A conceptual estimator prices a project from a program, a site and a conversation, before drawings exist. The skill is defining scope and stating assumptions, with a number and a range attached. Hire for judgment about what the owner has not said yet. Takeoff speed, the thing most estimating interviews test, matters least in this seat.
Pricing was never the slow part of estimating. Scope was. Give ten estimators a complete drawing set and their numbers land within a few percent of each other, and software now does the counting for all ten. Give the same ten a two-page narrative for a 150,000 square foot building on a site nobody has surveyed, and the spread is enormous, because each has imagined a different building. The conceptual estimator is the one whose imagined building turns out to be the one that gets built.
What does a conceptual estimator do that a quantity estimator does not?
A quantity estimator measures what is drawn and prices it. A conceptual estimator decides what will have to be drawn, prices that, and writes down the assumptions. The first works from documents. The second works from precedent, questions and an understanding of how buildings go together.
The output looks different too. A hard-bid estimate is a number. A conceptual estimate is a number, a range, a basis-of-estimate narrative and a list of the decisions that would move it. The owner or developer uses it to decide whether there is a project at all, and the contractor uses it to decide whether to chase one. That is why an estimating exercise for this seat should score the assumptions a candidate writes down, not the speed of the takeoff.
Our note on the commercial estimator covers the measured side of the discipline. The conceptual estimator sits earlier, usually reporting to the chief estimator or the preconstruction leader, and is the person an owner meets at the first pricing conversation on negotiated, CM-at-risk and design-build work. Owners judge the whole firm by that meeting. They are also, fairly, going to ask how good the number is. At a multifamily developer-builder the early number is read against the pro forma, which is why the multifamily preconstruction seat there is judged on cost per unit and yield on cost. Our note on moving from hard-bid to negotiated estimating describes the move into this kind of work from the estimator’s side.
How accurate should a conceptual estimator’s number be?
Accurate enough to make the decision in front of the owner, with an honest range. Estimating standards classify estimates by how much of the project is defined, and the earliest class carries a wide expected range. A conceptual estimator states the range and narrows it as scope firms up.
The best-known framework is AACE International’s estimate classification. In its process-industry version, a Class 5 estimate, prepared when project definition is at 0 to 2 percent, has an expected accuracy range as wide as 50 percent under to 100 percent over. A Class 3 budget estimate tightens to a high side of about 30 percent, and a Class 1 estimate, built on nearly complete documents, to about 15 percent. The Department of Energy reproduces the classification in its cost estimating guide, and the Government Accountability Office’s cost guide sets out the practices that make any estimate credible: a documented basis, stated assumptions and a risk range.
Commercial buildings are better understood than first-of-a-kind process plants, so a good conceptual estimate for an office or a school should sit well inside those outer bounds. The principle holds anyway. An estimator who presents a single number at concept stage is either inexperienced or telling the client what the client would like to hear. The strong ones say “$62 million, and the three things that could make it $70 million are the foundations, the electrical service and the facade you have not chosen.”
That sentence contains the whole job. It also explains why the newest estimating tools have not replaced the person who can say it.
Why has estimating software not replaced the conceptual estimator?
Because the tools speed up counting and pricing what is already defined, and a concept has almost nothing defined. Automated takeoff needs drawings. Cost databases need a scope to look up. The conceptual estimator supplies the scope, which is the input everything else waits for.
Automation has made the measured estimate faster and cheaper, which moves the value upstream. When the quantity work took three weeks, a contractor could hide a weak conceptual capability behind it. When it takes three days, the difference between firms is who understood the project before it was drawn. On design-build work a design manager then holds the designers to that early number, so the design that gets drawn is the design that was priced.
Allowances show the skill, or its absence. Every allowance in an early estimate is a decision somebody has not made yet: a lobby finish, a generator size, a soil condition. A weak estimator scatters allowances to cover uncertainty and presents the total as prudent. A strong one treats each as an open question with an owner, a deadline and a likely answer, and works the list down before the guaranteed maximum price is set. On a CM-at-risk job the architect counts them at design development, and a list that has grown is a warning. The cost of leaving them open is described in our note on the cost escalation lead, where undefined scope meets a moving market.
So the interview has to test something a takeoff exercise cannot reach.
How do you interview a conceptual estimator?
Give the candidate a short written program and ninety minutes, with no drawings. Ask for a number, a range, the five assumptions that matter most and the questions they would put to the owner. Then change one fact and watch how the estimate moves.
A workable exercise:
- The brief. One page: building type, gross area, number of floors, a site address, a target opening date and two sentences on quality.
- The deliverable. Total cost and cost per square foot, a low and high case, and a short basis of estimate.
- The follow-up. “The owner now wants structured parking in place of surface parking.” A strong candidate reprices in minutes and names the knock-on effects on schedule, foundations and site work.
- The conversation. Ask where the number came from. Listen for specific past projects, adjusted for time, location and scope, and for what they would verify first.
| Signal | Strong candidate | Weak candidate |
|---|---|---|
| First move | Asks about the site, use and schedule | Asks for drawings |
| The number | A range with a stated basis | One figure to four significant digits |
| Assumptions | Ranked by cost impact | A generic exclusions list |
| When challenged | Adjusts the specific driver | Defends the total or abandons it |
Check the estimator’s history against outcomes where possible. The useful reference question is how the candidate’s concept numbers compared with final GMPs on the same projects, and what explained the gaps. The BLS groups all of these people under one occupation, cost estimators, which is part of why title and pay data say little about who can do conceptual work.
Sector matters more than metro, though the two travel together. Semiconductor and data center work around Austin and Dallas, and life sciences work in Raleigh, are priced from narratives long before design, on programs where the mechanical and electrical systems are most of the cost. An estimator with deep office and multifamily history will misjudge them, and the reverse is also true. On those programs the systems budget is the work of an MEP estimator, who prices the mechanical and electrical scope from the narrative before any subcontractor will commit. Industrial and process work is priced differently again, in craft hours rather than square feet, by an industrial estimator.
Who this hire is not for
- A hard-bid contractor. If every job arrives as a complete set of documents with a bid date, conceptual skill has nowhere to work. Invest in measured estimating and bid-day discipline.
- A firm that wants a salesperson with a spreadsheet. A conceptual estimator who shades numbers to win the pursuit will cost the company at GMP. The seat needs independence from business development.
- A contractor with one negotiated project a year. Let the chief estimator carry it and bring in a cost consultant for a second opinion.
The number that starts the relationship
An owner’s first serious question to a contractor is some version of “what will this cost?”, asked long before there is anything to measure. The answer sets the budget, the financing and the owner’s opinion of the firm, and every later estimate is compared to it.
Estimators who can give that answer well are known to the owners and architects in their market, and most are mid-pursuit at any given time. Advertise for one and you have told competitors which work you are short-handed on. The alternative is to approach them directly: confidential executive search for mandates that cannot be publicly posted, as our note on confidential search describes. Selah Talent Partners works with contractors across the United States that are moving from bid work toward negotiated work. Start here if the first number your firm gives an owner is still a guess.
Sources and further reading
- Cost estimating guide, US Department of Energy, DOE G 413.3-21A
- Cost guide, Government Accountability Office, Cost Estimating and Assessment Guide (GAO-20-195G)
- Cost estimators, Bureau of Labor Statistics, Occupational Employment and Wage Statistics (13-1051)