Assess a construction estimating leader on decision quality rather than project outcomes. Ask about a bid they lost and one they won that should have been lost, choose your own references rather than accepting the list, and test scope judgment rather than pricing speed. Outcomes are noisy. Decisions are not.
Every contractor has hired a construction estimating leader who interviewed beautifully. Confident, fluent, twenty years of relevant projects, references that came back warm. Nine months later the bids are late and the forecasts are optimistic and nobody can say precisely when it went wrong.
The interview was not the problem. What the interview measured was.
What should you actually assess in a construction estimating leader?
Decision quality, not project results. A construction estimating leader is judged in practice by outcomes they only partly controlled, so assessment has to reach the calls they made with the information available at the time. Ask what they assumed, what they were wrong about, and how quickly they found out.
Outcomes are a terrible instrument at this level, and construction is particularly bad for it.
A bid can be lost because a competitor bought the work. A job can go badly because an owner changed their mind about the program in month four. A margin can hold because escalation went the right way. None of that tells you whether the person in front of you exercises good judgment, and every one of those events shows up on a résumé as a result they either did or did not deliver. What you want is the reasoning, and reasoning is only visible in specifics.
- Scope judgment. What did they carry as risk, what did they exclude, and what did that decision cost or save.
- Assumption discipline. Which assumptions were written down and which were held in someone’s head.
- Speed of discovery. How long between a bad assumption and knowing it was bad.
- Escalation behavior. What they raised early rather than absorbing quietly.
- Team calibration. Who they trusted on numbers, and how they decided that.
| What most interviews measure | What predicts performance |
|---|---|
| Project size and sector history | Scope and risk judgment |
| Win rate | Quality of the bids they walked away from |
| Software and process familiarity | How fast they detect a wrong assumption |
| Confidence in the room | Willingness to say what they got wrong |
| Reference warmth | What the person who reported to them says |
The left column is not useless. It is simply available on the résumé, which means it is not what the interview is for.
How do you interview a construction estimating leader properly?
Ask about a bid they lost and a bid they won that should have been lost. A construction estimating leader who describes both, with numbers and specific reasoning, is describing judgment. One who can only describe wins is describing a curated history, and a rehearsed answer collapses under follow-up.
Two questions do most of the work, and both are hard to prepare for.
The first question is the lost bid. A good answer includes the number, the gap to the winner, what they would price differently, and what they concluded about the competitor rather than about themselves. A weak answer blames the market. The second question is the won bid that should have been lost, and it is the more revealing of the two, because it asks a candidate to admit that a success on their record was partly luck. People who understand estimating say yes immediately and can name the specific piece of luck. People who do not, and there are many at senior level, will not concede the premise.
Rehearsal difficulty is the reason both work. Outcome questions have prepared answers because every candidate has been asked them. Decision questions can be followed for three or four turns before the candidate has to reason live, and reasoning live is exactly the skill you are buying. The same principle applies to reference checks on construction executives, which most firms run as a formality.
Do reference checks work for construction leadership hires?
Only if you choose the references. A construction estimating leader supplies a list of people who will speak well of them, which is not dishonest, it is what a list is. Ask instead for the project manager who reported to them on their hardest job, and for the owner on the project that went worst.
That request is itself a test, and the response tells you something before the call happens.
Candidates who are confident about how they operated will make the introduction, sometimes with a caveat that the job was difficult, which is fair and useful context. Candidates who deflect, or who explain why that person would not be a fair reference, have told you something the curated list never would. Neither response is disqualifying on its own. Both are data.
When you get the call, ask about behavior under pressure rather than about performance in general. What happened when the forecast moved. Whether they were told about problems early or late. Whether the reference would work for them again, and the pause before the answer matters as much as the answer. The counteroffer dynamic sits nearby: a leader whose current employer will fight hard to keep them is telling you something the reference call may not.
The market context is worth knowing before you calibrate expectations. BLS cost estimators and construction managers set the compensation bands, BLS JOLTS and construction employment data show how rarely these people move, and Census construction spending with federal transportation and FHWA programs explains why demand in Texas and the Southeast keeps outrunning supply. AGC and ABC survey the same shortage, CFMA publishes structure benchmarks, OSHA records travel with a leader’s project history, and NCCER covers the craft pipeline underneath.
Assessment at this level almost always happens quietly, because the strongest candidates are employed and their employer does not know they are talking to you. That is the ordinary condition of confidential executive search for mandates that cannot be publicly posted, and it constrains how many people can be involved in the process.
Who this does not apply to. A firm hiring its first estimator does not need this depth, it needs someone competent and available. A firm with a strong internal candidate should assess them against the same questions rather than assuming familiarity is knowledge. And a firm that has already decided who it wants is running a confirmation exercise, which is fine, as long as everybody knows that is what it is.
The candidate who can name what they got wrong is usually the one who will notice it next time. That is the person we go looking for when a client briefs us.