Design-Build vs. CM-at-Risk vs. Design-Bid-Build (2026)
Design-Build vs. CM-at-Risk vs. Design-Bid-Build (2026)
Which Delivery Method Actually Saves Money? From a GC Who Builds Under All Three
The wrong delivery method costs commercial developers 8–15% of total project cost. Here's how to pick the right one in 2026.
Every commercial construction project — every cold storage warehouse, every cannabis cultivation facility, every QSR drive-thru, every medical office building — is delivered through one of three dominant contracting methods: design-build, construction manager at risk (CM-at-Risk), or design-bid-build. Each one has a different cost profile, a different schedule profile, a different risk allocation, and a different point at which the project's total price gets locked in.
Owners who choose the wrong method pay for it. We've seen design-bid-build projects deliver 14 months late and 11% over budget because the architect designed something the market wouldn't price. We've seen CM-at-Risk projects come in 9% high because the GMP got set before scope was clear. And we've seen design-build projects beat the original budget by 6% because preconstruction and construction were one continuous handoff.
This is a straight comparison of the three methods, written by Terrapin Construction Group — a nationwide commercial general contractor that delivers under all three. We have skin in the design-build game (it's our preferred model), but we'll tell you when CM-at-Risk or design-bid-build is actually the right answer.
TL;DR — Which Method Wins in 2026
- Design-Build wins on speed, single accountability, and integrated design-and-construction creativity. Typical for cold storage, IMP-heavy industrial, cannabis, data centers, and any project where MEP/process integration drives outcomes.
- CM-at-Risk wins when the owner wants competitive subcontractor bidding, GMP cost certainty after design is mostly complete, and a collaborative team relationship without committing to design-build upfront. Common for healthcare, higher ed, and complex tenant improvements.
- Design-Bid-Build wins when scope is simple, design is fully complete before pricing, and the owner wants the lowest possible bid with strict cost controls. Best for public sector work and straightforward shell-and-core builds.
For projects in the $1M–$30M range with refrigeration, IMP envelope, process MEP, or fast-track schedules, design-build wins on cost and schedule in 70%+ of cases we see.
How each method actually works
Design-Bid-Build (DBB)
The oldest delivery model. Sequential, linear, and adversarial by design.
How it works: Owner hires architect/engineer to produce 100% complete documents. Documents go out for competitive bid. Lowest qualified bid wins. GC builds exactly what's drawn. Any change triggers a change order.
Contracts: Two separate — architect/engineer and GC.
Risk allocation: Owner retains design risk. GC retains construction risk. No shared interest in problem-solving.
Construction Manager at Risk (CM-at-Risk)
A hybrid that brings the construction manager into the project earlier, while preserving competitive sub-bidding at the back end.
How it works: Owner hires architect plus CM during design. CM provides preconstruction services. At 60–90% design, the CM provides a Guaranteed Maximum Price (GMP). CM then competitively bids work to subs and manages construction "at risk" of the GMP.
Contracts: Two separate — architect/engineer and CM — but CM is involved during design.
Risk allocation: Shared. Owner retains design risk. CM accepts construction risk at the GMP.
Design-Build (DB)
A single-source delivery model where one entity is responsible for both design and construction.
How it works: Owner hires one design-build entity (GC-led with integrated architecture/engineering, like TCG with 3rd Act Architecture and 9BA MEP). Team works through preconstruction, design, and construction as one continuous process. Pricing is established progressively, with lump-sum or GMP locked in at 60–80% design.
Contracts: One contract.
Risk allocation: Most design and construction risk transfers to the design-build entity. Owner retains scope clarity risk and external risks (utilities, permitting, market conditions).
Read our deeper explainer on what a design-build contractor does.
Head-to-head: the four metrics that matter
Cost certainty
| Design-Bid-Build | CM-at-Risk | Design-Build | |
|---|---|---|---|
| When price is locked | After 100% design | At 60–90% design | At 60–80% design |
| Change order rate | 5–12% | 3–6% | 1–3% |
| Final cost vs. original budget | +6 to +14% | +2 to +6% | -2 to +4% |
| Owner cost control | Weakest | Moderate | Strongest |
Schedule
| Design-Bid-Build | CM-at-Risk | Design-Build | |
|---|---|---|---|
| Design phase | 9–14 months | 8–12 months | 6–9 months |
| Bid/buy-out phase | 8–14 weeks | 4–8 weeks | Continuous |
| Construction phase | Standard | Standard or slightly faster | 12–22% faster |
| Total schedule | Baseline | -8 to -12% vs DBB | -18 to -28% vs DBB |
For broader industry schedule data, our Commercial Construction Material Lead Times 2026 piece shows where schedule risk concentrates today.
