- The delivery method is a risk-allocation decision first and a management-style decision second. It determines who holds design liability, who owns cost overruns, and who answers when a defect surfaces.
- Design-Bid-Build gives the owner maximum design control but leaves a contractual gap between designer and builder — the single most common fault line I encounter in defect litigation.
- Design-Build unifies responsibility but shifts design-quality control to the builder; the owner's protection lives or dies in the bridging documents.
- CM at Risk trades some bid-day price competition for early builder input and a Guaranteed Maximum Price — and moves disputes into GMP scope interpretation.
- IPD shares risk and reward across a multi-party agreement; it can work well, but the waiver and insurance provisions deserve careful legal review before signing.
When I am retained to investigate a building failure, one of the first documents I ask for is not a drawing, a submittal, or an inspection report. It is the contract. Before I can say anything useful about why a wall assembly leaked or a slab cracked, I need to know how the project was organized — because the delivery method chosen years earlier quietly decided who was responsible for coordinating the details that failed.
Page 1
After more than twenty-five years in building science and construction forensics, I have investigated defects on projects delivered every way the industry knows how. The defect itself rarely cares about the contract structure. But the dispute that follows the defect is shaped by it almost entirely. This article walks through the four principal delivery methods — Design-Bid-Build, Design-Build, Construction Manager at Risk, and Integrated Project Delivery — from the perspective of someone who usually meets these contracts after something has gone wrong.
What Is a Construction Delivery Method — and Why Does It Matter Before a
Shovel Hits the Ground? A delivery method is the contractual and organizational framework that defines how design and construction services are procured, sequenced, and held accountable. It answers four questions before any technical work begins: who designs, who builds, who holds which contracts, and who carries the risk when reality departs from the plan. Those answers matter to me as a forensic investigator because responsibility follows contract lines, not common sense. On a project with a divided structure, a coordination failure between the mechanical drawings and the structural steel is not automatically anyone's fault — it lands wherever the contracts put it. On a unified structure, the same failure has an obvious owner. Neither arrangement prevents the failure; each one determines how expensive and adversarial the aftermath becomes. The industry's standard contract families — the AIA documents, ConsensusDocs, DBIA forms, and EJCDC documents — each publish coordinated sets built around these delivery methods. Choosing a method effectively chooses a contract ecosystem, an insurance structure, and a dispute posture all at once.
How Does Design-Bid-Build Work, and Where Does It Break Down?
Design-Bid-Build (DBB) is the traditional sequence: the owner hires a design team, the design team produces complete construction documents, the project is competitively bid, and a general contractor builds what the drawings show. The owner holds two separate contracts — commonly an AIA B101 with the architect and an AIA A101/A201 pair (or EJCDC C-700 on civil work) with the contractor. Its strengths are real: maximum owner control over design, a complete set of documents to price, and transparent low-bid competition that public owners often require. Its structural weakness is just as real: there is no contract between the people who designed the building and the people who built it. When a defect traces to the boundary between design intent and field execution, the owner sits in the middle of that gap. This is where the Spearin doctrine — from the 1918 U.S. Supreme Court decision in United States v. Spearin — enters nearly every DBB dispute I touch. In broad terms, an owner who furnishes plans and specifications impliedly warrants their adequacy to the contractor. A builder who faithfully constructed defective drawings has a strong defense; a builder who deviated from adequate drawings does not. Much of my forensic work on DBB
Page 2
projects is exactly that sorting exercise: distinguishing design deficiency from construction deviation, detail by detail.
Figure 1 — The four contract structures. Note where the contract lines run — and where they don't.
What Does Design-Build Change About Responsibility?
Design-Build (DB) collapses the divide. The owner signs one contract — typically a DBIA 530/535 pair or an AIA A141 — with a single design-build entity that carries both design and construction responsibility. When something fails, there is, in the industry's blunt phrase, one throat to choke. Fast-tracking becomes natural because the same organization controls both design production and construction sequencing. From a forensic standpoint, the trade is straightforward: the owner gains single-point responsibility and gives up independent design advocacy. The architect no longer works for the owner; the architect works for the builder. The owner's real protection shifts upstream into the bridging documents — the owner's project criteria, performance specifications, and basis-of-design narratives prepared before the design-builder is selected. The DB disputes I investigate concentrate in two places. First, performance-specification ambiguity: when the criteria say a system must “achieve equivalent performance,” someone eventually has to decide what equivalent means, and the parties rarely agree after a failure. Second, conflicts between the bridging documents and the final design — particularly when owner criteria were thin, and the design-builder's cost pressures filled
Page 3
the silence. Owners who retain an independent advisor to review design development under a DB contract are, in my experience, meaningfully better positioned when questions arise later.
