Construction job postings fail on senior seats for a structural reason: they select for people who are looking, and the project managers and estimators you want are employed, busy, and not reading job boards. Rewriting the posting samples the same wrong population more attractively.
Your construction job postings have been live six weeks. Forty-one applications. Two worth a phone call, and one of those withdrew before the interview.
At this point most firms conclude the posting needs work. Better title, clearer description, add the salary, move it up the board. So the posting gets rewritten and reposted, and six weeks later there are thirty-eight applications and one worth a call, and everyone agrees the market is difficult.
The market is difficult. The posting is also the wrong instrument, and those are separate problems.
Why do construction job postings attract the wrong candidates?
Because a posting samples the population currently looking for work, and your best available project manager is not in it. They run two jobs for a competitor, they are reasonably happy, and they have not opened a job board in three years. The problem is distribution, not copy.
Think about who is actually reading a posted senior construction role on a Tuesday afternoon.
- People between roles. Some are excellent and were caught by a project ending or a firm contracting. Many are not, and sorting them takes real time.
- People who are unhappy. A genuinely useful group, but “unhappy where they are” correlates weakly with “good at the job”, and sometimes negatively.
- People applying broadly. The application is cheap for them, which is exactly why 41 of them arrived, 2 were relevant, and after the commute conversation just 1 was still interested.
Missing from that list is the group you actually want: competent, employed, not looking, and reachable only by someone who calls them. In a market where 91.7 percent of the 1,041 contractors answering the salaried-hiring question in the 2025 workforce survey reported difficulty filling salaried positions, that group is not idle. It is employed by your competitors, all of whom are also short.
So the funnel is not leaking. It was filled from the wrong source, and the numbers at the bottom are what that looks like.
When do construction job postings actually work?
When a real applicant market exists for the role. Project engineers, coordinators, assistant PMs, and any seat where people actively move and your firm is known are well served by a posting. It stops fitting where the qualified pool in your metro drops below a few dozen people.
That threshold is worth locating precisely for your own firm, because it is not a seniority line so much as a scarcity line.
| Role | Qualified pool in a metro | Posting works? |
|---|---|---|
| Project engineer | Large, actively moving | Yes, reliably |
| Project manager, commercial | Moderate, some movement | Sometimes, with a strong brand |
| Senior estimator, GMP work | Small, rarely looking | Rarely |
| Chief estimator or preconstruction lead | Very small, almost never looking | No |
Two things move a role up this table besides seniority. Delivery method is one: a project manager who has run guaranteed maximum price work on negotiated healthcare projects is a much smaller pool than “project manager”. Geography is the other, and it cuts both ways. A posting in Dallas or Houston reaches a deeper market than the same posting in Nashville or Raleigh, but it also competes against far more postings.
The practical version: before you post, estimate how many people within an hour of the office could actually do the job. If the answer is under fifty, the posting is a formality and you should plan the real search alongside it.
What should you do instead of reposting?
Change the instrument rather than the copy. A search that reaches employed people requires someone to identify them, call them, and give them a reason to take a conversation they did not ask for. That is a different activity from advertising, and it does not become one by being described better.
The sequence that works is unglamorous and mostly mechanical.
First, map the population. Which firms in your metro run the kind of work your seat requires, and who holds the equivalent seat at each. For most senior construction roles this list is between twenty and eighty names, and it is knowable.
Second, decide what you are offering that their current employer is not. Not the salary, or not only. Scope, project type, reporting line, the chance to build a group rather than inherit one. A strong PM at a good firm needs a reason, and “we are hiring” is not one.
Third, make the approach in a way that does not cost them anything to take. That last part is where most direct approaches by the hiring firm break down, because a call from you is a call from a known competitor and it carries risk for them.
There is also a class of search where posting is not merely ineffective but actively harmful. When the seat is currently occupied, when a bid is pending and any signal of instability in preconstruction would cost you the negotiation, or when your competitors would read the posting as a map of your weaknesses, the role cannot be advertised at all. That is confidential executive search for mandates that cannot be publicly posted, and it starts from the assumption that no posting will ever exist.
Which brings the argument back to where it started. The forty-one applications were not a failure of the posting. They were a correct answer to a question you did not mean to ask, which was: who is available right now. The question you meant to ask was: who is good. Those are different queries and they need different instruments.
Three weeks into a posting with nothing shortlistable, you already have your answer. Build the proper search brief instead of the fourth revision of the advert, and run the search as a search.
Sources and further reading
- AGC of America and NCCER, 2025 Workforce Survey, national results: AGC survey PDF
- US Bureau of Labor Statistics, Job Openings and Labor Turnover Survey: BLS JOLTS
- US Bureau of Labor Statistics, Occupational Employment and Wage Statistics, Construction Managers 11-9021: BLS OEWS
- AGC of America, construction data and industry surveys: AGC data