Risk allocation
| Risk Category | DBB | CM-at-Risk | Design-Build |
|---|---|---|---|
| Design errors | Owner / Architect | Owner / Architect | Design-Builder |
| Constructability issues | Owner (via change orders) | Shared | Design-Builder |
| Subcontractor performance | GC | CM | Design-Builder |
| Coordination schedule slip | Owner (most likely) | Shared | Design-Builder |
| Cost overruns during construction | Owner | Shared (above GMP) | Design-Builder |
| Material lead time exposure | Owner / GC | Shared | Design-Builder |
When each method actually wins
Design-Build Wins
- Significant MEP, process, or refrigeration integration
- Schedule is a primary success criterion
- Owner wants a single point of contact
- Manufacturer integration upfront (IMP, PEMB)
- Limited internal design/engineering capacity
- Complex subcontractor coordination
CM-at-Risk Wins
- Public-funded healthcare, education, government
- Design partially complete, preconstruction input needed
- Budget/schedule important but not dominant
- Owner has fixed design preferences
- Project $20M+
- Sophisticated owner team
Design-Bid-Build Wins
- Scope is simple and well-defined
- Lowest upfront bid is the criterion
- Schedule is not primary
- Design risk is low
- Repeat designs (prototype QSRs, retail)
- Experienced internal owner staff
Most TCG projects fall in the design-build column. Cold storage, cannabis cultivation, QSR, data center, and IMP-heavy industrial all favor design-build delivery for the same fundamental reason: design decisions and construction execution are too tightly coupled to be done sequentially.
What the academic and industry data say
Multiple research studies have benchmarked these three methods over decades. The most-cited industry data:
| Source | Finding |
|---|---|
| Construction Industry Institute (CII) | Design-build delivers 6% lower unit cost and 33% faster schedule vs DBB |
| Penn State / Charles Pankow Foundation | Design-build shows 5.2% better cost and 13.3% better schedule performance |
| Design-Build Institute of America (DBIA) | Design-build now ~47% of all non-residential US construction, projected 50%+ by 2027 |
| AGC of America CMAR Analysis | CM-at-Risk shows ~3% better cost outcomes than DBB, comparable schedule |
The data points consistently in one direction: integrated delivery methods beat sequential methods on both cost and schedule.
The five hidden costs of choosing the wrong method
Where the Money Actually Leaks
1. Coordination gaps in design-bid-build. Architect designs around clearance assumptions that subs can't actually execute. Result: $80K–$400K of rework on a typical $5M project.
2. Late-stage value engineering in CM-at-Risk. GMP comes in 12% high. Owner directs $1.2M of VE cuts. Critical refrigeration redundancy gets stripped. Three years later, the operator is paying for it.
3. Manufacturer lead-time exposure. Architect specs an IMP panel the manufacturer discontinued. Bid releases. GC discovers the issue. Substitution adds 14 weeks. We've seen this twice in 2026 already.
4. Change order erosion. Every change order has a markup. On DBB, change orders compound — the typical change order has a 12–18% markup on direct cost plus impact on adjacent scope.
5. Schedule penalty cascades. A 4-week design delay often becomes a 12-week occupancy delay because of cascading impacts on tenant moves, equipment commissioning, and operational ramp-up.
How AI estimating changes the calculation in 2026
Until recently, the case for design-bid-build was that competitive subcontractor bidding produced the lowest market price. That argument has weakened.
In 2026, AI-powered estimating tools like our TCG.ai estimator allow design-build and CM-at-Risk teams to benchmark subcontractor bids against market-calibrated data instantly. The competitive bidding premium of design-bid-build has shrunk from the historical 5–8% advantage to roughly 1–3% — and that 1–3% is more than wiped out by the higher change order and schedule slip exposure of sequential delivery.
This is why design-build is taking market share. The cost-control mechanism that justified design-bid-build for 50 years has been replaced by faster, better data — and design-build is the delivery model that takes the best advantage of that data flow. See our AI-Powered Commercial Construction Estimator article.
How to decide for your specific 2026 project
Five diagnostic questions to ask yourself:
- How complete is your design today? 0–30% complete → design-build usually right. 60–90% → CMAR or DBB.
- How critical is schedule? Every month of delay costs material money → design-build.
- How specialized is your facility? Refrigeration, IMP, process MEP, vertical-specific design (cannabis, data center, food processing, biotech) → design-build.
- How experienced is your team? First-time owner with limited capacity → design-build. Sophisticated developer with internal teams → any method may work.
- Is there a regulatory or funding driver? Public funding often mandates DBB. CMAR common in healthcare/higher ed. Private commercial in 2026 trends toward design-build.
For broader context on selecting a GC, see our How to Select the Best Commercial GC Guide.
Frequently asked questions
Is design-build always cheaper than design-bid-build?
No — but it's cheaper on average, and meaningfully cheaper for complex, MEP-intensive, or fast-track projects. For simple shell buildings with no specialized scope, the cost difference is smaller and the schedule difference becomes the deciding factor.
Can I switch delivery methods mid-project?
Rarely, and expensively. The contract structure is set at the outset. The closest thing is converting a CMAR contract to a lump-sum design-build at GMP — but this requires aligned parties and significant restructuring.
Does TCG do CM-at-Risk work?
Yes, when it's the right fit. We deliver under all three models. Our default is design-build because it's where we add the most value, but we'll deliver CMAR or design-bid-build when the project requires it. See our Commercial Construction Delivery Methods piece.
What's the typical fee for a design-build project?
Design-build fees vary, but a typical commercial design-build project carries 4–6% GC fee plus 4–7% design fee, totaling 8–13% of construction cost. Sequential design-bid-build often runs 10–14% combined when fully tallied across both the architect and GC.
How does design-build affect bond and insurance structure?
Design-build typically carries a single performance and payment bond from the design-build entity, and a single CGL policy covering both design and construction. Professional liability covers the design portion. Owners benefit from simplified claims handling — one entity is responsible.
Do you handle architectural and engineering design in-house at TCG?
Yes. We deliver design-build through 3rd Act Architecture, 9BA MEP Engineering, and in-house structural engineering. Single-source accountability is the entire value proposition.
