Hiring a construction manager for a GMP project is a commercial hire, not a delivery hire. The person you bring in inherits a fixed number, an allowance schedule and a contingency they did not set. Screen for who has carried that exposure before, on your delivery method and your sector, not for who has built similar square footage.
On a GMP project the guaranteed maximum price gets set by people who will not be there to defend it.
That is the structural fact behind most hires of a construction manager for a GMP project, and it is rarely how the job is described when the search starts. The requisition says the firm needs someone to run a $60M negotiated job. What it actually needs is someone who can hold a number assembled during preconstruction, priced against an unfinished design, and agreed with an owner who now treats it as a ceiling. The building is the easy part.
Under a lump-sum contract, a low estimate is a bid problem. Under a GMP, it becomes an operating problem that lands on one seat, month after month, until the job closes out. That is the seat you are filling.
What does a construction manager actually own on a GMP project?
On a GMP the construction manager owns the commercial position. That means the contingency draw, the buyout gap between estimate and awarded subcontract, the change-order argument over original scope, and the monthly cost report telling the owner whether the guarantee still holds.
Delivery is the part everyone screens for. It is also the part most candidates at this level can do. The differentiator sits in the contract. The BLS occupational profile describes a role spanning budget, schedule and contract administration, not field supervision alone.
Three exposures decide whether a GMP holds. None of them appear on a résumé:
- The buyout gap. The difference between what preconstruction priced a trade at and what it actually awards for. A construction manager who tracks this weekly finds the problem in April. One who reviews it at the monthly report finds it in July, when the trade is already mobilized.
- Contingency discipline. Every GMP carries contingency. The question is who may spend it, against what, and whether a drawdown gets recorded as a cost event or quietly absorbed. Firms lose GMPs to the second habit far more often than to one catastrophic scope miss.
- The change-order boundary. What counts as owner-directed change, against what counts as means-and-methods the contractor already priced. This is a contract-reading skill. It is also the one most often assumed rather than tested.
Ask a candidate which of those three cost them the most on their last GMP job. The answer separates people who carried the number from people who worked near it.
Why does hiring a construction manager for a GMP project go wrong?
Most fail on a mismatch the brief created. The role is written as delivery leadership and scoped as commercial accountability, so the search screens for project volume and sector experience while the actual job is holding a contract position. The candidate is competent and the hire is still wrong.
That outcome is worth sitting with, because it is the one nobody sees coming. The person interviews well. They have run bigger jobs than yours. Six months in, the reporting is late and the contingency is thinner than anyone expected, and no single decision explains it. There was nothing to catch, because the brief asked for the wrong thing and got it.
There is a second failure mode, and it is structural rather than individual.
The person accountable for the money is often on site twice: once when the ground is broken, once when it is handed over. Everything in between reaches them as a document. That is not laziness. It is how the roles are drawn, because the commercial function sits where the contracts, valuations and reporting live, and the execution function sits where the work is. Both are doing what the structure asks.
The result is two teams running one project on different information. Execution makes a call that is entirely sensible on site and carries a commercial consequence nobody flags for three weeks. Commercial holds a position that is correct on paper and lands on a sequence that stopped being possible in April.
Ask a site team what is still outstanding that has to go into the forecast, and the honest answer is often that they do not know yet. Not carelessness: the thing that will cost the money has not announced itself. So the number gets built from what was written down, and the part that would have changed it stays on site.
A GMP construction manager is the one hire that can close that gap, because the role sits on both sides of it. Which is why hiring for delivery skill alone leaves the gap exactly where it was. The same gap reappears after substantial completion, which is why a closeout leader is a separate question.
What should we screen for, and what can we ignore?
Screen for carried exposure on a comparable contract: the delivery method, the sector, the project stage they joined at, and whether they owned the contingency or reported on it. Ignore raw project value, total square footage, and the number of projects listed. Scale is the weakest predictor in the set.
| Screening signal | Why it predicts | What it replaces |
|---|---|---|
| Owned contingency on a GMP, not just reported it | Tests commercial authority, not proximity to it | ”Managed projects up to $80M” |
| Joined before the GMP was fixed on at least one job | They have set a number, not only defended one | Years of experience |
| Can describe a buyout gap they closed and how | Shows weekly cost control, not monthly reporting | ”Strong cost management skills” |
| Same delivery method and sector as your work | Negotiated healthcare is not hard-bid industrial | Total project count |
| Held the change-order boundary against an owner | Contract literacy under pressure | ”Excellent communication” |
The second row is the one most firms skip. A construction manager who has only ever inherited a GMP has never been on the hook for the assumptions inside it, and the first sixty days of their tenure go on reconstructing what was priced and why. If the submission is still ahead of you, bringing the hire in before the number is fixed converts that lost quarter into contribution. Selah runs this as a preconstruction manager search.
