Establishing Phase-Based Review Processes and Completion Criteria

How architects and project managers build the formal QC system: structuring review checkpoints at 30/60/90/100%, defining completion criteria, formalizing comment registers, and using phase gates to control quality, cost, and schedule progression.

ARE · Project Management

Overview

Quality doesn't happen by accident on a design project. It happens because someone built a system of structured checkpoints, defined what "done" looks like at each stage, and gave the project manager a framework for saying "stop" when the work isn't ready to move forward.

That's what this topic is about: the mechanics of the quality control system itself. Not what the reviews are checking for (that's a separate question), but how the review process gets structured, who does what, and what criteria trigger a go or no-go decision.

The standard industry framework uses percentage-complete milestones, typically at 30%, 60%, 90%, and 100%, as formal gates. At each gate, the project team evaluates deliverables against predefined completion criteria. If the criteria are met, the project advances. If not, it holds until deficiencies are corrected.

For the ARE, you need to analyze how review processes should be structured for different project conditions. A small tenant improvement and a federal courthouse don't need the same review apparatus. The DOE calls this a "graded approach," and the exam tests whether you can match review rigor to project risk. You also need to understand the contractual dimension: under AIA B101, phase completion triggers billing rights. The approval letter at each phase gate is simultaneously a quality checkpoint and a contractual milestone.

This topic covers the system architecture of QC. Think of it as building the engine. Other topics cover what the engine measures.

Essential

QA vs QC: The Distinction the Exam Will Test You On Before getting into the mechanics, you need to separate two concepts that candidates constantly confuse. Quality Assurance (QA) is the system. It's the policies, procedures, checklists, and organizational structures that a firm puts in place to prevent errors. QA is proactive. It happens before and during the work. Quality Control (QC) is the checking. It's the reviews, inspections, and evaluations that verify the work meets standards. QC is reactive. It happens after some portion of the work is done. A phase gate review is a QC activity. The review process that defines when it happens, who conducts it, and what criteria apply is a QA system. The exam may present scenarios where the distinction matters, for instance, asking whether a firm should invest in better QA procedures or more QC reviews to address a pattern of coordination errors.

Professional

Scenario: A Growing Firm Formalizes Its QC Process Consider a 40-person architecture firm that has grown from 15 staff in three years. The founding principals built quality into their projects through personal oversight, checking every set of drawings before they went out. But with five concurrent projects and several recent senior retirements, that approach no longer scales. Two recent projects went to bid with coordination errors that produced change orders. The firm needs a formal QC system. The quality director starts by distinguishing QA from QC activities. The QA side includes creating standardized checklists for each phase (what must be verified at SD, DD, 60% CD, and 100% CD), documenting the firm's review procedures in a process manual, and establishing rules for who reviews what. The QC side includes conducting the actual reviews, logging findings in comment registers, and tracking resolution. The checklists become the firm's institutional memory.

TLDR

QA is the system (checklists, procedures, policies). QC is the checking (reviews, inspections, evaluations). The exam tests whether you know the difference.
Standard review milestones: conceptual, 30%, 60%, 90%, 100%, bid document verification. "Percent" means budget spent, not time elapsed.
Completion criteria must be binary and measurable across five dimensions: technical completeness, budget alignment, schedule conformance, regulatory status, and owner authorization.
A phase gate is a go/no-go decision, not a progress report. Three outcomes: hold, advance with conditions, or loop back.
The review comment register tracks every finding with discipline, type, severity, responsible party, and resolution deadline. Unresolved comments escalate.
Review depth scales with project risk. A dental office renovation and a hospital wing don't get the same review apparatus. The DOE calls this a "graded approach."
VE timing is a process design decision: at or before 30% design for projects over $2 million in construction cost.
Phase completion under AIA B101 triggers billing rights and requires owner authorization. The approval letter is both a quality gate and a contractual milestone.
Cost estimate accuracy tightens with design maturity: ROM (+/- 50%), Budgetary (+/- 20-25%), Definitive (+/- 10%).
Peer reviewer independence matters. GSA requires reviewers from a national register; DOE requires External Independent Reviews for projects over $100 million.

ELI5

Imagine you're baking a wedding cake for a really important event. You wouldn't just throw ingredients together and hope it turns out great, right? You'd have a plan. First, you'd create a checklist. After mixing the batter: did you add the right amount of sugar? After baking: did the layers come out the right height? After stacking: are the layers straight? After frosting: does it match the design the customer wanted? That checklist is your quality system. It tells you WHEN to stop and check, WHAT to look for, and WHAT to do if something's wrong. Architects building a new building need the same kind of system, except their "cake" takes months or years to finish and costs millions of dollars. So they break the work into stages. At roughly 30% done, they stop and check. At 60%, they check again. At 90%, one more time.

AREProject Management

Establishing Phase-Based Review Processes and Completion Criteria

2 min read239 words

Building the QC Machine That Catches Problems Before They Cost Real Money

This topic covers the system architecture of QC. Think of it as building the engine. Other topics cover what the engine measures.

Want to track your progress and access more study tools?

Create a free account