Every building gets delivered one of three ways. The industry names them badly, explains them worse, and then asks owners to pick. Here is what they actually mean, in terms of the only two questions that matter to you: who is holding the risk, and when do you find out the real price?
Design-bid-build
You hire an architect. The architect designs the building and produces a complete set of construction documents. Those documents go out to bid. General contractors price them, and you select one, usually the lowest responsible bid. That contractor builds it.
When you learn the real price: after the design is finished. This is the defining feature and the defining problem.
Who holds the risk: you do. You hold the gap between what the architect drew and what the market will build it for. If the bids come in over budget, that is your problem to solve, and your options at that point are to cut scope, find money, or redesign. You also hold the gap between the drawings and the field: anything ambiguous, missing, or conflicting becomes a change order priced when you have no leverage.
When it is right: when your procurement rules require competitive sealed bidding on complete documents, when the scope is genuinely simple and well understood, or when the political need for a visible low-bid process outweighs the cost risk. For many public agencies this is not a preference, it is a statute.
Construction manager at risk
You hire an architect and, separately and earlier than in design-bid-build, a construction manager. The CM advises on cost and constructability while design is still moving, then at some agreed point converts to a guaranteed maximum price and builds the project.
When you learn the real price: partway through design, which is a genuine improvement.
Who holds the risk: shared, and the sharing is the complication. You are still managing two contracts with two sets of incentives. The CM's price is only as good as the design documentation it was based on, so the negotiation over what was and was not included in the GMP is where the friction concentrates. You are also, in practice, the referee between the designer and the builder, which is a role most owners are not equipped for.
When it is right: on large, complex, phased projects where early cost input is valuable and the owner has enough in-house capacity, or a good owner's representative, to manage two contracts actively.
Design-build
You hire one firm. That firm designs the building and builds it under a single contract.
When you learn the real price: before construction, and because the same organization owns both sides, cost is an input to design rather than a verdict on it.
Who holds the risk: the design-build firm. There is no gap between the drawings and the field to fall into, because one party owns both. That is the entire point.
When it is right: when you want single-point accountability, a price you can defend to a board, and a completion date you can plan around. It fits owners without construction staff especially well, because it removes the coordination burden rather than transferring it to you.
The honest caveat: design-build concentrates a great deal in one relationship, so selection matters more than in the other two. You are choosing a partner, not a low bid. Ask who leads the team and what standard of care they work under. In architect-led design-build, the architect leads, which means the person running your project carries a fiduciary duty to your interest. In builder-led design-build, they generally do not.
Side by side
- Contracts you hold: design-bid-build two, CM-at-risk two, design-build one
- Real price arrives: after design, during design, before construction
- Who absorbs the design-to-field gap: you, shared, the design-build firm
- Who referees a dispute: you, you, nobody needs to
- Selection basis: low bid, qualifications plus fee, qualifications and value
The question behind the question
Owners usually ask which method is cheapest. That is the wrong frame, because all three can produce the same building for roughly similar hard cost. What differs is who absorbs the variance when reality diverges from the plan, and reality always diverges.
So ask instead: if this goes 15 percent over, whose problem is that? Then pick accordingly. If your organization cannot absorb a 15 percent overrun without a crisis, and most school districts, nonprofits, and municipalities cannot, you should be choosing a method that puts that risk somewhere else.
How to decide without guessing
A Design Framework Report includes a delivery-method comparison against your actual project, your funding source, and your procurement constraints. It is a far better basis for the decision than a general article, including this one, because the right answer genuinely depends on your statute, your board, and your building.
Keep reading on why building projects go over budget and what a guaranteed maximum price actually guarantees.