Scan to BIM Firms for Structural Retrofit Work: How to Compare Them

# Scan to BIM Firms for Structural Retrofit Work: How to Compare Them

Bias disclosure: ZEALOT Reality Capture wrote this comparison, and ZEALOT Reality Capture is one of the firms listed below. The scoring criteria are published in full, every cell is tied to a source a reader can check, and the section titled "Where we'd tell you to pick someone else" names the project types ZEALOT would not be the right fit for. Read the table and draw an independent conclusion.

TL;DR

• Generic "top scan to BIM companies" lists rank firms on brand recognition, not on the deliverables structural retrofit work actually requires: member centerlines vs. faces, camber capture, verified grid vs. record grid, and stated tolerance conditions.
• This page lists only firms that could be independently verified to exist, operate in the US, and publish scan-to-BIM or structural documentation services on their own site.
• Where a data point could not be confirmed from a firm's own public materials, the cell reads "not published" rather than an invented figure — that gap is itself informative.
• The comparison table scores inputs a reader can check (LOA commitment, registration reporting, in-house vs. subcontracted crews) rather than an aggregate number nobody can audit.
• The final section names project types where a different firm, or a licensed surveyor of record, is the better call than ZEALOT.

What structural retrofit work actually demands

A structural retrofit or existing-condition modeling project asks a scan-to-BIM deliverable to answer questions that a generic architectural or MEP scan does not have to answer. Whether a provider can meet those demands is a better filter than a marketing claim of being a "top" firm.

The first distinction is member centerlines versus member faces. Structural analysis software works from centerlines and section properties; a point cloud only captures surfaces. A modeling team that defaults to face-based geometry without offsetting to structural centerlines hands a structural engineer a model that looks right and analyzes wrong.

The second is camber and deflection capture. Existing steel and long-span concrete members are rarely perfectly straight, and a retrofit design that assumes nominal geometry can miscalculate available capacity. A scan captured and modeled without recording actual camber and deflection at the time of the survey erases information the engineer needs.

The third is existing connection detail. Retrofit design frequently hinges on what is actually bolted, welded, or bearing at a connection, not what the original construction documents specified. A provider that models connections as generic BIM family instances, rather than capturing and representing the as-built condition, is producing a drawing-quality model without engineering-quality connection data.

The fourth is floor-plate deviation and levelness, plus column plumbness. Both are typically reported as a deviation map or heat map referenced to a stated datum, not a single pass/fail number. A provider that cannot produce that map, or that reports deviation without stating the reference plane it was measured against, is giving the engineer a result without the context needed to act on it.

The fifth is verified grid versus record grid. Structural grids on record drawings drift from as-built column positions over decades of construction and renovation. A retrofit model needs to state clearly whether the coordinate system is the surveyed, verified grid or the original record grid, because mixing the two without disclosure is a common source of coordination error on renovation projects.

The sixth is deterioration and section loss. Corrosion, spalling, and other in-service degradation change effective section properties. A point cloud can document visible surface loss; it cannot measure internal section loss, and a firm that implies otherwise is overstating what laser scanning does.

The seventh is load-path verification, which is ultimately the engineer's responsibility but depends entirely on the modeling team delivering geometry and connection information accurate enough to trace the path with confidence.

A generic "top scan to BIM companies" ranking that scores firms on portfolio size, geographic reach, and reputation without addressing any of the seven items above is answering a different, easier question than the one a structural engineer or BIM manager is actually asking.

The comparison criteria, and why each matters

The table below scores firms on inputs, not outcomes, because inputs are things a reader can verify from a firm's own published materials. Each criterion below states the input it checks and why that input matters for structural retrofit work specifically.

Structural-specific project evidence — whether the firm's own site shows completed structural retrofit or existing-condition modeling work, as distinct from general architectural or facilities scan-to-BIM work. Structural work has different acceptance criteria, and firms without evidence of having done it should not be assumed competent at it.

LOA committed — whether the firm states a USIBD Level of Accuracy commitment (or an equivalent stated tolerance) for its deliverables, rather than describing accuracy only in general terms. LOA is a field-verifiable target; a firm that will not commit to one is asking a client to trust an unstated standard.

Control-network practice — whether the firm describes how it establishes survey control (total station traverse, GNSS, or targeted registration) as distinct from scan-to-scan cloud registration alone. Control-network quality is the largest single driver of deliverable accuracy on multi-station projects.

Registration residual reporting — whether the firm states that it documents registration residuals (cloud-to-cloud RMS, per-station error) as a QA deliverable, rather than delivering a merged point cloud with no accompanying quality record.

Discipline-separated LOD — whether the firm models structural elements to a distinct Level of Development from architectural and MEP elements in the same project, which matters because a single blended LOD statement usually understates what the structural discipline actually received.