How Does CM at Risk Balance Cost Certainty and Collaboration?
Construction Manager at Risk (CMAR) keeps the owner's two-contract structure — designer under one agreement, construction manager under another, commonly the AIA A133 or A134 or ConsensusDocs 500 — but changes the timing. The CM is hired during design, provides preconstruction services such as constructability review, cost modeling, and scheduling input, and then commits to a Guaranteed Maximum Price (GMP), usually before the drawings are fully complete. The appeal is a genuine middle path: the owner keeps an independent design team, gains a builder's practical input while the design can still respond to it, and receives a cost commitment earlier than a DBB bid would provide. Early trade packages — earthwork, foundations, long-lead equipment — can start before the full document set is finished. The forensic pattern follows the GMP. Because the price is set against incomplete documents, the GMP amendment typically includes qualifications, assumptions, and allowances describing what the CM understood the design to include. Years later, the dispute becomes an interpretation exercise: was the failed scope “reasonably inferable” from the documents at the time of the GMP, or was it a legitimate change? Contingency drawdown records and the drafting of those qualifications frequently decide who absorbs the cost of correction. When I review CMAR project files, the GMP amendment and its exhibits get as much of my attention as the drawings do.
Page 4
Figure 2 — Sequence and responsibility. Illustrative sequencing only; actual durations depend on the project.
What Makes Integrated Project Delivery Different?
Integrated Project Delivery (IPD) is the furthest departure from tradition: the owner, the design team, and the builder sign a single multi-party relational agreement — the AIA C191 or ConsensusDocs 300 are the common vehicles. Instead of allocating risk to individual parties, the agreement pools it. Profit is typically tied to shared project outcomes against a target cost, key participants often waive many claims against one another, and major decisions run through a joint management team. On complex projects with sophisticated participants — healthcare work is the frequent example — IPD can align incentives in a way the other methods struggle to match. The team that designs the building is financially invested in how it builds, and vice versa. My cautions come from the same features that make it attractive. The waiver-of-claims provisions that keep an IPD team collaborative can also narrow an owner's recovery path if a defect emerges after closeout, and insurance products for shared-liability structures are still less standardized than the coverage supporting the traditional methods. IPD also depends heavily on the experience of every signatory; a multi-party agreement with one inexperienced party is a multi-party problem. None of this makes IPD unsound — it makes the presignature legal and insurance review non-negotiable.
Where Do Disputes Concentrate in Each Method?
No delivery method eliminates risk; each one relocates it. When I map the defect disputes I have worked on against the delivery method behind them, the patterns are consistent enough to summarize — not as statistics, but as the recurring shapes an investigator learns to look for.
Page 5
Figure 3 — Recurring dispute patterns by delivery method, drawn from forensic practice.
The common thread is instructive: disputes concentrate wherever the method leaves a seam — between
designer and builder in DBB, inside the performance criteria in DB, around the GMP boundary in CMAR, and within the risk-pool accounting in IPD. Quality documentation practices, independent inspection at critical hold points, and disciplined contract administration do not change the method's seams, but they determine how well the record answers questions when those seams are tested.
How Should an Owner Choose a Delivery Method?
There is no universally superior method — only a better or worse match between the method's risk profile and the owner's priorities, sophistication, and capacity to manage what the method leaves on the owner's side of the table. I encourage clients to start from the constraint they cannot compromise, then verify the method's known weak points before signing.
Page 6
Figure 4 — A practical selection framework: lead with the non-negotiable constraint, then verify the method's known seams.
Side-by-Side Comparison
Attribute
Design-Build
CM at Risk
IPD
Contracts held by owner
Two (designer +
builder)
One (design-builder)
Two (designer + CM)
One multi-party
agreement
Design liability
Architect/engineer; owner impliedly warrants plans to builder
Design-builder
Architect/engineer, with CM constructability input
Shared per
agreement, often with claim waivers
Cost commitment
Lump sum or unit
price at bid
Lump sum or GMP, often early
GMP, typically set
during design
Target cost with
shared risk/reward pool
Schedule structure
Fully sequential
Overlapped / fasttrack
Overlap via early
packages
Continuous
integrated team
Owner design control
Highest
Lowest without
strong bridging documents
High, with builder
input
Collaborative, consensus-driven
Common contract forms
AIA A101/A201, B101; EJCDC C-700
DBIA 530/535; AIA
A141
AIA A133/A134; ConsensusDocs 500
AIA C191; ConsensusDocs 300
Page 7
Attribute
Design-Build
CM at Risk
IPD
Typical dispute pattern
Design-adequacy and
change-order claims across the designer/builder divide
Performance-spec
ambiguity; bridgingdocument conflicts
GMP scope and
contingency disputes
Risk-pool accounting; waiver enforceability
COMPOSITE CASE STUDY — ILLUSTRATIVE ONLY
The following is a composite drawn from recurring fact patterns across multiple engagements. It does not describe any single project, client, or party.
