What Is Drawing QA/QC? A Plain-English Guide for AEC Teams

Short answer: Drawing QA/QC is the discipline of making sure a set of architectural drawings is correct, coordinated, and safe to issue before anyone builds from it. QA (quality assurance) is the system — the standards, templates, and review process that prevent errors. QC (quality control) is the act of checking a specific drawing set against those standards and catching the errors that slipped through. Archi Check is desktop QC software that turns that checking into a tracked loop, so every error found becomes an item with an owner and a status rather than a note that gets lost.

Almost every architect, engineer, and coordinator does some form of drawing QA/QC, but very few would describe it the same way. For one person it is a senior reviewer red-penning a set the night before issue. For another it is a coordination meeting where disciplines compare layouts. For a third it is a checklist taped to the wall. All of these are part of the picture, and none of them is the whole thing.

This guide sets out what drawing QA/QC actually means, where QA and QC differ, what a proper drawing set review covers, and what errors cost when the process fails. It also makes one point that gets lost in most explanations: finding an error is the easy half. The hard half — the half that protects a project — is closing it.

What drawing QA/QC means

Drawing QA/QC is the combination of two related but distinct activities. Quality assurance is everything you do to stop errors from being created in the first place: drawing standards, layer and naming conventions, title-block templates, a defined review process, and trained people who know how to use them. Quality control is what you do to catch the errors that get created anyway: the actual checking of a specific drawing or set against the standard, sheet by sheet, before it goes out the door.

The shorthand most teams use is that QA is about the process and QC is about the product. QA asks, "is our way of producing drawings sound?" QC asks, "is this drawing correct?" You need both. A practice with excellent standards but no checking still issues mistakes, because standards do not enforce themselves. A practice that checks heavily but has no standards burns enormous effort catching the same avoidable errors on every job.

In drawing work the two are tightly coupled, which is why the field is almost always written as one term — QA/QC. The standards (QA) define what "correct" means, and the check (QC) measures the drawing against it. When people say "we need to QC the set," they usually mean the control half, but the assurance half is what made the check meaningful.

QA vs QC in one line

If you remember one distinction, make it this: QA is preventive and built into the system; QC is detective and applied to the output. A naming convention is QA. Opening the set and confirming the sheets follow that convention is QC. The two are not in competition — QC is how you find out whether your QA is working. For a fuller treatment of the difference on construction documents specifically, see QA vs QC in construction drawings.

Why drawing quality control matters

Drawings are instructions. A construction drawing set tells contractors, fabricators, and trades exactly what to build and where. Unlike a model that a coordinator interrogates interactively, an issued drawing is acted on directly: someone reads a dimension, cuts a steel member to it, and pours concrete around it. If that dimension is wrong, the error is now physical, and unwinding it costs far more than the few seconds it would have taken to catch on a screen.

This is the core reason drawing QC matters more than it appears to. The cost of an error does not stay constant as it moves downstream — it multiplies. An inconsistency caught during checking costs a corrected line. The same inconsistency caught on site costs a stop, an RFI, a redesign, possibly demolition and rework, and the schedule slip that ripples through every trade waiting behind it. The earlier the catch, the cheaper the fix, and QC is the last cheap chance before the drawing becomes someone else's expensive problem.

There is a second, quieter reason. A checked set is a defensible set. When a defect is questioned months later, the first thing anyone looks for is the record of who checked the drawing, what they checked, and when. A QC process that produces that record protects the practice as much as it protects the building. A process that lives in red pen on a printout that has since been recycled protects no one.

The cost of errors, concretely

It helps to make the multiplier tangible. The same error has a very different price depending on the stage it is caught.