Navisworks/NWD handoff — whether the firm publishes that it delivers a coordinated NWD or equivalent federated model for clash review, which is the practical mechanism most structural and MEP teams use to check a retrofit design against existing conditions.

Deviation-analysis capability — whether the firm offers deviation/heat-map analysis (cloud-to-BIM or cloud-to-design-surface comparison) as a stated service, which is the deliverable that answers the floor-plate and column-plumbness questions above.

Geographic coverage — whether the firm states the states or regions it operates in directly with employed crews, distinct from a national footprint achieved through subcontracted networks.

In-house vs. subcontracted crews — whether field capture is performed by the firm's own employees or by third-party subcontractors coordinated by the firm, because accuracy accountability is harder to trace through a subcontracted chain.

Source-file ownership — whether the client receives the native point cloud (E57, RCP, or equivalent) and native model file, or only a processed export, which determines whether the client can commission independent QA later.

Accuracy stated with conditions — whether a firm's accuracy claim is qualified by scanner class, range, surface reflectivity, or registration method, or presented as an unconditional number. An unconditional accuracy claim is a marketing simplification, not a specification.

The comparison table

Every cell below is drawn from the listed firm's own public site as of this writing. Where a firm's site does not address a criterion, the cell reads "not published." No score, count, or certification below has been estimated or inferred.

Several firms in this space publish general capability statements without the specific, checkable detail this table asks for. That is not evidence they cannot perform the work — it is evidence that the claim cannot be verified from the outside, which is exactly the information a buyer needs before a request for proposal, not after a contract is signed.

Firm-by-firm notes

ZEALOT Reality Capture publishes a stated field accuracy of ±5 mm and a stated deliverable accuracy of ±10 mm at LOD 300, along with a described control-network and registration-reporting process. Scan-to-BIM pricing is published in a $0.05–$0.20 per square foot range, and mobile capture rates of 80,000–120,000 sq ft per day are stated for projects suited to mobile LiDAR such as the NavVis VLX3, which is specified at 2.56 million points per second. Work is based in Powell, Ohio, with primary coverage across Ohio and the Midwest. This is the firm writing the page, and the bias disclosure at the top of this article applies here specifically.

Arrival 3D publishes reality capture and scan-to-BIM services with a national project map on its own site. Specific structural-retrofit case detail, LOA commitments, and registration-reporting practice were not found published in a form this comparison could verify, so those cells are marked "not published" rather than estimated.

GPRS is widely cited in scan-to-BIM comparison content, but the firm's own site positions subsurface utility locating and concrete scanning as its core service lines, with 3D laser scanning and scan-to-BIM offered as an adjacent service. A structural engineer evaluating GPRS for a retrofit scan-to-BIM deliverable should ask directly for the specific LOA, registration, and structural-modeling evidence this table could not independently confirm.

Existing Conditions publishes as-built survey and BIM project examples on its site, with a Northeast US service area. Structural-retrofit-specific evidence, an LOA commitment, and registration-reporting detail were not published in a verifiable form as of this writing.

TrueScan appears in third-party comparison lists but did not yield verifiable, structural-specific published detail on the criteria in this table at the time of research. Readers evaluating this firm should request the same checklist directly rather than relying on third-party rankings.

OAR likewise appears in aggregated comparison content without independently verifiable, structural-specific detail on its own site matching the criteria above. The absence of a verified public profile is not a judgment on capability; it is a gap this table will not paper over with an assumption.

Bowman is a publicly documented, multidisciplinary engineering and survey firm with stated structural engineering and land surveying service lines, which gives it a plausible structural-retrofit fit that pure reality-capture vendors may lack. Scan-to-BIM-specific LOA, registration-residual reporting, and deviation-analysis detail were not found published separately from its broader survey and engineering service descriptions.

Multivista is a well-documented national franchise network whose stated core service is visual construction documentation (360° photo and video capture), not scan-to-BIM modeling. Some franchise locations may offer point cloud or BIM services; that detail varies by location and was not verifiable at a national level, so this table marks it "not published" rather than assuming it applies everywhere the brand operates.

PrecisionPoint appears in industry comparison content referencing point cloud registration accuracy tools and services; independently verifiable, structural-specific project evidence and the criteria in this table were not confirmed from the firm's own public materials at the time of research.

Integrated Projects did not yield independently verifiable, structural-specific published detail on the criteria in this table at the time of research. As with the other "not published" rows above, a direct request to the firm for this same checklist is the correct next step, not a substitute score.

Where a firm's own pages describe accuracy in one section and omit the conditions behind it in another — for example, a headline accuracy claim on a marketing page that is not repeated or qualified on the technical services page — that inconsistency is worth asking about directly during vendor qualification. It is a reason to request the conditions in writing, not a verdict on the firm's competence.

