Quick Summary
- Design review should happen at every stage: Schematic, DD, 50%, 90%, and IFC
- Comment tracking across submissions ensures issues are actually resolved
- Revisions can introduce new issues while resolving old ones—track both
- AI tools can automate re-review and comment closure verification
Projects rarely get it right on the first submission. Design evolves through multiple stages— Schematic Design (SD), Design Development (DD), and Issued for Construction (IFC)—with interim reviews at 50%, 90%, and beyond. Each submission is an opportunity to catch issues before they become expensive field problems. But here's the challenge: how do you ensure previous comments were actually addressed?
This guide covers best practices for reviewing designs across stages, tracking comment implementation, and preventing issues from slipping through the cracks between submissions.
Why Multi-Stage Reviews Matter
Most coordination issues don't appear at a single stage—they emerge as designs become more detailed. A structural layout that looked fine at SD may create MEP routing conflicts at DD. A fire protection concept that worked on paper may clash with architectural ceilings at IFC.
The Cost of Skipping Stage Reviews
- Issues discovered at IFC cost 10× more to resolve than at SD
- Field-discovered issues cost 50–100× more than design-phase catches
- Untracked comments often resurface as RFIs during construction
Typical Design Stages & What to Review
Schematic Design (SD)
- • Building layout and massing
- • Code zoning compliance
- • Structural system selection
- • Major MEP concepts
- • Site & civil layout
Design Development (DD)
- • Detailed floor plans & sections
- • MEP routing & coordination
- • Structural detailing
- • Fire/life safety systems
- • ADA/accessibility compliance
50% / 90% Progress Sets
- • Verify prior comments resolved
- • Cross-discipline coordination
- • Specification alignment
- • New issues from revisions
- • Constructability review
IFC / Pre-Construction
- • Final coordination check
- • All comments closed
- • Spec-to-drawing alignment
- • Permit readiness
- • Trade-specific review
Comment Tracking Across Submissions
The most common failure mode in multi-stage review isn't missing issues—it's losing track of issues that were already identified. Comments get raised at DD, drawings get revised, and then nobody verifies whether the revision actually addressed the concern.
What to Track for Each Comment
| Field | Purpose |
|---|---|
| Comment ID | Unique identifier for tracking across submissions |
| Discipline | Architectural, Structural, MEP, etc. |
| Sheet Reference | Where the issue was found |
| Severity | Critical, Major, Minor, or Informational |
| Status | Open, Addressed, Partially Addressed, Closed |
| Stage Raised | Which submission the issue was first identified |
| Response/Resolution | How the design team addressed it |
The Re-Review Process
When a revised submission arrives, the review process has two distinct objectives:
1. Verify Prior Comments
Check each previously raised comment against the updated drawings. Is it fully resolved, partially addressed, or still open? Did the fix create new issues?
2. Flag New Issues
Revisions often introduce new problems. A structural change to address one comment may create a new MEP conflict. Always review for regression.
Common Multi-Stage Review Failures
1. "Closed" Comments That Weren't Fixed
Design team marks comments as resolved, but the actual fix is incomplete or incorrect. Without verification, these slip through.
2. New Issues from Revisions
Fixing one problem creates another. A structural member relocation resolves a clash but now conflicts with ceiling heights.
3. Lost Comment History
Comments tracked in email or disconnected spreadsheets get lost between stages. No single source of truth for what was raised and resolved.
4. Inconsistent Review Depth
Full review at DD, cursory review at 90%. Critical issues missed because "we already reviewed this discipline."
How AI Automates Stage Reviews
Manual re-review is time-consuming and error-prone. AI-powered review tools can automate much of the verification process:
Automated Re-Review
AI compares previous and current submissions, checking whether each comment location has been modified and if the stated issue is resolved.
Regression Detection
Flags new issues introduced by revisions. Catches the "fix one, break another" pattern that often slips through manual review.
Progress Tracking
Generates reports showing comment closure rates, outstanding issues by discipline, and design quality trends across submissions.
Best Practices for Multi-Stage Review
Stage Review Checklist
Typical Pricing for Multi-Stage Reviews
Review effort and pricing naturally differs across stages. Initial reviews require full analysis; re-reviews focus on verification and delta checking.
| Stage | Typical Effort |
|---|---|
| Initial Review (DD or IFC) | 100% of base fee |
| 2nd Submission Re-Review | ~40–50% of initial fee |
| 3rd+ Submission Re-Review | ~25–35% of initial fee |
| Volume Programs | Multiple iterations bundled at lower effective cost |
Conclusion
Design review isn't a one-time event—it's a process that spans the entire design phase. Projects that implement rigorous multi-stage review with proper comment tracking catch issues earlier, reduce rework, and arrive at construction with cleaner documents.
The key is systematic tracking: every comment needs a unique ID, a clear status, and verified closure. Whether you use manual processes or AI-powered tools, the discipline of following comments from identification to resolution is what separates projects that discover issues in the field from those that catch them on paper.
The Bottom Line
Multi-stage review with comment tracking isn't extra overhead—it's the minimum viable process for delivering buildable documents. Every untracked comment is a potential RFI waiting to happen.