Building digital twin: how to choose among 4 technologies (an honest comparison)
By Olivier Kerremans ·
Digitising a building in 3D can mean four very different things depending on the chosen technology. Each has its strengths, limits, costs and timelines. This article honestly compares the four options available in 2026, without trying to push a single answer, and without naming brands. Our goal is that you leave with a clear decision framework for your project.
For a broader overview of the breakthrough technology that is 3D Gaussian Splatting, see our complete 3DGS guide. To understand the specific difference between 3DGS and 360°, see our detailed comparison.
What is a building digital twin, really?
The term “digital twin” is used for everything. Here is an operational definition.
A digital twin is a 3D representation of a real building, faithful enough that decisions can be made about it without travelling there. It can be more or less rich in data (visual, metric, domain), but its function is always to replace or augment a physical visit.
A useful distinction: a virtual tour (360° or other) shows a space. A digital twin lets you act on it: measure, compare, archive, share with different stakeholders.
By this definition, the 360° virtual tour is actually borderline; it is more a visual experience than a true digital twin. But because it is often compared with the other three technologies, we keep it in this comparison.
The 4 technologies available today
1. Virtual tour with 360° panoramic images
Principle, panoramic photographs taken from fixed points with a 360° camera, assembled into a point-to-point navigable path.
Fidelity, excellent on captured angles, nil between fixed points (the camera did not see).
Navigation, by teleportation; you jump from one point to another, without free movement through the space.
Maturity, a well-established technology, democratised for the public over ten years.
2. Point-cloud laser scanning
Principle, a laser scanner emits millions of points per second and measures their return to reconstruct a metrically precise 3D point cloud.
Fidelity, millimetre geometry. Variable texture depending on the quality of the associated colour photos.
Navigation, possible but requires a specialised viewer. The rendering is not photorealistic; it is a cloud of coloured points.
Maturity, a well-established technology in AEC for fifteen years, the standard for scan-to-BIM.
3. Photogrammetry / classic 3D mesh
Principle, from photos taken at several angles, software reconstructs a 3D surface (a polygonal mesh) with applied textures.
Fidelity, good on objects with uniform, well-lit materials. Weak on glass, water, vegetation, reflective materials.
Navigation, smooth once the mesh is exported into a 3D engine or web viewer. Files are often heavy.
Maturity, a technology mature for twenty years, well tooled, but limited by the very nature of the polygonal mesh.
4. 3D Gaussian Splatting (3DGS)
Principle, from hundreds to thousands of photos taken in continuous motion, an algorithm reconstructs a cloud of 3D Gaussians that renders light and materials photorealistically.
Fidelity, photorealistic from every angle, including on difficult materials (glass, vegetation, skin, changing light).
Navigation, free and continuous in a web browser, on any recent device.
Maturity, an emerging technology (reference paper in 2023). Stable production since 2025. Commercial adoption accelerating in 2026.
Detailed comparison table (12 criteria)
| Criterion | 360° virtual tour | Laser scanning | Photogrammetry / mesh | 3D Gaussian Splatting |
|---|---|---|---|---|
| Visual realism | Good on captured angles | Weak (point cloud) | Medium (difficult materials) | Photorealistic from every angle |
| Freedom of navigation | Teleportation between fixed points | Limited, specialised viewer | Good in a dedicated viewer | Free and continuous |
| Metrological precision | None | Millimetre, reference | Centimetre depending on conditions | Centimetre, usable but not metrological |
| Capture time | Long on large volumes (multiplies points) | Long (hours to a day) | Variable with photo coverage | Fast, continuous motion |
| Delivery time | A few days | Several days to weeks | Several days to weeks | 24 to 48 hours |
| Device compatibility | Maximal, including old ones | Dedicated viewer | Dedicated viewer, heavy files | Web browser, recent GPU required |
| Long-term archiving capacity | Good (stable formats) | Good, but software dependency | Good, but software dependency | Good, open formats (PLY, SOG) |
| Gamification potential | Limited | Very limited | Possible but heavy | Native (paths, points of interest, narration) |
| Meshing with BIM | Very limited | Excellent (scan-to-BIM) | Possible | Being structured |
| Relative cost | Low | High | Medium | Moderate, accessible |
| Technological maturity | Very mature (10+ years) | Mature (15+ years) | Mature (20+ years) | Emerging (3 years in production) |
| Devices required to view | Any device | Pro software | Software or dedicated viewer | Standard web browser |
This table is a guide, not a verdict. The right choice depends on the use.
Which use case for which technology
Here are four typical situations, with the technology best suited to each.
”I want to quickly document a large catalogue of standardised properties at controlled cost”
→ 360° virtual tour. It is the format recognised by the public, technically accessible, and works on all devices. For properties whose visual differentiation value is not critical, it is enough.
”I need to provide precise technical drawings to an architect or feed a BIM”
→ Point-cloud laser scanning. It is the only technology that produces millimetre metric surveys, required by engineering firms and heritage services for sensitive projects.