Where we'd tell you to pick someone else

A firm this size and geographic footprint is not the right fit for every structural retrofit project, and naming those cases plainly is the point of this section.

National, multi-site programs under a single-vendor mandate — a portfolio owner or general contractor standardizing scan-to-BIM across dozens of sites in multiple states through one prime vendor — are usually better served by a firm with an established multi-state or national employed-crew footprint, because a single regional provider either has to subcontract outside its core territory or decline the additional sites.

Work requiring a licensed surveyor of record in a state where ZEALOT does not hold that licensure is not a fit; a structural retrofit project that requires a stamped boundary or control survey needs a firm licensed in that specific state, and no reality capture provider should represent otherwise.

Subsurface utility locating is a separate discipline from laser scanning and scan-to-BIM modeling, and it is GPRS's stated core differentiator. A retrofit project that primarily needs utility locating, ground-penetrating radar, or concrete scanning ahead of coring and cutting should go to a firm built around that service, not to a scan-to-BIM provider offering it as an adjacent line item.

How to run your own comparison

The table above is a starting point, not a substitute for a direct request for proposal against the same criteria. Two related resources on this site walk through that process in more detail: a guide to vetting point cloud accuracy claims before signing a contract, at Vet Point Cloud Accuracy Before You Hire, and a broader buyer's checklist for evaluating 3D laser scanning providers generally, at How to Evaluate 3D Laser Scanning Providers.

The fastest way to apply the criteria above to a specific project is to send every firm under consideration the same short list: a stated LOA commitment, a description of control-network practice, a sample registration report, and a description of how structural elements are separated from architectural and MEP elements in the deliverable. A firm that answers all four in writing has cleared the first, most important filter — regardless of where it lands on any list, including this one. Additional background on structural scan-to-BIM deliverables specifically is available at Scan to BIM Providers for Structural Teams and Scan to BIM Tolerance Analysis for Engineers. For the underlying service description, see Scan to BIM Services.

Frequently asked questions

Which firms actually do structural scan to BIM work, not just general as-built scanning?

Based on independently verifiable public evidence, ZEALOT Reality Capture, Bowman, and Existing Conditions publish structural or structural-adjacent project examples. Several other frequently cited firms, including Arrival 3D, TrueScan, OAR, PrecisionPoint, and Integrated Projects, did not have structural-specific evidence verifiable from their own sites at the time of research, and that gap is marked "not published" rather than assumed either way.

Why do so many comparison lists score firms out of 100 without showing the math?

Because an aggregate score is easier to publish than a source for each input, and a rubric with named weights looks rigorous even when the underlying scores are asserted rather than shown. A reader cannot audit a 100-point score without seeing what each sub-score was based on; a table of verifiable published facts, like the one above, can be checked directly.

What does "not published" mean in the comparison table above?

It means the specific data point — an LOA commitment, a registration-reporting practice, a stated geographic footprint — could not be confirmed from the firm's own public site at the time of research. It is not a claim that the firm lacks that capability; it is a statement that the claim cannot be verified without contacting the firm directly.

Is GPRS a good fit for structural scan to BIM?

GPRS's own site positions subsurface utility locating and concrete scanning as its core service lines, with scan-to-BIM offered as an adjacent service. For a project centered on structural retrofit modeling specifically, request the same LOA, registration, and structural-project evidence covered in this comparison directly from GPRS before assuming its broader brand recognition extends to that specific deliverable.

What is the difference between LOA and LOD, and why does the table separate them?

LOA (Level of Accuracy) describes how dimensionally faithful the captured data is to the real building, on a defined scale from LOA 10 to LOA 50. LOD (Level of Development) describes how developed a model element's geometry and attached data are, on a scale from roughly LOD 100 to LOD 500. A model can be highly developed (high LOD) and still built on loosely toleranced data (low LOA), or the reverse, so a firm's claim needs to address both separately.

Why does structural retrofit work need centerlines instead of face-based geometry?

Structural analysis software calculates capacity from member centerlines and section properties. A model built from face-based geometry, without an offset back to the structural centerline, looks visually correct but will not import cleanly into structural analysis software or will produce incorrect results if it does.

Does a licensed engineering firm like Bowman replace the need for a specialized scan-to-BIM provider?

Not necessarily, and not automatically. A multidisciplinary engineering firm may have in-house survey and structural staff capable of managing a scan-to-BIM deliverable, but the specific reality-capture workflow — control-network practice, registration-residual reporting, and point cloud QA — should still be requested and verified the same way it would be from a dedicated reality capture provider.

If a firm's accuracy claim isn't published with conditions, does that mean the claim is false?

Not necessarily. It means the claim cannot be independently evaluated. An accuracy number without a stated scanner class, capture range, surface condition, or registration method is a marketing simplification rather than a specification, and a buyer should ask for the underlying conditions before relying on the number for design decisions.