Early-stage design models are useful for testing whether a development could work on a site. They help teams explore building height, footprint, scale, and the relationship with neighbouring properties.
The difficulty is that not everyone can interpret a simple massing model.
To an architect or developer, a collection of blocks may communicate plenty of information. To a client, landowner, planning consultant, or local stakeholder, the same model can feel abstract.
A more realistic image can help people understand what is being proposed. It can also make early conversations more focused by giving everyone something clear to respond to.
Projected lets you turn a view of your massing model into an architectural render without rebuilding the project in separate visualisation software.
What is AI architectural rendering?
AI architectural rendering uses an existing image, such as a view of a 3D massing model, as the basis for a more realistic visual.
The aim is not to create a completely new design. It is to add materials, lighting, landscape, and architectural detail while retaining the broad form and composition of the original model.
In Projected, you can choose:
- A view of the proposal
- A material direction
- A season
- A time of day
- A camera style
The result is an illustrative architectural image that can be used to discuss the project, compare ideas, or add context to an early feasibility study.
Why render a massing model?
Massing models are good at communicating size and form, but they often leave important questions unanswered for non-specialist audiences.
People may find it difficult to imagine:
- What the building could feel like at street level
- How a proposed material might affect its appearance
- How the development relates to nearby buildings
- How the site might look in different seasons
- Whether the proposal feels appropriate for its setting
An architectural render can make these questions easier to discuss.
It does not provide a final answer, but it can reveal preferences, concerns, and misunderstandings earlier in the process.
That can be valuable during feasibility work, when a proposal is still open to change.
A simpler route from model to image
Traditional architectural visualisation can involve exporting a model, preparing materials, adding lighting, building the surrounding scene, and adjusting the camera in separate software.
That level of work is justified for important planning submissions, sales material, and final presentations. It may be excessive when a team simply needs to communicate an early option.
Projected offers a shorter workflow:
- Open the massing model.
- Position the view.
- Choose the visual settings.
- Generate the render.
- Review the result against the original model.
The image remains connected to the feasibility study, making it easier to move between the visual and the underlying proposal.
Early client discussions
Some clients find plans and massing diagrams difficult to read. A realistic image can help them understand the broad direction of a proposal and give more specific feedback.
Instead of asking whether they “like the design”, you can discuss:
- Whether the building feels too dominant
- Which materials suit the location
- How the entrance should be expressed
- Whether the landscaping feels appropriate
- Which parts of the proposal need further development
Comparing material directions
Materials can significantly change how the same building form is perceived.
A darker façade may make the building feel more solid. Timber may soften the appearance. Brick may create a stronger relationship with neighbouring properties.
Generating several options from the same viewpoint makes these differences easier to compare.
The renders should not be treated as accurate material samples, but they can help narrow down a direction before more detailed design work begins.
Feasibility reports
A feasibility report often needs to communicate both evidence and possibility.
Plans, site data, and development calculations provide the evidence. A visual can help explain what those findings could mean in practice.
Used carefully, a render can make a report more accessible without suggesting that the design is more developed than it really is.
Internal design reviews
Early images can help a team identify issues that are easy to overlook in a simple model.
A render may prompt questions about:
- The scale of the ground floor
- The rhythm of the façade
- The relationship with the street
- The balance between hard and soft landscaping
- The visibility of servicing or parking
- The treatment of neighbouring properties
The image is not necessarily correct, but the reaction it creates can still be useful.
Testing atmosphere and context
The same proposal can feel very different depending on weather, light, planting, and season.
Testing these conditions can help a team avoid relying on a single idealised image. It can also encourage discussion about how a place might feel throughout the year, rather than only on a bright summer afternoon.
Comparing options from the same viewpoint
Visual comparisons are most useful when the viewpoint remains consistent.
If one material option is shown from street level and another from a distant aerial view, it becomes difficult to judge the actual differences between them.
Projected saves the viewpoint used for each render, allowing images created from the same position to be grouped together.
This is useful when comparing:
- Different materials
- Alternative massing options
- Day and evening conditions
- Summer and winter settings
- Changes made during the design process
Using the same view helps keep attention on the design decision rather than the presentation.
Choosing a useful viewpoint
The quality of a render depends partly on the view you choose.
A dramatic angle may create an attractive image, but it may not be the most informative. For feasibility work, it is usually better to choose a view that helps answer a specific question.
