“Plan review” means two different things in construction, and conflating them is the source of a lot of confusion \u2014 and a lot of missed conflicts.<\/p>\n
The first meaning is regulatory: an Authority Having Jurisdiction (AHJ), typically a municipal building department, reviews your submitted documents for code compliance before issuing a permit. The second meaning is internal: the design team, GC, or an independent reviewer reads the drawing set to catch coordination errors, specification mismatches, and cross-discipline conflicts before construction starts.<\/p>\n
Both are important. But they’re not the same activity, they catch different problems, and they happen on different timelines. Understanding where each one falls short is how you understand why construction projects still run over budget even when they pass code review.<\/p>\n
When you submit construction documents to a building department, a plans examiner reviews them for compliance with the adopted building code \u2014 the IBC, local zoning ordinances, fire and life-safety requirements, accessibility standards, and similar regulatory frameworks. The examiner’s job is to confirm that what you’re proposing to build meets the minimum standards the jurisdiction requires.<\/p>\n
This review is mandatory for permitted work, and it can be extensive. Common rejection reasons include:<\/p>\n
Permit review timelines vary enormously by jurisdiction, project type, and current backlog \u2014 ranging from a few days for over-the-counter residential permits to several months for complex commercial projects in busy departments.<\/p>\n
What AHJ review does not<\/strong> do: it does not read every sheet against every other sheet. It does not verify that the structural drawings and the mechanical drawings are coordinated. It does not catch a dimension that conflicts between the architectural floor plan and the structural layout. Those are internal coordination problems, and no building department is staffed to find them. They’re your problem.<\/p>\n The second type of plan review is the one the project team runs \u2014 or should run \u2014 before the documents are issued for bid or construction. This is where cross-discipline conflicts, specification mismatches, and coordination gaps are supposed to get caught.<\/p>\n In practice, it looks something like a senior project architect or PM walking through the drawing set looking for problems. On well-resourced projects it’s formalized as a design review gate; on most projects it’s squeezed into whatever time exists before the issue deadline. Either way, the core task is the same: read the full document set \u2014 every discipline, every sheet, plus the project manual \u2014 and identify places where things don’t agree.<\/p>\n This is the review that a construction document review checklist<\/a> is designed to structure. And it’s the review where the most expensive conflicts get caught or missed.<\/p>\n The conflicts that drive change orders and rework are distributed problems. They don’t live on one sheet \u2014 they live between<\/em> sheets, between disciplines, between the drawings and the specifications. That’s what makes them hard to catch manually.<\/p>\n The specification section calls one product or assembly. The drawing detail shows another. Neither is “wrong” on its own \u2014 they just disagree. When the field hits it, someone has to issue a clarification, price the difference, and often demo work that’s already in place.<\/p>\n A dimension or elevation on one sheet doesn’t match the same element on another. Architectural and structural floor plans carry different slab thicknesses. A section cut references a detail that was never updated after a design revision. These conflicts are obvious in retrospect and genuinely hard to catch when you’re reading one sheet at a time under deadline pressure.<\/p>\n The canonical case: structure, MEP, and architecture each solve their own scope correctly, and the seams between them don’t line up. A structural shear wall is placed exactly where the mechanical drawings route a main supply duct. Neither drawing is internally inconsistent. Together, they’re a $28,000\u2013$45,000 rework if caught in the field \u2014 and a coordination note if caught in review.<\/p>\n On one real project that Flikt reviewed, more than 40 change orders traced back to conflicts between 2D plan sheets. These weren’t exotic failures; they were routine coordination gaps that didn’t surface until construction had started. The MEP coordination failures in multifamily<\/a> post documents what this pattern looks like at a portfolio scale.<\/p>\n Some conflicts are absences, not contradictions. A renovation adds a structure, but the fire-alarm device layout isn’t extended onto it. An ADA clearance is required but no sheet accounts for it. A detail is referenced in the drawings but never actually drawn. These are the hardest type to catch because there’s no internal inconsistency to notice \u2014 you only find them by knowing what should be there.