Hard bid, CMAR and progressive design-build

Why this is the first decision, not a procurement detail

The delivery method is chosen long before there is a contractor, usually by people who will not be on site, and it constrains everything that follows: what the price means, what a change order is, who is responsible when the drawings turn out to be wrong, and — most importantly on a live plant — how early anyone with a shovel gets to look at the sequence.

That last point is the one this page exists for. On a retrofit, the buildability of the sequence is the project. A delivery method that puts the contractor in the room during design catches an unbuildable sequence while it is still a drawing. A delivery method that does not catches it as a change order.

All three methods below are legitimate and we work under all three. Which is right depends on how well-defined the scope is, how much risk the owner wants to hold, and what the procurement rules allow — which for a Colorado public entity is frequently the binding constraint.

Hard bid

The design is completed, the documents are issued, contractors bid a lump sum, and the lowest responsive bid wins. It is the default for public work and for good reasons: it is transparent, it is defensible to a board, and it produces the lowest price for a well-defined scope.

It works well when the scope really is well defined — a pipeline replacement in a known corridor, a new structure on a greenfield site, work where the unknowns are few and small.

It works badly when the scope is not, and a plant retrofit is the classic case. The bidder prices the documents as issued, including a sequence they had no part in writing. Everything discovered afterwards — the abandoned line shown as live, the invert a foot off, the tie-in that cannot be done in the window the drawings assume — is a change order, and change orders on a live plant are negotiated under time pressure with the plant already exposed. That is the worst possible moment to be negotiating anything.

If a project has to be hard bid and it is a retrofit, the mitigation is to spend money on investigation before the documents go out: potholing, CCTV, condition assessment, and accurate flow data. It is the cheapest money on the whole job.

Construction manager at risk (CMAR)

The owner selects the contractor on qualifications early, during design, and pays them to advise on constructability, cost and sequence while the design develops. At an agreed point the contractor gives a guaranteed maximum price and the relationship converts to a construction contract.

What this buys on a plant retrofit is the thing hard bid cannot: somebody who has actually done a tie-in reading the sequence while it is still cheap to change. A sequence that cannot be built in the outage window the design assumes is a conversation in month four instead of a claim in month fourteen.

It also lets the bypass be designed as part of the project rather than priced as an afterthought, which on a retrofit is frequently where the real money and the real risk are.

The trade-off is that the price is not fixed at the start, and the owner has to be comfortable with an open-book process and with selecting on qualifications rather than on number. For a district whose board is used to low-bid procurement, that is a governance conversation as much as a technical one, and it is worth having before the project rather than during it.

Progressive design-build

One contract covers both design and construction, procured on qualifications, with the price developed progressively as the design advances. The owner holds one party responsible for the whole thing.

The strongest argument for it is that it eliminates the gap the owner otherwise stands in. Under any split-responsibility method, when a facility does not perform, the owner is between the engineer who says it was built wrong and the contractor who says it was designed wrong. Under design-build there is no gap: one entity is responsible for the outcome.

It also allows the fastest schedule, because construction on early packages can start while later design is still developing.

The trade-offs are real. The owner gives up a degree of control over design decisions, gets less competitive price tension at the outset, and needs the internal capacity to administer a progressive process — which is more work for the owner, not less. It also requires clear performance criteria written up front, because the criteria become the acceptance test.

What we will tell you before you choose

We do not have an interest in pushing one of these. What we have an interest in is not being handed an unbuildable sequence under a method that gives us no way to say so until it is a claim.

So: if you send us a set of hard-bid documents for a retrofit and the sequence in them will not work, we will say so during the bid period, in writing, as a question to the engineer. That sometimes makes us unpopular and it is a great deal cheaper for everybody than the alternative.

And if you are at the stage of choosing a method rather than choosing a contractor, telephone and describe the project. That conversation costs nothing and it is the one with the most leverage on the outcome of anything that follows.