A mid-rise mixed-use project was delivered CM at Risk, with the GMP set at roughly 75 percent design
completion. The exterior wall assembly later exhibited widespread water intrusion at window-to-wall interfaces. The CM's position: the final flashing details differed materially from what was “reasonably inferable” at GMP, making the corrective scope a change. The owner's position: weather-tight window integration was plainly within the GMP's scope, however the details evolved. The investigation turned less on building science than on paper. The GMP amendment's list of qualifications had never been reconciled against the final construction documents; the design team's detail revisions were issued without pricing confirmation; and no one had documented which contingency draws related to envelope scope. The physical cause of the leakage was resolvable in weeks. Allocating its cost took far longer — and nearly every hour of that fight traced back to how the delivery method's central document, the GMP amendment, had been administered. The lesson generalizes across all four methods: the delivery method's signature document — the bid set in DBB, the bridging criteria in DB, the GMP amendment in CMAR, the target-cost and waiver terms in IPD — deserves the same rigor as the drawings, because that is the document the dispute will be fought over.
Frequently Asked Questions
Which delivery method is fastest? Methods that overlap design and construction — Design-Build, CM at Risk with early packages, and IPD — generally compress overall duration relative to fully sequential Design-Bid-Build. Whether that compression is realized on a specific project depends on design maturity, permitting, and procurement realities, so treat speed claims as project-specific rather than inherent.
Which method is cheapest?
Page 8
None of them, reliably. DBB offers bid-day price competition on complete documents; DB and CMAR offer earlier cost commitment and fewer design-error change orders; IPD targets cost through shared incentives. Each mechanism controls a different category of cost risk. The expensive outcome in my files is usually not the method — it is the mismatch between the method and the owner's ability to administer it.
Can a public owner use anything other than Design-Bid-Build?
Increasingly, yes. Many states have authorized Design-Build and CMAR for public work through alternative procurement statutes, though enabling legislation, project-size thresholds, and selection procedures vary significantly by state and agency. Confirm the current statutory authority in your jurisdiction before assuming a method is available.
What are bridging documents, and do I really need them on Design-Build?
Bridging documents are the owner's criteria package — performance requirements, basis-of-design narratives, and sometimes partial design — that define what the design-builder must deliver. On DB projects they are the owner's primary quality lever. In my forensic experience, thin bridging documents are among the most consistent predictors of later scope and quality disputes on DB work.
Does the delivery method change how a defect investigation is performed?
The physical investigation — testing, openings, moisture mapping, code and standard comparison — is methodagnostic. What changes is the documentary side: which contracts, amendments, and records I request, and how causation maps onto responsibility. The method determines whose file answers the question.
Pre-Selection Checklist: Ten Questions to Answer Before Committing
☐ Which single constraint — design control, speed, cost certainty, or unified responsibility — is genuinely non-negotiable for this project? ☐ Does our organization have the staff capacity to administer this method, including the documentation burden it creates? ☐ If public, is this method authorized under the applicable procurement statutes, and what selection procedure does the statute require? ☐ Which standard contract family (AIA, ConsensusDocs, DBIA, EJCDC) will govern, and which specific forms and amendments? ☐ Under DBB: how will we manage the designer/builder gap — and who reviews design adequacy before bid? ☐ Under DB: are our bridging documents complete enough to protect design quality, and who provides independent owner-side design review? ☐ Under CMAR: at what design percentage will the GMP be set, and how will its qualifications and assumptions be reconciled against final documents?
Page 9
☐ Under IPD: has counsel reviewed the waiver-of-claims and insurance provisions, and does every signatory have relevant IPD experience? ☐ How will independent quality assurance and hold-point inspections be structured, regardless of the method chosen? ☐ If a defect emerges in year six, what does the recovery path look like under this contract structure — and are we comfortable with it?
Which delivery method has served your projects best — and which one taught you the hardest lesson? I
read every comment.
Page 10