Street-level views
Use a street-level view to examine:
- How the building meets the pavement
- The perceived height of the proposal
- Entrances and ground-floor activity
- The relationship with neighbouring buildings
- The likely experience of pedestrians
Wider context views
Use a wider view to understand:
- The overall scale of the development
- The relationship between multiple buildings
- Open space and landscape
- The position of the proposal within the site
- The surrounding urban pattern
Elevated views
An elevated view can help explain:
- Building depth and footprint
- Courtyards and roof areas
- Access and circulation
- The arrangement of open space
- The relationship between different parts of a larger site
Where possible, use several practical viewpoints rather than relying on a single presentation image.
How to get more useful results
AI-generated architectural images are not exact, but a few simple choices can make them more useful.
Start with a clear massing model
The image should make the proposal easy to distinguish from its surroundings. If the starting view is crowded or unclear, the generated result is more likely to become confusing.
Frame the question before generating
Decide what you are trying to learn from the image.
For example:
- Does brick feel appropriate for this street?
- Is the building too dominant from this viewpoint?
- Would more planting soften the edge of the site?
- Does the entrance need to be more visible?
- How does the proposal feel in winter?
A clear question makes it easier to judge whether the image is useful.
Compare one change at a time
If you change the material, viewpoint, season, and lighting together, it becomes difficult to understand what influenced the result.
For a clearer comparison, keep most settings consistent and change one element at a time.
Keep the original model nearby
Always compare the finished image with the underlying massing model.
Check whether the render has altered:
- The apparent building height
- The number of floors
- The footprint
- The roof shape
- The position of the building
- The relationship with neighbouring properties
If it has, the image should not be used to support conclusions about those parts of the proposal.
Label the image clearly
AI-generated images should be presented as illustrative material.
A simple label can explain that the image represents an early design direction and may include details that have not yet been designed or verified.
This is particularly important when sharing images outside the immediate project team.
What these renders cannot tell you
A realistic image can look convincing even when parts of it are inaccurate.
The rendering process may invent or alter details such as:
- Windows and doors
- Floor divisions
- Façade patterns
- Roof details
- Planting
- Street furniture
- Reflections
- People and vehicles
- Features on neighbouring buildings
It may also make small changes to the building’s proportions or surroundings.
For that reason, an AI architectural render should not be used as proof of:
- Planning compliance
- Accurate building dimensions
- Daylight or sunlight performance
- Verified views
- Material performance
- Construction detail
- Accessibility
- Buildability
- Development viability
Those questions should be answered using the model, drawings, site information, technical studies, and professional review.
The render is a communication tool, not a substitute for evidence.
Avoiding false certainty
Photorealistic images can make an early proposal appear more resolved than it is.
A massing study may only define the building envelope, but the generated image can add windows, balconies, entrances, landscaping, and finished materials. Viewers may assume those details have already been designed.
To avoid that impression:
- Describe the image as illustrative
- Explain which decisions have and have not been made
- Show the massing model alongside the render
- Avoid presenting invented details as approved design features
- Include more than one option where appropriate
- Keep technical conclusions separate from visual impressions
The purpose of the image should be to support discussion, not to give an uncertain proposal a false sense of completion.
AI rendering and architectural visualisation
AI rendering and traditional architectural visualisation are useful at different stages.
AI rendering is well suited to:
- Early feasibility work
- Fast option testing
- Internal reviews
- Initial client conversations
- Exploring broad material directions
- Communicating massing to non-specialists
Traditional visualisation remains more appropriate for:
- Carefully controlled planning images
- Verified viewpoints
- Final marketing material
- Detailed façade representation
- Accurate material specification
- Images requiring precise art direction
One does not have to replace the other.
An early AI render can help a team decide which ideas are worth developing. A visualisation specialist can then produce a more controlled image once the design has enough detail to justify the work.
What comes next
The first version focuses on residential massing, a defined set of materials, a small number of camera styles, and a limited range of seasonal options.
Future improvements may include:
- More material and façade choices
- Better preservation of windows and floor divisions
- Additional landscape controls
- Clearer comparison tools
- Automatic retries for temporary generation failures
- More detailed render histories
- Stronger warnings when generated geometry appears to diverge from the source
As image models improve, some of the prompt constraints may become less necessary. The broader workflow should remain the same: capture the live design view, preserve its context, generate the image, store it securely, and make the result easy to reproduce.
The goal is not to make feasibility studies look more finished than they are. It is to help more people understand the proposal while there is still time to question it, compare options, and make changes.
A tool for better early conversations
The value of an architectural image is not simply that it looks realistic.
Its value comes from helping people ask better questions while the project is still flexible.
Does the building feel appropriate for the site? Is the scale understandable? Which material direction deserves further work? What needs to change before the proposal moves forward?
Projected’s rendering tools are designed to support those conversations directly from the massing model.
The result is not a finished design and should not be presented as one. It is a way to make an early proposal easier to understand, compare, and question.
That is often exactly what a feasibility study needs.