An in-house construction recruiting function is cheaper than search fees only above roughly eight to ten leadership hires a year. Below that line the seat costs more than it displaces, and it cannot run the one search you most need run quietly: replacing a leader who is still sitting in the chair.
The case for your own recruiting function writes itself on a whiteboard. Four leadership hires last year, twenty percent each, call it US$120,000 in fees. Hire a recruiter for US$95,000 and keep the change. Every contractor president who has paid three invoices in eighteen months has drawn that box. The arithmetic is clean, and it is wrong in a way that only shows up in month nine, when the recruiter has filled two of the four seats and the third has been open since March.
What the whiteboard leaves out is not the salary. It is what the seat can and cannot do.
What does a construction recruiting function actually cost?
A fully loaded in-house construction recruiting function runs roughly US$140,000 to US$180,000 a year for one person: salary, payroll burden, a sourcing tool subscription, a job board contract, and the management time to run them. That fixed cost recurs whether or not you have four searches open.
Call the midpoint 160 thousand dollars. Against it sit four contingency fees on a US$150,000 base, which come to 120 thousand dollars, and the recruiter salary alone at 95 thousand dollars. Break those apart, because the salary is the part everyone estimates correctly and the rest is the part nobody books.
The US Bureau of Labor Statistics reports median pay for corporate recruiters, the occupational category that carries them, and construction firms competing against tech and healthcare for the same skill sit above that median, not below it. Add payroll taxes and benefits at the standard loading. Add a sourcing license, because a recruiter without a candidate database is a person with a phone and a hope. Add a job board contract you will discover you need in month two.
Then add the part that never reaches a spreadsheet: your VP of Operations spends four hours a week managing this person. That is a hundred and eighty hours a year of your most expensive operational judgment, redirected into pipeline review meetings.
The comparison that matters is not salary against fees. It is a fixed annual cost against a variable one, and fixed costs are unforgiving in a business whose hiring volume tracks a backlog that moves.
When does an in-house recruiting function beat a search firm?
Above roughly eight to ten leadership hires a year, sustained across at least two years. At that volume the fixed seat costs less per hire than fees, the recruiter has continuous live work, and the firm is large enough to have the second thing an internal function needs: a talent brand candidates already recognize.
That threshold moves with your average salary, so run it on your own numbers rather than ours. At a US$150,000 average base and a 20 percent fee (charged on first-year cash compensation, a base worth reading carefully), each search displaces US$30,000. A US$160,000 fully loaded seat needs to displace between five and six searches simply to break even, and break-even is not a reason to hire anyone.
Volume alone does not settle it, though. Three conditions have to hold together.
- Sustained volume, not a spike. A contractor that doubles backlog and hires nine leaders in one year, then hires two the next, has bought a fixed cost against a variable need. The backlog moved; the seat did not.
- Repeatable role types. An internal recruiter who fills project managers repeatedly builds a genuine bench. One asked to fill a project manager, then a controller, then a safety director, then a preconstruction lead is a generalist with four shallow networks.
- A brand candidates answer. A regional contractor everyone in the metro knows can source inbound. One that shares a name with two other firms in the state cannot, and its recruiter starts every call explaining who you are.
Miss any one of those and the seat underperforms the fee it replaced. Miss the third and something more specific goes wrong, which is where most of these decisions actually turn.
Why can an internal recruiter not run a confidential search?
Because the caller is the company. An internal recruiter phoning a competitor’s chief estimator identifies your firm on the first sentence, so the approach becomes market intelligence the moment it is declined. When the seat you are quietly replacing is still occupied, that exposure is the whole risk.
Consider the situation concretely, because the abstraction hides what is at stake. Your preconstruction lead is not working out. You have a negotiated healthcare project with a guaranteed maximum price submission in fourteen weeks, and the same person is leading the estimate. You cannot post the role. You cannot tell the estimating team. You cannot tell the owner, whose confidence in your preconstruction group is currently the reason you are in the negotiation at all.
Now send your internal recruiter to find the replacement. Every call they make carries your company name into a market where three of the people they call know your preconstruction lead personally, and one plays golf with your owner’s project executive. The search is a rumor inside two weeks. This is confidential executive search for mandates that cannot be publicly posted, and the defining constraint is that the searching party cannot be identified as you.
An outside firm carries no return address. That is not a service feature. It is a structural property of being a third party, and no amount of discretion training gives an internal seat the same property.
| Search type | In-house recruiter | Search firm |
|---|---|---|
| Posted project manager role, high volume | Strong: brand, speed, cost per hire | Weak: fee against a role that fills itself |
| Repeat superintendent hiring in one metro | Strong: network compounds locally | Moderate: better on the hard ones only |
| Replacing a seated chief estimator | Not viable: caller is identifiable | Strong: no return address on the approach |
| First hire in a new region | Weak: no local network to start from | Strong: network is the product |
Read the table as a portfolio rather than a verdict. Most contractors hiring at any scale need both, and the interesting question is not which one wins but which searches belong on which side of the line.
How should a contractor split searches between the two?
Put volume, posted, local and junior-to-mid roles in-house, where brand and repetition compound. Send confidential, senior, cross-market and first-in-region searches outside, where anonymity and an existing network are the product. The split is by search property, not by seniority alone.
The firms that get this right stop asking “should we hire a recruiter” and start asking “what fraction of our hiring is postable”. A contractor with fourteen annual hires where eleven are postable project engineers and field staff has an obvious in-house case, plus three searches a year that will never belong there. One with six hires where four involve a seated incumbent or a market they have never worked in has no in-house case at all, regardless of the whiteboard.
There is a quieter cost to getting the split wrong in the ambitious direction. An internal recruiter given a confidential mandate will try, because they were asked. They will make the calls. The search will leak, and the failure will read as a recruiting failure rather than a structural one, which means you will replace the recruiter instead of the model. The 2025 workforce survey found that 91.7 percent of the 1,041 contractors answering the salaried-hiring question reported difficulty filling positions, which means the market is unforgiving of a second attempt at the same seat.
Firm size does not change this, and neither does geography, though the reason differs by market. A Houston contractor competes for estimators against a deep Texas bench that also knows everyone in it. A Nashville firm has a thinner bench where a single indiscreet call reaches most of it. Either way, the searches that go outside are the ones where being identifiable is the disqualifier.
If you are weighing a first recruiting hire against a search partnership, the honest first step is to count your last two years of leadership hires and sort them into postable and not. If you want that conversation with someone who has no interest in the answer being fees, our approach to construction search starts from the same count.
Sources and further reading
- US Bureau of Labor Statistics, Occupational Employment and Wage Statistics: Human Resources Specialists 13-1071: BLS OEWS
- US Bureau of Labor Statistics, Occupational Employment and Wage Statistics: Construction Managers 11-9021: BLS OEWS
- AGC of America and NCCER, 2025 Workforce Survey, national results: AGC survey PDF
- AGC of America, construction data and industry surveys: AGC data