Short answer: a complete review covers four things: code and compliance, coordination between the disciplines and against the specifications, completeness of everything needed to permit and bid, and constructability. The expensive misses are almost always in the second one. Each discipline’s package can be internally correct and still contradict another’s, which is why the cross-discipline section below is the one to slow down on.
A set of construction documents is never just one document. It’s an architect’s drawings, a structural engineer’s sheets, an MEP consultant’s layouts, a civil set, a project manual, and dozens of detail sheets — all produced by different teams, often on different timelines, expected to agree perfectly when field crews rely on them.
They rarely do. And the conflicts hiding in the seams between those documents — a shear wall that lands where a duct has to run, a spec that calls one product while the drawing shows another, a fire-alarm layout that was never extended onto the new structure — are exactly where change orders come from. The measured picture: design errors and omissions alone account for 1.5% of contract value in field rework on building projects, and 1.8% across all building types (sources). That delta is the entire case for doing document review well.
This checklist is organized around the four dimensions a complete review has to cover: coordination, completeness, constructability, and code/compliance. Use it as a working tool — and pay close attention to the cross-discipline section, because that’s where the expensive conflicts hide.
What document review actually verifies
Before the checklist, a framing point. Most review time is spent on a single discipline at a time — the architect checking the architectural set, the structural engineer checking structure. That kind of internal review is necessary but not sufficient. The conflicts that become change orders are usually between disciplines, not within them: the mechanical drawings are internally coherent, and the structural drawings are internally coherent, and they still disagree about what goes where.
A complete construction document review has to cover four things:
- Code and compliance — does the project meet the applicable building code, ADA/accessibility requirements, and AHJ-specific requirements, and can the documents demonstrate it?
- Coordination — do the disciplines agree with each other, and does the drawing set agree with the specs?
- Completeness — is everything required to permit, bid, and build actually in the documents?
- Constructability — can a competent trade actually build this as drawn, without ambiguity or contradiction?
The checklist below tracks all four.
The checklist
General / TOC and document control
Architectural
Structural
Mechanical, Electrical, Plumbing (MEP)
Civil and site
Specifications vs. drawings
The cross-discipline coordination section: where conflicts actually hide
Every category above is necessary. This one is where the expensive surprises live.
Cross-discipline coordination failures don’t show up within any single discipline’s package — they appear only when you read two or more packages together. On one real project, more than 40 change orders traced back to conflicts between 2D plan sheets. A single structural conflict — a shear wall placed where the mechanical ductwork had to route — carried $28,000–$45,000 in rework and 14–21 days of schedule. Neither the structural drawings nor the mechanical drawings were internally wrong. They just didn’t agree with each other.
The cross-discipline checks that catch these conflicts:
Structure vs. MEP
Architecture vs. MEP
Architecture vs. structure
Specifications vs. drawings (cross-discipline)
Review milestones: 60%, 90%, and 100% CDs
Document review isn’t a one-time gate — it’s most effective layered across the design process.
60% CDs — completeness and coordination check At this milestone the set should be complete enough to validate scope and layout. The review goal is big-picture: are all disciplines present? Do the floor plans, structural grid, and MEP routing decisions already have the seeds of coordination conflicts? Catching a structural-vs-duct conflict at 60% means a coordination note, not a redesign.
90% CDs — conflict and compliance check This is the primary conflict-detection pass. The drawings should be complete enough to permit review. Every cross-discipline check in the section above applies here. Spec sections should be drafted. This pass catches the bulk of the conflicts before bidding.
100% CDs / permit set — final verification The final check is for completeness and consistency: all addenda and revision clouds reconciled, spec table of contents matches sections included, all reference notes resolve. This is not the moment to find the first structural-vs-MEP conflict — by 100% CDs, those should have been resolved weeks earlier.
