pay cut against another, and a candidate in Greater Boston can see both.

Industry matters even more than geography. The same BLS data shows what mechanical engineers earned at the median in different industries:

• Scientific research and development services: $132,040

• Computer and electronic product manufacturing: $106,870

• Transportation equipment manufacturing: $105,560

• Machinery manufacturing: $98,620

Same job title. A $33,420 spread between the top and bottom of that list. The skilled trades show the same pattern: BLS puts the median tool and die maker at $78,200 in transportation equipment manufacturing and $56,600 in primary metal manufacturing.

Benchmark against an all-industries number and you get a band that reads fine on paper and loses every competitive situation.

Headcount was mistaken for supply

This is the one that catches people hardest.

BLS counted about 298,500 mechanical engineers in the U.S. in 2025. Sounds like plenty. Now ask the questions that matter. How many work in your industry? About 14 percent are in machinery manufacturing, 10 percent in transportation equipment, 6 percent in computer and electronic products. How many have the specific experience you need, whether that's precision machining, controls, validated processes, or a particular CAD platform? How many would consider a move? How many of those live within a reasonable commute?

Run those filters and a healthy-looking talent pool shrinks to a short list, nearly all of them employed and not looking.

The skilled trades are tighter. BLS projects tool and die maker employment to fall 9 percent over the next decade. It also expects every one of the roughly 30,400 annual openings for machinists and tool and die makers to come from replacing people who leave or retire, not from growth. Nobody is adding to the pool. Deloitte and The Manufacturing Institute estimate U.S. manufacturing could need 3.8 million new workers by 2033, with as many as 1.9 million of those jobs going unfilled.

The headcount wasn't wrong. It was just answering a different question than the one that mattered.

The real competition was never mapped

Employers map their competition by looking at companies like themselves. A machine shop lists other machine shops.

That's usually the wrong list.

On the Seacoast and in southern Maine, the real competitor is often the shipyard. Portsmouth Naval Shipyard has nearly 8,000 civilian employees, and a recent congressional letter notes it is working to add more. Bath Iron Works, Maine's largest manufacturer, had about 6,700 employees at the start of 2023, and a study of its hiring found starting wages rose an average of 15 percent a year from 2021 to 2023. The Navy's own leadership has said it needs 250,000 new shipyard workers over the next decade.

Beating another shop's hourly rate doesn't touch that problem. The shipyard offers scale, job security, benefits, and a name that looks good on a resume. If you don't know what you're actually competing against, you can't build a message that beats it.

The territory crossed a line candidates won't

A posting covers two plants, or one role is written to serve two sites, and the assumption is that one hire covers both.

Most people won't cross it. About 9.8 percent of U.S. workers commuted an hour or more each way in 2019, the Census Bureau reports, and the national average was 27.6 minutes. Two sites an hour apart are usually two separate labor markets. Building one search across both means running two searches badly instead of one well.

None of this requires a phone call to discover

That's the part worth sitting with. Every one of these is findable in a week of research, before a single candidate is contacted. Wage data, industry differentials, workforce size, competing employers, commute times: all of it is public, and almost none of it gets checked.

So the search launches, the posting goes up, six weeks pass, and the conclusion is that the market is tight.

Sometimes it is. More often the market was fine and the search was aimed wrong.

I'd rather know in week one.

If you're about to open a requisition, send me the job description. I'll send back a one-page read on pay, supply, and competition before you post.