None of this is exotic. It is simply a different screen from the one a generalist recruiter runs, which is title, salary band and project value, and it is why résumés that clear an initial filter do not survive a technical conversation. On a GMP with heavy mechanical scope, the MEP coordination manager carries a large share of the same exposure.
Who is this hire not for?
A GMP construction manager is the wrong hire for a firm running hard-bid work, where exposure is set at bid day and the delivery seat genuinely is a delivery seat. It is also wrong for a first negotiated project where nobody internally has held a GMP before.
The second case deserves a sentence more, because it looks like the right hire. Bring in a construction manager as the only person in the building who understands the contract, and there is nobody to check their judgment against.
Negative qualification matters here more than it looks.
If your firm is moving from hard bid into negotiated work for the first time, the hire that de-risks that transition is usually a preconstruction leader who can build the estimating discipline, not a construction manager who can run the job once the number exists. The order matters, and getting it backward is expensive in a way that shows up two projects later. We have written separately on what construction management recruiters should ask before a preconstruction search, which covers that sequencing.
Equally, if the seat is currently occupied and the firm has decided to replace the incumbent, this stops being a hiring problem and becomes a confidentiality problem. Advertising the role tells the incumbent’s own team, the subs on the project, and frequently the owner. Selah Talent Partners runs those as confidential executive search for mandates that cannot be publicly posted, which is a different process from a posted search rather than a discreet version of the same one.
What does the market for this hire actually look like?
Tight, and tightest exactly where the GMP exposure sits. In the 2025 workforce survey from the AGC, 92 percent of firms reported difficulty filling open positions, and 45 percent said labor shortages were delaying projects.
The salaried breakdown is the part worth reading twice. Across 1,342 responding firms, the AGC analysis records superintendents as the hardest salaried role to fill at 81 percent of firms seeking them, estimating personnel at 77 percent, and project managers and supervisors at 76 percent. For 10 of the 15 salaried positions surveyed, at least half of firms with openings reported difficulty.
Those three roles are the commercial spine of a GMP job. The scarcity is not spread evenly across construction hiring: it is concentrated in the seats that carry the number. That is also why the posted-role route underperforms at this level, a pattern we cover in why good construction candidates are not applying.
Two further pieces of context are worth holding. Census Bureau construction spending data tracks where the work is going by sector, which tells you which candidate pools are being drained locally. The ENR top 400 list tells you who is competing for the same people in your metro. Construction compensation and availability are strongly regional. A Dallas negotiated-healthcare pool and an Atlanta heavy-civil pool are different markets at different prices, and a national salary band will mislead you in both.
Selah works with contractors and construction consultancies across the United States, on preconstruction and estimating, project and construction management, and cost and commercial management.
The hire is a contract decision
The temptation on a GMP search is to hire the person who has built the most of what you are about to build. It is the safe-looking answer and it screens for the wrong thing, because the risk on a guaranteed maximum price is not whether the building goes up. It is whether the number survives the way it goes up.
Square footage is the easiest thing on a résumé to verify and the least useful thing to know. Carried exposure is the hardest to verify and the only thing that predicts this seat. Most briefs screen for the first because it is countable.
So the question worth asking is not who can run this project. It is who can hold a number they did not set, in front of an owner who treats it as a ceiling, when the sequence changes in month four. Ask it before you write the job description, not after the second hire has not worked.
If you are filling this seat and want to compare notes on how to scope the brief, get in touch.
Sources and further reading
- AGC 2025 workforce survey news release
- AGC analysis salaried role difficulty detail
- AGC results national survey response data
- BLS occupational data for construction managers
- BLS OEWS construction managers wage data
- BLS estimators cost estimator occupational data
- Census Bureau construction spending data
- JOLTS job openings and labor turnover
- ENR top 400 contractor rankings
- CFMA construction financial benchmarking
- AGC outlook hiring and business outlook