Manual vs. AI-assisted review for large sets
For a 20-sheet renovation package, a careful senior reviewer can catch most of the conflicts above in a day. For a 400-sheet multifamily building or a 1,000-sheet commercial project, that same reviewer faces a combinatorial problem: every sheet has to be checked against every other relevant sheet, across every discipline, plus the project manual. Fatigue and deadlines guarantee that conflicts will be missed.
This is where AI construction plan review changes the economics. AI reads the full 2D PDF set comprehensively — structure against MEP, drawings against specs, sheet against sheet — without fatigue and without sampling. It surfaces the same class of conflicts in the checklist above: spec-vs-drawing mismatches, cross-discipline spatial conflicts, missing-coordination gaps. It doesn’t replace the judgment calls that require engineering expertise; it eliminates the reading-comprehension failures that result from a 500-sheet set and a two-day deadline.
The same review that historically required senior staff hours on a large set runs in hours on the full document package. What comes back isn’t a model or a clash report — it’s a prioritized conflict log, tied to sheet numbers, readable by anyone on the project team. For the full technical and workflow comparison, see building plan review process and clash detection without BIM.
For projects that are integrating this into a broader pre-construction discipline, the pre-construction QA checklist covers the upstream workflow — from drawing release through RFI triage — that this checklist feeds into. And if you want to see what real conflict counts look like across real projects, the conflicts per 100 sheets benchmark puts this checklist in quantitative context.
Using this checklist in practice
The full checklist above is most useful in two configurations:
-
Internal quality gate — run it before issuing drawings to GC or bidders. The cross-discipline section in particular is designed to catch the coordination failures that generate RFIs and become change orders.
-
Pre-bid review — GCs and owner’s reps can use it to assess a drawing set before committing to a price. Incomplete or uncoordinated documents at bid time are a risk multiplier; knowing the conflict density before you bid lets you price it or push back.
Either way, the value is in the cross-checking, not the per-discipline checklists. The per-discipline items confirm that each package is internally complete. The coordination items are where the real review happens.
Flikt reviews your 2D construction plan sets for cross-discipline conflicts before you build — no BIM required. See what AI plan review finds on real projects or contact us to review your next set.
Frequently asked questions
What should a construction document review cover?
Four dimensions: code and compliance, coordination between disciplines and between drawings and specs, completeness of everything needed to permit, bid and build, and constructability, meaning a competent trade can build it as drawn without ambiguity.
Which section of the checklist matters most?
Cross-discipline coordination. Each discipline’s package can be internally correct and still contradict another’s. On one real project more than 40 change orders traced back to conflicts between 2D plan sheets, and a single shear wall in a duct route carried $28,000 to $45,000 in rework and 14 to 21 days of schedule.
What are the most common cross-discipline checks?
Duct mains, waste lines and conduit runs clearing beams, shear walls and framing; equipment rooms and shafts sized to the equipment schedules; finish-floor elevations agreeing with structural slab elevations; and fire-rated assemblies on the drawings matching the listed assemblies in the specifications.
When should document review happen?
Layered across the design process rather than as one gate: a completeness and coordination check at 60% CDs, the primary conflict and compliance pass at 90%, and a final sweep at 100% before the set goes out for bid or permit.
How much more does a conflict cost when it is found in the field?
There is no reliable published multiplier, but the mechanism is clear: the field version includes demolition, rework, a change order and schedule impact, none of which a markup on a PDF carries.
This checklist covers the design-review side. For the bid-stage version of the same problem, where a gap between the drawings and the specifications turns into money left on the table, see how to find missed scope in bid documents before bid day.
Ready to catch conflicts early?
Upload your plan set and get AI-powered conflict detection — no BIM required.
A checklist tells you what to look for. For what an automated pass actually hands back — the five finding types, with examples taken from published reports — see what we check.
The pre-IFC coordination checklist
50 checks before you issue for construction, ordered by what actually goes wrong — built from 1,516 findings across 34 real plan sets. Comes as an assignable tracker (Excel) with status, owner and due date, plus a printable PDF. No call, no card.