• Caught in checking. A coordinator spots a door that is dimensioned 900 on the plan and 1000 on the schedule. Cost: a corrected number and a re-check. Minutes.
• Caught at tender. A contractor's estimator queries the same clash in an RFI. Cost: a written clarification, an addendum, and a small delay to pricing. Hours to days.
• Caught on site. The frame is set to one dimension and the schedule procured to the other. Cost: a stop, a variation, possibly a reorder and rework, plus the trades held up behind it. Days to weeks, and real money.
• Caught after handover. The discrepancy surfaces in a defect or a dispute. Cost: investigation, possible remediation, and reputational exposure. Open-ended.

None of these errors is exotic. They are the ordinary inconsistencies that creep into any set produced by several people under deadline. QC exists because the cheapest column in that list is the one you control.

What a drawing set review covers

A drawing set review — the QC pass — is more than proofreading. A thorough check looks at the drawing on several levels at once, from the individual sheet up to the whole coordinated set. The exact scope varies by practice and project stage, but a competent review covers the same broad categories every time.

Completeness

Is everything that should be in the set actually present? This is the drawing-list check: every sheet referenced exists, every detail callout points to a detail that has been drawn, every schedule is populated, and nothing is marked "to follow" that should now be resolved. Missing information is the most common and most expensive QC failure, because the gap is invisible on the sheets that are present.

Correctness

Is the information on each sheet right in itself? Dimensions add up, levels are consistent, annotations match what is drawn, references point to the correct sheet, and the title block carries the right revision and status. This is the sheet-level check, and it catches the internal contradictions that survive because no single person drew the whole sheet.

Coordination

Do the disciplines agree with each other? The architectural plan, the structural layout, and the services routes have to occupy the same building without colliding. Cross-discipline coordination is where the most serious errors hide, because each discipline's drawing can be internally perfect while being wrong about the others. Comparing sheets side by side is the heart of this check.

Compliance and standards

Does the set follow the office standard and the relevant codes? Layer and naming conventions, drawing scales, line weights, the agreed status codes, and the regulatory requirements that apply to the project. This is where QC measures the product against the QA system, and where a good standard makes the check fast.

For a practical walkthrough of how to run this review in order rather than at random, see how to check construction drawings.

QA and QC across the drawing lifecycle

The two halves do not happen at one moment; they run through the whole life of a set. Mapping them onto the stages of a project shows how they interlock — assurance setting the ground rules, control verifying the output at each gate.

The pattern repeats: QA defines what good looks like at each stage, and QC measures the work against it. The pre-issue review is the one most people picture when they hear "QC," but it is only the most visible gate of several. The full sequence of those gates is set out in the drawing QA/QC process in 4 stages.

Who does the checking

QC is not a solo activity, and the most important rule in it is independence: the person who checks a drawing should not be the person who drew it. An author cannot reliably see their own errors, because the same assumption that created the mistake hides it from them. Separating the roles is what makes a check worth running.

In practice the work divides across a few clear responsibilities. A checker reviews the drawing and raises items but does not change the drawing. A corrector — usually the original author or someone in the production team — makes the fixes. An approver or project lead signs off that the set is ready to issue. A manager owns the standard and the process. Keeping these roles distinct is what prevents an error from being marked fixed by the same person who introduced it, which is how unresolved problems slip through. The full split of duties is covered in who does what in drawing QC.

Finding errors is not the same as closing them

This is the point that most explanations of QA/QC skip, and it is the one that determines whether the process actually protects a project. A red mark on a drawing is not a fix. It is a question — "is this right?" — that someone now has to answer, act on, and confirm. The check that finds the error is the cheap, satisfying half. The expensive, unglamorous half is making sure every one of those marks is resolved and verified before the set is issued.

This is exactly where most real QC processes leak. A reviewer covers a set in comments, the comments go back to the team, some get fixed, some get missed, a few get fixed wrongly, and at issue nobody can say with confidence which is which. The set goes out with open items buried in it, not because the check was bad, but because there was no system tracking each item from raised to closed. A defect tracked on a printout that has been superseded is, for practical purposes, a defect that was never found.

