GenAnywhere Generate Now

Select a map area. Generate a 3D city model with AI.

No drone survey required. Turn map data into editable 3D buildings, terrain, water and vegetation for BIM, GIS and 3D workflows.

Generate now

Alternative workflows for site models: what to compare, GenAnywhere

Five steps from question to handover, and the points where one workflow quietly costs more than another.

Alternative workflows for site models are usually compared by their output, which is the least useful comparison. The differences sit in the input, the edit, the export and the handover, and each of those steps has a cost that shows up later. A short comparison makes those costs visible before a team commits.

Aerial view of a waterfront district with cranes and warehouses

1 Define the workflow

1.1 What counts as a workflow?

A workflow is the sequence from a question to a scene someone else can use. It includes the input, the edits, the export and the handover, and each step has a cost.

Comparing workflows means comparing those steps, not comparing the final screenshots. A scene that looks good and cannot be edited is a poor workflow for early work.

Write the sequence down before comparing. A list of five steps makes the missing ones obvious, and it gives the discussion a common structure. A workflow written down can be compared; one held in memory cannot.

1.2 Why compare at all?

Because the right workflow depends on the stage, the site and the team. A method that suits a feasibility study may be wrong for a planning submission.

Comparison also exposes assumptions. A team that assumes a flight is available may find the workflow depends on a budget line nobody owns.

The comparison is not a competition between tools. It is a way to see which steps the team can actually complete. The comparison should be repeated when the team changes, because a workflow depends on who runs it.

2 Compare the input

2.1 What input does each need?

A drawn area, a flight, a survey or a set of drawings. The input determines what the workflow can produce, and it is often the constraint that decides.

GenAnywhere needs a polygon on a map and reads map and satellite data behind it. No flight and no survey are required for the initial context.

That does not make the input free of limits. Map data has a date and a level of generalisation, and both belong in the note. The input list is also a checklist for what could go wrong.

Brick waterfront buildings and a promenade beside a river

2.2 Which input is hardest to get?

A flight needs access, weather and processing. A survey needs site time and a scope. A drawn area needs a decision about extent, which is cheaper but still a decision.

Name the hardest input before comparing outputs. A workflow that depends on the hardest input is fragile at the stage where questions move quickest.

The hardest input is also the one most likely to be skipped. A plan that assumes it will arrive on time is a plan with a hidden dependency. A dependency nobody owns is a risk, not an input.

3 Compare the edit

3.1 How editable is each?

A base can replace, move and resize blocks without rebuilding context. A capture needs a new flight or a rebuild for most changes.

Editability matters when the proposal is still moving. A workflow that cannot keep up with the design conversation becomes a record of an earlier idea.

Test the edit before the meeting. Move one block and see what else changes, because that is the cost the team will actually pay. A base edit is reversible; a capture rebuild is a new record.

3.2 What does an edit cost?

The cost is the number of parts that must be touched. A base edit touches one block; a mesh edit may touch the surface around it.

Count the parts before the change, not after. A change that looks small on screen can be large in the file.

The cost also includes the record. An edit that is not noted makes the scene harder to explain, even when the change itself was simple. That record is part of the cost, and it is usually unpaid.

4 Compare the export

4.1 What does the export carry?

The base exports a general GLB with geometry and materials. A capture exports a mesh, often large and often without layers.

Compare what survives the export. Units, origins and names can change, and a workflow that loses them costs the receiver time.

Test the export in the tool the receiver will use. Standards published by the OGC are a useful reference for how geospatial data is described across systems. A file that opens in the receiver's tool is worth more than one that needs conversion.

Aerial view of logistics warehouses beside a highway

4.2 What is lost?

Authored meaning is lost in a general export. A wall becomes a surface, and a layer becomes a group of objects.

That loss is acceptable for context and unacceptable for coordination. Say which use the file serves before the receiver discovers the difference, and put that sentence in the note.

A short note about the loss is more useful than a long list of formats. The receiver needs to know what the file can do, not what it cannot.

5 Compare the handoff

5.1 What does the handoff need?

A source, a date, a coverage drawing and a note that names the checker. Those four items make a scene usable by someone who was not there.

A handoff without them is a file, not a record. The receiver cannot tell what was observed from what was assumed.

The handoff also needs a named contact. A question with nowhere to go becomes a guess. A contact name turns a dead end into a question, even when the sender expects to be available.

5.2 What does the receiver do?

Open the file, test one dimension, check the layers and read the note. That routine takes minutes and prevents most surprises.

Record the result with a date. The next handoff starts from a known baseline, and the check does not have to be repeated from scratch. A recorded check can be trusted by someone who was not present.

If the test fails, stop and ask. A scene that is used before it is checked tends to be defended rather than corrected.

6 Choose

6.1 Which workflow fits an early stage?

The one that starts soonest and stays editable. A map-derived base usually wins on both counts, because it needs no flight and no survey.

The one that carries the context the question needs. A base that includes the next street is more useful than a detailed site alone.

The one the team can repeat. A method used once is an experiment; a method used on every site becomes a workflow. Repeatability is the difference between a method and an anecdote.

6.2 Which fits a later stage?

The one that carries authored elements and coordination. At that point a design model or a BIM workflow takes over.

The base can remain as context, but it should not be stretched. A context scene asked to carry coordination will fail at the earliest revision.

The transition should be planned. Naming the earliest authored element marks the point where the job changes hands, and a planned transition is cheaper than an improvised one.

6.3 When do you revisit?

Revisit when the stage changes or the site changes. A workflow that suited concept may not suit submission.

Revisit when a new input arrives. A survey can change the answer more than a new tool.

Revisit when the handoff keeps failing. A repeated problem is a sign the workflow, not the file, needs attention. A workflow that is never revisited becomes a habit.

Draw the area in the GenAnywhere app and compare the workflow with the one the team already uses.

7 FAQ

FAQ-1 What is being compared?

Five steps: input, edit, export, handoff and the record that travels with the scene. Comparing outputs alone hides the costs that appear later. A workflow that looks convincing in a screenshot may be the hardest one to repeat. That is why the list starts with inputs rather than outputs in practice.

FAQ-2 Is a flight required?

No. A map-derived base needs a drawn area and map data, and it can supply the arrangement without a flight. Where surface condition matters, a capture is still the right record and should be treated as a separate source, with the note saying which answered which question on the same site.

FAQ-3 Can workflows be mixed?

Yes. A base can carry the arrangement while a capture supplies surface evidence and a design tool carries the proposal. The mix works when each source keeps its label and the note says which one governs which question. The labels are what make the mix work, along with the note itself.

FAQ-4 What is the most common failure?

The handoff. Teams spend effort on the scene and little on the note, so the receiver cannot tell what was checked. A one-page note with the source, date, coverage and checker prevents most of the trouble. The note takes an hour and saves a rebuild in most site reviews today.

See how a map area becomes an editable site model in the GenAnywhere workflow.

Content and technical notes

This article was prepared by the GenAnywhere team with AI-assisted drafting and human review.

Product features, formats, and availability change as the platform evolves; refer to the official GenAnywhere website for current details.

For technical questions or corrections, contact support@genanywhere.com.

Last reviewed: 2026-11-06.