Most RFI-reduction advice focuses on processing RFIs faster — better logs, quicker turnaround, cleaner submittals. That’s workflow management. It doesn’t reduce RFI volume; it just handles the backlog more efficiently.
This post is about something different: closing the drawing gaps that generate RFIs in the first place, before bid, before mobilization, before anyone touches the site. The distinction matters because an RFI resolved in the field is still a disruption — to the schedule, to the GC’s budget, to the subcontractor waiting on an answer. An issue caught during plan review is a redline on a PDF.
The four steps below are a practical sequence for a pre-issue coordination review. They run on your 2D drawing set. No BIM required.
RFI volume is a quality indicator, not just a workflow problem
High RFI counts are a symptom. The underlying condition is a drawing set that doesn’t fully agree with itself — sheets that contradict each other across disciplines, specs that call something different from what the drawing shows, coordination gaps where required elements simply aren’t on the sheet they belong on.
What an RFI actually costs breaks down the economics: every RFI consumes hours from the GC, the architect, and the engineer before anyone types a formal response. Multiply by a hundred RFIs on a mid-size project and you’re looking at a material schedule and cost impact before a single change order is written.
The projects with the lowest RFI volumes typically share one thing: the drawings were reviewed comprehensively, across disciplines and against the specs, before they were issued for construction. That’s the window this playbook targets.
Step 1: Run a cross-discipline coordination check
The most expensive conflicts are the ones no single discipline’s reviewer would catch, because they’re distributed across two or more drawings that each look correct in isolation.
A shear wall and a duct main that occupy the same space don’t show up on the structural sheets as a problem, and they don’t show up on the mechanical sheets as a problem. They only show up as a problem when someone reads both sheets against each other — or when the crew in the field discovers they can’t build both at once. On one Flikt-reviewed project, a structural conflict of exactly this type carried an estimated $28,000–$45,000 in rework plus 14–21 days of schedule impact. Caught in review, it’s a coordination note. Caught in the field, it’s a change order.
The coordination check should explicitly ask:
- Structure vs. MEP: Do beam depths, shear walls, and footing locations leave routing clearance for the duct, pipe, and conduit runs shown on mechanical, plumbing, and electrical?
- Architecture vs. structure: Do floor-to-floor heights, slab thicknesses, and wall locations on the architectural drawings match the structural assumptions?
- Architecture vs. MEP: Are penetrations, chases, and equipment rooms coordinated between the architectural and MEP disciplines, or is the architectural design silent on what it needs to accommodate?
- Civil vs. architecture: Do site grading, utility routing, and access alignments match the building’s ground-floor layout?
This isn’t a theoretical checklist — it’s the set of seam-points where the most expensive conflicts consistently hide. The Proceso de coordinación de MEP goes deeper on how these seam-points form and what a structured coordination workflow looks like.
Step 2: Run a completeness and spec-vs-drawing check
Spatial conflicts get most of the attention, but some of the most RFI-productive problems in a drawing set are simply omissions — something required that’s absent from the sheet it should appear on.
Spec-vs-drawing conflicts are a parallel category: the specification section and the drawing call different products, assemblies, ratings, or finishes. Neither is “wrong” on its own; they just disagree. The GC hits the discrepancy in the field, sends an RFI, and waits. Multiply by how many times a finish schedule, door schedule, or equipment spec drifts from what the drawing shows, and you have a predictable RFI queue.
Run this check against:
- Finish and door schedules vs. plan and detail sheets — do referenced materials and types actually appear?
- Equipment schedules vs. MEP plans — is every scheduled piece of equipment shown, and are utility connections drawn?
- Fire protection and life safety — is coverage complete on the current footprint, including any additions or renovations?
- Spec sections vs. drawing callouts — when a note or callout references a spec section, does the spec section actually specify what the drawing shows?
Missing-coordination gaps are the most insidious version of this: a fire-alarm device layout that was never extended onto the renovation addition; an ADA clearance that no sheet actually accounts for. These aren’t conflicts between two contradictory sheets — they’re an absence, and absence is harder to see.
Step 3: Do a constructability pass
A constructability review asks whether what the drawings show can actually be built as drawn — in the sequence, with the clearances, at the dimensions specified. It’s distinct from a coordination check (which asks whether the disciplines agree) and a code check (which asks whether the documents comply with standards). Constructability asks whether a real crew can execute the work as documented.
