You can usually name the person before the owner suspects there is a problem. The submittal log, the RFI turnaround and the tone of the last three OAC meetings all say it. What almost no owner’s rep has is a considered process for raising it, which is why the conversation tends to happen two months late and once the schedule has already moved.
The tell an owner’s rep notices is rarely dramatic. It is a two-week RFI turnaround that used to be four days, a superintendent answering questions that should go to the project manager, and a look passed across the table when the schedule update comes up. None of it belongs in a report. All of it means the same thing.
You are the first person on the project who knows. That is the position the job puts you in, and it is more uncomfortable than the contract makes it sound.
What does an owner’s rep actually see first?
Coordination behavior, ahead of any cost or schedule number. RFI turnaround stretching, submittals resubmitted for the same reason, decisions escalating to the field because the office is not making them, and OAC meetings where the same item reappears for a third cycle. Cost and schedule report the consequence a month or two later.
That lead time is the whole value of the seat, and it is also the whole problem. The evidence is behavioral, it is qualitative, and the owner’s rep is contractually a monitor rather than a manager of the contractor’s staffing.
So a person with the earliest information has the least direct authority to act on it. That is not an accident of any particular contract. It is how the role is drawn.
Why is this so hard to raise?
Because the observation sits between two roles. The contractor manages its own staff, the owner holds the contract, and the owner’s rep sits between them with a view that neither has and an authority that matches neither. Raising it early risks the relationship. Raising it late risks the project, and your standing.
There is a third risk that goes unnamed more often than it should: being right too early. A concern raised in week six, without a documented pattern, reads as a personality conflict. The same concern raised in week fourteen with three cycles of evidence reads as diligence. The material difference between those two conversations is not the underlying facts. It is what has been written down.
That points at the practical answer, which is unglamorous and mostly about documentation.
Keep the record on the ordinary project instruments rather than in a private file. Submittal logs, RFI response times, meeting minutes and schedule updates are already being produced, already shared, and already contractually meaningful. A pattern visible in documents the contractor also holds is much harder to characterize as one consultant’s opinion.
Three tests separate a bad fortnight from a pattern worth raising:
- Recurrence. The same category of failure across three cycles, not three different problems in one month. Different problems point at a project. The same problem repeating points at a person or a process.
- Response to being flagged. A team that corrects after the first flag has a workload problem. A team that repeats the failure after two flags has a capability or capacity problem, and those need different answers.
- Concentration. Whether the failures follow one person’s scope across the job. If the same coordination gap appears wherever a single individual is accountable, that is a much narrower finding than “the team is struggling.”
Once a pattern clears all three, the question changes from whether to raise it to how.
How should an owner’s rep raise it with the owner?
Lead with schedule and cost consequence, not with a personality. Describe what is not happening, what it has cost in days or dollars, what has already been flagged and when, and what the pattern will do to the next milestone if nothing changes. Recommend a response rather than a person.
The distinction between behavior and person is not a matter of politeness. It is what keeps the conversation inside your professional standard of care and out of the territory where you are directing the contractor’s employment decisions.
A concrete version reads roughly like this. Submittal responses on a negotiated CM-at-risk healthcare project in the Southeast have run past contractual review windows on six of the last nine packages. Two were flagged in OAC minutes in consecutive months. The pattern sits entirely within one scope area. On the current sequence, the next long-lead procurement date is exposed by about three weeks. That is a report an owner can act on without anyone being accused of anything.
What the owner does with it is genuinely their decision, and the options are not equivalent.
| What the owner can do | What it fixes | What it costs |
|---|---|---|
| Ask the contractor to add support around the seat | Capacity problems, quickly | Nothing contractually; leaves a capability problem in place |
| Ask the contractor to replace the individual on this project | The immediate delivery risk | Relationship friction, and a ramp-up on a live job |
| Do nothing until the next milestone slips | Nothing | The recovery is more expensive than the intervention would have been |
| Raise it as a structural staffing question with the contractor’s leadership | The underlying gap, if it recurs across jobs | Requires the contractor to be willing to hear it |
Most owners choose the first option, and most of the time that is correct. The interesting case is the fourth, because it is the one that reaches beyond the project.
What happens when the contractor already knows?
Frequently they do, and the constraint is not awareness but supply. A contractor that knows it has the wrong person in a seat on a live job cannot advertise the replacement: the posting reaches the project team, the subs and often the owner. The gap persists for reasons unrelated to denial.
This is where the conversation stops being a project-management question and becomes a hiring one. Selah Talent Partners runs confidential executive search for mandates that cannot be publicly posted, and a seated project manager on a live job is close to the defining case. The search runs by direct approach, the incumbent stays in the seat until the transition is planned, and the market learns nothing until the change is real.
Owner’s reps across the United States sit at the point where this problem is first visible, which is why the referral matters. The pattern holds whether the job is a Texas industrial build-out or a Nashville hospital expansion. A contractor told in a project meeting that a seat needs to change, and given no route to change it discreetly, will usually do nothing. A contractor given the same message and a name is in a different position by the end of the week.
Two things this is not. It is not a route for an owner’s rep to select the contractor’s staff, which is neither your role nor a place you want to be standing if the replacement also struggles. And it is not for the case where the real problem is the brief: a “project manager” seat that is actually a project executive’s scope will fail with any occupant, and hiring against the wrong brief simply repeats the failure with a new name. Interrogate the scope before anyone concludes it is the person. On mission-critical work that scope question is sharper still, as our note on the data center project executive sets out.
Where the pattern is real and the seat is genuinely wrong, the useful contribution from the owner’s rep is not the diagnosis. Everyone gets there eventually. It is the timing: two months earlier, when the schedule can still absorb it, and quietly enough that the contractor can act without a public admission. On a program with several delivery teams, the person who sees the failure first is the owner’s program manager, whose job is the owner’s decision queue.
If a delivery team on one of your projects has stopped performing and the contractor needs a discreet route to fix it, that is worth a conversation before the next milestone rather than after it.
Sources and further reading
- FHWA guidance on federal-aid highway construction program administration
- Census data monthly value of construction put in place
- BLS managers Occupational Employment and Wage Statistics, Construction Managers
- AGC survey 2025 AGC and NCCER workforce survey, national results
- ACEC on engineering and owner’s representative professional practice