1. The three options
Estimated ranges. Same output in all three cases: every plant modelled in the portal. What changes is who does the work, what it costs, and when it is ready.
| Library + assembler | All outsourced | In-house (Revit + DE) | |
|---|---|---|---|
| Total cost | ≈ 150 k€ | ≈ 367 k€ | 60–125 k€ visible |
| Digital Engineering time | none | none | 6–12 months |
| Who builds a plant | Sales, 1 h | Studio, 1.5 d | Revit engineer, 1 d |
| Materials and look | Included, done once | Included, redone often | Separate studio work |
| Committed delivery date | Yes, contractual | Depends on supplier | None — internal priorities |
| Visual consistency | Guaranteed, one library | Varies by plant | Varies by author |
| Cost of one more plant | 35 € | 1 200 € | ≈ 400 € |
2. What is being bought
Ranges, not quotes. Each phase is estimated from the information available today and is confirmed at the end of the study, before anything beyond it is committed.
| Phase | Deliverable | Price |
|---|---|---|
| 1 | Study, asset specification, viewer mockup | 12–15 k€ |
| 1 | Viewer design — screens, components, interaction states | 3 k€ |
| 2 | Library | 47–66 k€ |
| 3 | 5 pilot plants, delivered while the tool is built | 8–9 k€ |
| 4 | Assembler, training, documentation | 32–46 k€ |
| Development | 80–159 k€ | |
| Perpetual licence, first year of support included | 26–30 k€ | |
| Total, one-off | 106–194 k€ | |
| Support and maintenance after year one, optional | on request |
How the library is counted. 15 to 20 component types, 4 to 5 design variants each — 60 to 100 variants in total. Sizes are not counted: each variant is parametric, so a 5 m and a 10 m grate are the same asset, and adding a size costs nothing.
Two ways to buy it. Either as a whole, at the price above, or variant by variant: a one-off 6 k€ for the parametric kit, then 800 € per variant, 700 € beyond 30, sizes included. Batches of 10 minimum. Variants with no usable reference are quoted separately. Buying by variant means you can start with 20 and extend later, at the same rate.
The library and the specification belong to Kanadevia Inova. The assembler is licensed, perpetually and non-exclusively, for the delivered version. Blender itself carries no licence cost, on any number of seats.
3. The return
Estimated on mid-range figures. Clients already have Pulse 360; the 3D model of their own plant is sold on top of it. The question is not how many plants exist, but how many clients take the add-on.
| With the assembler | Outsourced | |
|---|---|---|
| Clients taking the 3D add-on | 100 | 100 |
| Revenue | 1 000 k€ | 1 000 k€ |
| Production cost | 121.5 k€ | 282 k€ |
| Margin | 878 k€ — 88 % | 718 k€ — 72 % |
| Difference | +160 k€ | — |
| Break-even | 62 plants | — |
| Covered by | 15 clients — 8 % of the fleet | — |
4. What the numbers leave out
The in-house column looks comparable on cost alone. It carries no committed delivery date, no tool a salesperson can use, and no guarantee that two plants look like they belong to the same fleet.
It also takes 25 to 30 component families away from Digital Engineering's own roadmap, and still needs a studio for materials.
Two questions settle it: on what schedule can Digital Engineering absorb those families, and who produces the materials afterwards?