<\/p>\n Manual review catches some of these conflicts. It doesn’t catch all of them \u2014 not because reviewers aren’t competent, but because the task is genuinely hard at scale. A large drawing set means hundreds of sheets across architectural, structural, mechanical, electrical, plumbing, fire protection, fire alarm, low-voltage, telecom, civil, landscape, interior design, food service, and signage disciplines, plus a project manual with dozens of specification sections. Reading every sheet against every other sheet, without fatigue, without missing a cross-reference, is not something humans do reliably under deadline conditions.<\/p>\n Errors caught during construction cost far more<\/strong> to fix than the same errors caught during plan review. The industry’s cost-of-change curve (Construction Users Roundtable, 2004) is the standard statement of it, and it is a principle rather than a measured multiplier. The math is straightforward: a conflict caught on a PDF costs a markup and a design coordination meeting. The same conflict caught in the field costs demolition, rework, a change order, and schedule impact. That gap is why thorough internal review isn’t a nice-to-have \u2014 it’s the highest-leverage spend in preconstruction.<\/p>\n A constructability review<\/a> extends this logic further, asking not just “are the documents coordinated?” but “can this actually be built as documented?” \u2014 catching sequencing and constructability problems that coordination review might not surface.<\/p>\n AI document review is built specifically for the internal QA task: reading the full drawing set comprehensively, cross-referencing every discipline against every other, and surfacing conflicts without fatigue or deadline pressure. This is AI construction plan review<\/a> \u2014 and it runs on the 2D PDF documents the project already has, with no BIM or 3D model required.<\/p>\n That last point matters because the window where internal review is most valuable \u2014 during design review, before bid, before mobilization \u2014 is frequently the window where a 3D model doesn’t exist yet, or isn’t coordination-ready. For more on what AI-based 2D review catches that model-based clash detection<\/a> doesn’t, see clash detection without BIM<\/a>.<\/p>\n The practical output is a conflict report: specific findings, keyed to sheets and specification sections, organized by type (spatial conflict, spec mismatch, missing coordination, code\/clearance gap). The benchmark Flikt uses across projects is conflicts per 100 sheets<\/strong> \u2014 a normalized drawing-quality metric that lets owners and GCs compare document quality across design teams and project phases rather than relying on gut feel.<\/p>\n Here’s how the two types of plan review fit into a typical project timeline:<\/p>\n Design development through CD completion<\/strong> \u2014 Internal QA review runs iteratively as discipline sets are issued. Cross-discipline conflicts and spec mismatches are most efficiently caught here, when revisions are still cheap and the design team is still assembled.<\/p>\n Before bid \/ CD issue<\/strong> \u2014 Final internal review pass. This is where a construction document review checklist<\/a> is most directly useful \u2014 a structured sweep that confirms the documents are ready to go out without the coordination gaps that generate pre-bid RFIs.<\/p>\n Permit submission and AHJ review<\/strong> \u2014 Runs in parallel or immediately after CD completion. AHJ review is largely independent of internal QA; passing permit review does not mean the documents are free of coordination conflicts.<\/p>\n Pre-construction \/ GMP negotiation<\/strong> \u2014 GCs running their own document review before committing a Guaranteed Maximum Price have a particular interest in finding conflicts that will show up as change orders. Flikt for general contractors<\/a> is aimed directly at this moment.<\/p>\n The separation is worth stating plainly: a stamped, permitted set of documents is not the same as a coordinated set of documents. Permit review is a code-compliance check. Coordination review is a buildability check. Both have to happen; they serve different purposes.<\/p>\n The conflicts hiding in a typical drawing set aren’t hypothetical. Flikt tracks them across real projects and the pattern is consistent: coordination gaps between disciplines, specification mismatches between drawings and specs, missing-coordination problems that no single discipline’s review would surface. The individual structural conflicts we’ve documented carry $28,000\u2013$45,000 in rework per occurrence \u2014 and they occur more than once per project.<\/p>\n For owners and GCs evaluating how much document quality matters before construction starts, the evidence page<\/a> has the project-level data. For the broader framework of where coordination failures show up and what they cost, the AI construction plan review<\/a> guide covers the full picture.<\/p>\n Flikt reviews 2D construction plan sets for cross-discipline conflicts before construction \u2014 no BIM required. See the evidence<\/a>, estimate your exposure<\/a>, or schedule a demo<\/a>.<\/em><\/p>\nInternal QA \/ coordination review<\/h3>\n
Where conflicts actually hide<\/h2>\n
Spec-vs-drawing conflicts<\/h3>\n
Cross-sheet conflicts<\/h3>\n
Cross-discipline conflicts<\/h3>\n
Missing-coordination gaps<\/h3>\n
The catch rate problem<\/h2>\n
How AI plan review fits into this picture<\/h2>\n
Who runs which review \u2014 and when<\/h2>\n
The practical stakes<\/h2>\n
\n