Scan-to-BIM Workflow: From Field Scan to Revit Model
The scan-to-BIM workflow moves a building from a field laser scan to a usable Revit model through a fixed set of stages: intake and scope, field capture, point cloud registration, Revit setup, modeling, internal QA, client review, and delivery. The model that comes out the other end is only as good as the scope decisions made before the scanner reaches the site. Get the scope right at intake and the process is predictable. Skip it, and you can end up with a model that is technically clean but wrong for what the buyer needs to do with it.
- Intake and scope
- Site prep and field capture
- Registration
- Revit setup and modeling
- QA, client review, and delivery
Stage 1: Intake and scope
The workflow starts before anyone unpacks a scanner. Intake defines the deliverable and is the single biggest predictor of how smoothly the project runs. The questions that matter are concrete: which areas are in scope, interior or exterior or both, what Revit version the receiving team needs, which model categories are included (shell only, or structure and MEP too), what LOD or use case is expected, and whether 2D sheets or PDFs are part of the deliverable.
LOD is the lever most worth pinning down. LOD 200 captures approximate size and location and reads as a coordination-grade massing model; LOD 300 carries accurate geometry for design and renovation documentation; LOD 350 adds connection and interface detail. Each step up adds modeling hours, so naming the target up front keeps the quote and the result aligned. Accuracy is a separate axis, governed by the USIBD Level of Accuracy (LOA), which describes how closely the model must match measured reality. Good intake names both. Our scan-to-BIM LOD guide breaks down where each level earns its cost.
See scan-to-BIM services for the full service scope.
Stage 2: Site prep and field capture
Field capture is WeAre Capture’s core strength. We scan with a survey-grade terrestrial laser scanner, and a full scan day on site runs up to roughly 12 hours depending on building size and access. The scanner records only what it can see from each setup, so coverage is a planning problem. Locked rooms, blocked walls, reflective glass, active work areas, and high ceilings without access all turn into gaps in the data, and gaps become assumptions later in the model.
Site prep is not about making the building spotless. It is about removing preventable blind spots: confirm access and parking, unlock rooms in scope, brief on-site security, flag restricted or hazardous zones, and confirm roof or exterior access if the scope needs it. A scan planned for broad floor plans does not need the same setup density as one feeding detailed ceiling or facade modeling, so the field plan follows the model scope rather than chasing a large dataset. Our laser scanning site prep checklist covers this, and 3D laser scanning services shows how field capture is structured.
Stage 3: Registration
Registration aligns the individual scan positions into one coordinated point cloud, and everything downstream is built on it, because the Revit model is traced from registered data. If alignment drifts, modeling slows and measurements get less trustworthy. Review here means checking position-to-position alignment and overlap, confirming the scoped area is fully covered, and catching gaps before they reach a modeler. The deliverable is the registered cloud in an agreed format, typically E57 as a vendor-neutral exchange file, or RCP and RCS for direct use in Autodesk tools. When a client supplies their own cloud, the team reviews it first: a file that opens is not the same as a file that is production-ready. See point cloud registration explained for the deeper version.
When capture and modeling happen as separate phases
Scanning and modeling don’t have to be scoped as a single step, and when they run as two connected phases, the quality of the handoff between them is what determines how smoothly the project goes. That split works fine when the handoff is clean and falls apart when it is not. Moving from a registered cloud to a modeled file needs the cloud and its format, the scan date, the site scope, known missing areas, coordinate and origin assumptions, any existing drawings, and the requested model scope in writing. Field photos help resolve ambiguous conditions without a second site visit. Without that package, paid hours go into rediscovering issues that were already known at the time of capture. The handoff document is cheap insurance. See 3D laser scanning services (registration included) for how registered data is prepared for downstream use.
Stage 4: Revit setup and modeling
Before any geometry gets traced, the Revit file is set up: correct version, units, levels, grids where required, the linked point cloud, a project template, file naming, and a defined model origin or shared coordinate system. This is unglamorous, but it is what makes the file usable the moment the architect opens it. A client BIM standard belongs here, before modeling starts, not as a rework request afterward.
