What CMAR Actually Is
Construction Manager at Risk is the progressive contract model I managed hands-on during the Bowmanville Train Line Extension — a $2 billion rail extension in Ontario, Canada, delivered under one of the most ambitious CMAR engagements in North American transit history.
The model is two-phased, and understanding both phases is essential to understanding why CMAR produces the outcomes it does — both the successes and the failure modes.
Phase 1: Preconstruction
The owner selects a Construction Manager based on qualifications and fee — explicitly not on lowest bid price. This is the first and most important distinction from traditional contracting. The CM is chosen for who they are and what they can contribute during design, not for how aggressively they will price the work at tender.
Once engaged, the CM joins the design process as an active participant. Their core preconstruction deliverables are constructability reviews (applying field construction knowledge to design decisions before they are locked), cost estimating (building and maintaining an open-book estimate of the project as design evolves), value engineering (identifying alternative approaches that reduce cost or improve buildability without compromising the owner’s requirements), schedule development (building a construction schedule that reflects how the project will actually be built, not a theoretical programme), risk identification (surfacing and quantifying construction risks while there is still time and design flexibility to mitigate them), and procurement planning (identifying long-lead materials and subcontracting strategies that reduce cost and schedule risk).
This phase is where the value of CMAR is created. The contractor’s field knowledge shapes the design before it gets locked in. Problems that would have become change orders in traditional contracting get solved collaboratively during preconstruction.
Phase 2: GMP and Construction
When the design reaches sufficient maturity — typically 60-90% complete, depending on the project type and the owner’s tolerance for residual uncertainty — the CM and owner negotiate a Guaranteed Maximum Price. The GMP is built on transparent, open-book cost data: the actual cost of the work (labour, materials, subcontractors, equipment), plus the CM’s fixed fee (typically 3-8% of cost of work), plus shared contingency.
The financial logic of the GMP is based on transparency and shared risk. If the project comes in under the GMP, the savings are shared between owner and CM in a pre-agreed ratio — this is the ‘gain.’ If the project exceeds the GMP, the CM absorbs the overage — this is the ‘pain.’ This structure fundamentally changes the CM’s incentive compared to traditional contracting. They are now motivated to find efficiencies and avoid problems, not to identify claim opportunities.
What Conditions CMAR Requires
CMAR is not a magic fix for construction delivery. It is a model with specific conditions that need to be in place for it to perform. When those conditions are absent, CMAR can actually underperform traditional contracting — because you have added cost and time without getting the collaborative benefit.
The owner needs to be capable of active participation in preconstruction. An owner who treats the preconstruction phase as a contractor activity to be observed rather than a collaborative process to be participated in will not get the benefit of early contractor involvement. The design decisions that preconstruction is supposed to inform get made without the CM’s input, and the preconstruction becomes a billing exercise.
The CM needs to have real preconstruction capability — not just estimators who can produce a GMP, but constructability specialists, procurement strategists, and schedule analysts who can genuinely contribute to design development. A general contractor who wants early access to a project but has thin preconstruction capability will deliver thin preconstruction value.
The governance structure needs to allow fast decision-making. One of the most persistent failure modes I saw in CMAR delivery was an owner’s governance framework designed for traditional procurement — sequential decisions, multi-committee approval — meeting a contract model that required collaborative, fast decisions during preconstruction. The friction was immediate and ongoing. The contract’s potential was constrained by the governance.
The contract needs clear GMP amendment procedures. CMAR does not freeze scope at GMP establishment. When the owner adds scope or conditions change, the GMP needs to be adjusted through a clear, pre-agreed process. Ambiguity in this process is the single most common source of CMAR disputes I have observed.
What Goes Right and What Goes Wrong
When these conditions are in place, CMAR produces outcomes that traditional contracting consistently fails to achieve: cost certainty at GMP establishment, fewer change orders during construction, faster problem resolution, and a project team that functions as a partnership rather than an adversarial relationship.
When the conditions are absent — and they often are, particularly on first CMAR engagements — the model creates overhead without creating value. I will continue to share specific failure modes from my experience in coming articles, because understanding what goes wrong is as important as understanding what the model is designed to do.