Most owners bring a builder in once there is a set of drawings to price. By then the biggest cost decisions are already made. The site layout is fixed, the structural system is chosen and the floor-to-floor heights are drawn. The builder's job shrinks to counting what someone else decided. We think preconstruction should start earlier, while the project is still a program, a budget and a set of goals.
Cost is set early and spent late
A building's cost is largely locked in by a handful of early choices. How many stories. Steel or concrete. Where the central plant sits. How much of the envelope is glass. Each of those decisions is cheap to change in a meeting and expensive to change once it is on the drawings, and close to impossible once it is in the ground.
The money, though, gets spent much later, during procurement and construction. That gap is where projects get into trouble. A design that grew a little at every milestone shows up as a bid that comes in over budget, and the only tools left are value engineering and scope cuts. Neither one makes a better building. Both cost the design team time it has already spent.
When a builder is at the table from the first workshop, the estimate moves with the design instead of chasing it. We price options while they are still options, so the owner chooses with real numbers rather than finding out at the bid.
Target value design: set the budget, then design to it
Our preferred way to run early design is target value design. The idea is simple. Instead of designing a building and then estimating it, the team agrees on an allowable cost first, breaks it into targets for each system, and designs to those targets.
- Set the target. The owner's business case sets what the project can afford. We split that number into cost targets for structure, envelope, MEP, interiors and sitework.
- Estimate continuously. Our estimators update the cost model as the design moves, often weekly, so every cluster team can see where it stands against its target.
- Trade across systems. If the envelope team needs more money for performance glass, the savings come from somewhere the team agrees on, not from a cut made under pressure at bid day.
- Make decisions visible. A running decision log records what was chosen, why and what it cost, so nobody relitigates settled questions three months later.
The result is a design that lands on budget because it was built to the budget, not trimmed to it. It also tends to produce better buildings, because tradeoffs are made by the people who understand them, with time to think.
Bring the trades in before the details are drawn
An architect and an engineer can draw almost anything. The people who know what it costs to build, and how long it takes, are the trade contractors who install it. Their input is worth the most before the details are drawn.
On projects where we can, we bring key trade partners on early under design-assist agreements. Mechanical, electrical, plumbing, curtain wall and structural steel are the usual candidates, because those systems drive cost, schedule and coordination risk. A curtain wall contractor can tell the design team which unit sizes ship efficiently and which panel joints will leak. A mechanical contractor can suggest a duct route that frees up ceiling space and lets the floor-to-floor drop a few inches. Across a tall building, a few inches per floor adds up to real money in structure and skin.
Early trade input also feeds our in-house VDC team. A federated model built with the installers in the room catches clashes before they become RFIs. It lets us plan prefabrication of racks, risers and headwalls while there is still time to design for it. And it gives the schedule a basis in how the work will actually be sequenced, not a guess made from square footage.
The cheapest RFI is the one that never gets written because the installer saw the problem in a design meeting.
Preconstruction manager, MarkWester
From early estimate to a GMP the owner can trust
Early involvement pays off most at the guaranteed maximum price. A GMP built on a design the team has priced from the start is a different document from one assembled under deadline from a bid set.
By the time we propose a GMP, the owner has seen every major system priced more than once. Allowances are few and clearly defined. Contingency is sized to the risks that remain, not padded to cover a design nobody has read closely. Our qualifications and assumptions are short because most of the questions were answered in the design process. Long-lead items like switchgear, generators and air handlers have already been identified, and where it makes sense, released early so they arrive when the schedule needs them.
That GMP is also easier to manage through construction. When the owner, designer and builder have made decisions together, change orders drop, because fewer things were left unsaid. Buyout goes faster, because trade partners already know the project. And the schedule holds better, because the sequence was planned with the people doing the work.
None of this requires a particular contract. It works under CM at-risk, design-build and progressive design-build, and much of it works under a negotiated general contract. What it does require is an owner willing to bring a builder in before the drawings exist, and a builder willing to do real work in that window. That is the conversation we want to have with every client, as early as they will let us.