Modeling then traces Revit elements from the cloud per the agreed scope: walls, floors, ceilings, roofs, structure, doors and windows, stairs and railings, and selected MEP where it is in scope. This is where LOD decisions become visible work. Should an out-of-plumb wall be modeled as a clean plane or as observed? Is ceiling grid in scope? Are generic families acceptable, or does the project need custom ones? Each answer moves cost and schedule, which is why they are settled at intake rather than discovered mid-production. Our point cloud to Revit workflow and the LOD 200 vs LOD 300 comparison lay out those tradeoffs.
Stage 5: QA, client review, and delivery
Internal QA reviews the model against the scan, the scope, and the expected deliverable, not against a spellcheck. It confirms that modeled categories match scope, that major dimensions and alignment track the point cloud, that levels and views are set up correctly, that the file opens cleanly in the target Revit version, and that exclusions are documented. The goal is not to pretend every existing condition is fully known, but to deliver a model that matches the agreed scope and clearly flags its limits.
Client review is where the receiving team confirms the model supports its intended use. Specific comments move the project forward: “add ceiling grid in rooms 201 to 210 where visible,” “confirm whether these columns are in scope,” “we need Revit 2024, not 2025.” Vague comments like “make it more detailed” usually signal scope that was never defined, and if scope changes during review, schedule and price change with it.
Delivery typically bundles the RVT file, the linked point cloud, the E57 or RCP/RCS data, any PDF sheets and DWG exports, QA notes, and a written list of exclusions and assumptions. IFC is available when the receiving team works outside Revit. The handoff should state plainly what is included and what is not, which matters most when the model passes to another architect, engineer, owner, or contractor.
Where the workflow breaks
Scan-to-BIM work struggles for a small, repeatable set of reasons.
| Problem | What it causes |
|---|---|
| Scope is vague | Model includes too much, too little, or the wrong categories |
| LOD and LOA undefined | Review comments become subjective and rework follows |
| Point cloud has gaps | Modeling falls back on assumptions or exclusions |
| Wrong Revit version | The receiving team cannot use the file cleanly |
| Sheets are assumed | The client expects drawings that were never scoped |
| Review is skipped | Issues surface too late to fix cheaply |
The fix in every case is not more buzzwords. It is a clearer scope agreed before production starts.
FAQ
What is scan-to-BIM?
Scan-to-BIM turns a 3D laser scan of an existing building into a Building Information Model, usually in Revit. Field scanning captures conditions as a point cloud, registration aligns the data, and modelers trace accurate building elements from it at an agreed level of detail.
How does 3D laser scanning work?
A survey-grade terrestrial scanner sits at a fixed position and sweeps a laser across the space, recording anywhere from several hundred thousand to a couple million points per second depending on the instrument, to build a 3D point cloud. The scanner moves through multiple setups so overlapping scans cover the whole site, then those scans are registered into one coordinated cloud.
How do I import a point cloud into Revit?
Revit reads Autodesk RCP and RCS files. An E57 or other raw scan is first indexed through Autodesk ReCap to produce the RCP project file, which is then linked into Revit through Insert, Point Cloud. Linking against the correct shared coordinates keeps the cloud aligned. See our guide to importing a point cloud into Revit for the full steps.
How much does scan-to-BIM cost?
No single dollar-per-square-foot figure applies, because cost follows element density and the target LOD set at intake, not floor area. Field scanning is usually billed by day or hour, while modeling is priced by area and detail level. For the factors that drive a quote, see our scan-to-BIM cost guide.
Start your scan-to-BIM project
The best projects begin with one alignment question: what will the receiving team do with the model after delivery? That answer drives every stage above. WeAre Capture handles field capture with a survey-grade terrestrial scanner and supports the full path through to Revit delivery, and we can also work from client-provided point cloud data when the scan quality holds up. Request a quote with your building type, area, and target LOD, and we will scope it against the workflow on this page.