A real QC loop closes that gap by treating every error as an item with a lifecycle, not a mark. Each comment is raised, assigned to an owner, corrected, verified by someone other than the corrector, and only then closed — with a record of what changed. The check starts the loop; closure ends it; and nothing falls out in between because every item has a status you can see. The difference between a set that is "checked" and a set that is genuinely safe to issue is whether that loop was closed on every item, not whether the errors were found.

Where Archi Check fits

Archi Check is purpose-built QC software for architectural drawing sets, and it exists to make the closing half of QA/QC as reliable as the finding half. It is a standalone desktop application, not a Revit plugin — you load your drawing set, review it, and run the whole quality-control loop in one place. That loop has four stages: Check, Correct, Verify, Close.

In the Check stage a reviewer marks up the drawing. Every markup becomes a tracked comment, not just ink — it carries the sheet and revision it sits on, the reviewer's name, and a status. In the Correct stage each comment is assigned to an owner who makes the fix. In the Verify stage someone other than the corrector confirms the fix is right, which is the independence rule built into the workflow. Only then, in the Close stage, does the item close, leaving a full audit trail of what was raised, who acted on it, and what changed. A REVISE comment cannot quietly disappear, because it stays open until it is verified and closed.

A few design choices make this practical on real projects. The two-screen workflow puts the register of tracked comments on one monitor and the full-screen drawing markup on the other, so you can see every open item beside the sheet it belongs to. Storage is folder-first and local — your QC data lives in your own project folders rather than a vendor cloud, so the audit trail belongs to you. Archi Check runs on Windows today, with macOS coming soon, and there is a 14-day free trial so you can run a real set through the loop before committing.

FAQ

What is drawing QA/QC?

Drawing QA/QC is the discipline of making sure a set of architectural drawings is correct, coordinated, and safe to issue. QA (quality assurance) is the system that prevents errors — standards, templates, conventions, and a defined review process. QC (quality control) is the act of checking a specific drawing set against those standards and catching the errors that got through. The two work together: QA defines what correct means, and QC measures the drawing against it.

What is the difference between QA and QC in drawings?

QA is preventive and applies to the process; QC is detective and applies to the product. QA is everything you set up to stop errors being created — naming conventions, title-block templates, trained reviewers, a defined workflow. QC is the actual checking of a drawing or set against that standard before it is issued. A useful test: a naming convention is QA, and opening the set to confirm the sheets follow it is QC.

Why does drawing quality control matter?

Because drawings are instructions that get built directly, and the cost of an error multiplies the further downstream it is caught. An inconsistency found during checking costs a corrected line; the same one found on site costs a stop, an RFI, rework, and a schedule slip. QC is the last cheap chance to catch the error before it becomes physical, and a checked set is also the record that protects the practice if a defect is questioned later.

What does a drawing set review cover?

A thorough review covers four broad areas: completeness (every referenced sheet, detail, and schedule is present), correctness (dimensions, levels, annotations, and references are right on each sheet), coordination (the disciplines agree and do not collide), and compliance (the set follows the office standard and relevant codes). It is more than proofreading — it checks the drawing from the individual sheet up to the whole coordinated set.

Is finding errors the same as fixing them?

No, and the difference is where most QC processes fail. Finding an error is the cheap half; a red mark is a question, not a fix. The work that actually protects a project is making sure every raised item is corrected, verified by someone other than the corrector, and closed before the set is issued. Without a system that tracks each item from raised to closed, found errors leak back into the issued set unresolved.

Close the loop, not just the check

Understanding QA/QC is the easy part; the hard part is making sure every error you find is actually resolved before you issue. If your checks still end as marks on a printout, see how Archi Check turns each one into a tracked item with an owner, a status, and an audit trail: Try Archi Check free for 14 days.

Related guides

Keep building out your drawing QC process with Archi Check and these related guides:

• QA vs QC in construction drawings
• The drawing QA/QC process in 4 stages
• How to check construction drawings
• Who does what in drawing QC

Archi Check is an independent product by Archi for architectural drawing QA/QC. Product names and trademarks referenced belong to their respective owners and are used for identification only.