”I want to model an object to integrate it into a game, an animation, or a 3D print”
→ Photogrammetry or manual modelling. The polygonal mesh remains the standard format for classic 3D pipelines. 3DGS will reach these pipelines, but it is not yet the most efficient use.
”I want to capture a place photorealistically for communication, archiving or experience”
→ 3D Gaussian Splatting. This is what it does better than the other three. Heritage, events, healthcare, high-end real estate, visual site monitoring, immersive marketing.
Combining several technologies on the same project
The right reflex on an ambitious project is not to choose a single technology, but to combine them. A few concrete examples.
Laser scanning + 3DGS on a heritage site
The laser scan produces the surveys required by the heritage authorities. 3DGS produces the photorealistic archive for mediation, communication and public viewing. The two feed different uses that do not compete.
360° + 3DGS on a property portfolio
360° quickly covers standard properties at scale. 3DGS is reserved for premium properties, exceptional venues, and spaces where the quality of experience justifies the investment.
BIM + 3DGS for site monitoring
BIM describes what should be (the as-designed model). 3DGS documents what is (the real as-built). The gaps between the two become visible, measurable, opposable. It is a powerful workflow for project owners who want to lock in execution quality. See our 3D site monitoring guide for the detail.
Photogrammetry + 3DGS for a museum
Photogrammetry to model collection objects in 3D (ideal for catalogue records, e-commerce, reconstructions). 3DGS to capture entire exhibition rooms and create an immersive tour. The mesh objects can be reimplanted in the 3DGS model as an interactive layer.
How to decide in practice
Three questions to ask before choosing.
Question 1, is the final use visual, metric or domain-specific?
If it is visual (communication, archiving, experience), 3DGS is probably the right answer. If it is metric (technical drawings), it is laser scanning. If it is domain-specific (structured BIM, maintenance, simulation), it is a BIM + laser scanning combination.
Question 2, what is the life horizon of the digital twin?
If you need a one-off deliverable to show in a month and forgotten in six, 360° may suffice. If you want an asset that lives over time, across varied channels, and that can be updated, 3DGS is better suited.
Question 3, do your target visitors have technical constraints?
If your audience includes many people with old devices, or if you need it to work in a public kiosk on an old Windows machine, 360° works better. For a B2B audience on recent hardware, 3DGS has no compatibility problem.
When to request a test on a sub-scope
On a high-stakes project, do not hesitate to order a pilot capture on a limited area before deciding on the full rollout. The cost of a pilot is limited, the reality check is valuable, and the technical choices are made on solid ground.
In summary
Four technologies, four dominant uses. The 360° virtual tour for volume at controlled cost. Laser scanning for metrological precision. Photogrammetry for classic 3D pipelines. 3D Gaussian Splatting for the navigable photorealistic experience, the reference archive and immersive communication.
The right choice is not a matter of fashion; it is a matter of fit between the technology and the expected final use. And on the majority of B2B projects in Switzerland today, 3DGS is the format that brings the most value perceived by decision-makers and end users, without head-on competition with the other technologies on their home ground.
To go further, see our projects that combine these approaches depending on the project, or our three engagement packages to understand how we structure assignments.
Frequently asked questions
What sets a digital twin apart from a 360° virtual tour?
A 360° virtual tour is a path of panoramic images taken from fixed points, navigable by teleportation. A digital twin is a complete 3D reconstruction of a space, freely navigable from every angle, sometimes enriched with domain data. 3D Gaussian Splatting and laser scanning produce digital twins. 360° is an intermediate format, simpler and more limited.
Can 3D Gaussian Splatting replace a laser scan?
For most everyday uses (estimating surfaces, distances, spatial organisation), yes. For millimetre-accurate metric surveys intended for an architect or engineering firm, the laser scan remains the reference. On the same project, the two can be combined: laser scanning for technical drawings and 3DGS for communication, archiving and visual experience.
How much does a building digital twin cost?
The cost varies greatly with the chosen technology, the size of the building, the customisation options, and long-term hosting. A 3DGS capture of a medium-sized building is delivered in a few days and remains economically accessible. A full laser scan for precise technical surveys takes more time and a larger budget. See our packages to understand the pricing structure.
Can a 3DGS digital twin be integrated into a BIM environment?
Yes, but with limits. 3DGS integrates well as a photorealistic visual layer in a BIM workflow, as a navigable as-built reference. To feed the structured BIM data directly (IFC, attributes), it is more complex and laser scanning remains better suited. The two technologies are complementary more than competing.
Which technology to choose to archive a heritage building?
For a heritage archive, 3DGS is today the best compromise between photorealistic fidelity, long-term accessibility and cost. A laser scan can complement it for areas where precise technical surveys are expected by the heritage services. 360° and classic photogrammetry are less suited to long-term archiving.
Read next
3D Gaussian Splatting: the complete guide to the technology redefining 3D
Understand 3D Gaussian Splatting, the 3D capture technology reshaping digital twins. Principle, use cases, limits, future.
3D Gaussian Splatting or 360° virtual tour: which to choose for your project?
360° panoramic virtual tour or 3D Gaussian Splatting, which technology suits your project? An honest comparison, use cases and limits.