Common constructability issues that generate RFIs:
- Sequence conflicts: a detail that requires work to happen in an order the framing sequence won’t allow
- Tolerance gaps: adjacent assemblies drawn at exact dimensions with no tolerance or connection detail, leaving the field to figure out the interface
- Insufficient detail at transitions: the plan shows what each system does in isolation; it doesn’t show how the transition is executed where systems meet
- Equipment access: large mechanical equipment shown in a space with no clear rigging or installation path
Constructability issues rarely appear in a single sheet. They show up when you read the structural sequence against the MEP rough-in against the architectural finishes. A revisión de constructibilidad run at design development — before the documents are locked — is the highest-leverage point to catch them. A post-CD review still catches them before bid, which is still worth doing.
Step 4: Run automated conflict detection on the full 2D set
Manual review is good. The limitation is comprehensiveness: a senior reviewer working through a 600-sheet set on a deadline will catch the conflicts they know to look for, in the areas they focus on. They will miss things.
The case for automated review isn’t that it’s more insightful than an experienced reviewer — it’s that it’s more exhaustive. It reads every sheet against every other sheet, every drawing callout against every spec section, without fatigue and without the time pressure that compresses a real review.
Flikt runs Plan de revisión de construcción con IA directly on 2D PDF drawing sets and surfaces conflicts across four categories: spatial clashes, spec-vs-drawing mismatches, missing coordination, and code or clearance issues. On one real project, the review flagged more than 40 conflicts that eventually traced to change orders — conflicts distributed across architectural, structural, and MEP sheets, none visible from a single discipline’s drawings.
The benchmark Flikt tracks across projects is conflictos por cada 100 hojas — a normalized drawing-quality metric that lets you compare document quality across design teams, project types, and revision cycles, and measure whether a redline cycle actually improved the set. See conflictos por cada 100 hojas for what that benchmark looks like in practice.
This step works best as a complement to Steps 1–3, not a replacement for them. Automated review catches what’s distributed and hard to see comprehensively. Manual coordination review catches what requires judgment about the building’s intent. Together, they close the gaps the other misses.
The metric: what to measure before and after
RFI volume per project is a lagging indicator — you only know it after the fact. Drawing quality measured before construction is a leading indicator you can act on.
Two metrics worth tracking at plan review:
Conflicts per 100 sheets — how many real, actionable conflicts the set contains, normalized by sheet count. Lower is better; trends over revision cycles tell you whether the coordination process is working.
RFIs per 100 sheets (post-construction) — how many RFIs the project generated during construction, normalized the same way. Over time, projects that go through rigorous pre-issue review should show meaningfully lower post-construction RFI rates.
Industry studies estimate that errors caught during construction cost roughly diez veces más para arreglar than the same errors caught during plan review. The goal of pre-construction QA is to shift that ratio — catch as many issues as possible on paper, where they cost a fraction of what they’ll cost in the field.
A structured approach to this is in the pre-construction QA checklist, which organizes the full review sequence — from document receipt through final coordination sign-off — into a trackable format.
Putting it together
The four steps above aren’t a new discovery. Every experienced superintendent knows that RFIs cluster around coordination gaps. What changes when you run them systematically, before issue, on the full document set:
- You convert future RFIs into present-day redlines, when fixes are still cheap.
- You give the design team a specific, documented conflict list to resolve before bid — not a vague “this set needs work.”
- You reduce the knowledge gap between what the GC prices and what they’ll actually build.
- You generate a pre-construction baseline for drawing quality that makes every subsequent project comparison meaningful.
The economics are documented on la página de evidencia. If you want to quantify what catching conflicts earlier is worth on your specific project, the ROI calculator runs the numbers.
For architects carrying coordination liability, what Flikt’s review means for your practice covers how the workflow integrates with standard deliverable schedules and what the conflict documentation looks like.
Ready to see what’s hiding in your drawing set before the first RFI hits? Contact us to review a project — no BIM, no modeling software, just your PDFs.
¿Listo para detectar conflictos a tiempo?
Sube tu conjunto de planos y obtén detección de conflictos con IA, no se requiere BIM.