How Much Detail a Concept Image Should Show

The hardest judgement in concept visualization is not technical. It is deciding how much to show, and both failure directions are common enough that most concept imagery lands in one or the other.

Too little and the audience cannot read the building, which defeats the purpose of commissioning anything. Too much and they read commitments nobody has made, which causes a specific and expensive problem later.

This guide is about finding the line, and about why it sits considerably lower than instinct suggests.

The test that settles most cases

Look at any element in the image and ask when it was decided. If somebody on the project can name the meeting, the drawing or the moment, it belongs. If the honest answer is that nobody has decided it, it should not be shown as though they had.

Applied strictly, that test removes a great deal from a typical concept image: the specific brick, the window mullion spacing, the canopy profile, the paving pattern, the tree species, the signage.

What survives is the volume, the height, the position, the openings as areas rather than as designs, and the relationship to everything around it. Which sounds thin and is in fact the entire content of a concept decision.

The discomfort this produces is real and it is worth anticipating. The images look unfinished because the design is unfinished, and representing that accurately is the value rather than a shortfall.

Categories instead of choices

An image cannot show nothing, so the undecided items have to be represented somehow, and the workable approach is to show a category rather than a selection.

Predominantly glazed rather than a specific curtain wall system. A masonry mass rather than a particular brick and bond. A darker base and lighter upper rather than two named products. Openings as proportioned areas rather than as resolved windows.

That communicates enough for a viewer to understand the character being proposed while remaining honest about the fact that the specific decisions are open.

It also survives value engineering, because nothing specific was promised, which removes one of the most common sources of late stage friction in a development.

Why more resolution makes the decision worse

This is counterintuitive and it holds up consistently, and it is worth understanding rather than simply accepting.

A resolved image invites a different kind of attention. Viewers stop comparing volumes and start reacting to the facade, the entrance, the material, the atmosphere, all of which are placeholder decisions that will change.

So the discussion moves away from what is actually being decided and toward what merely happens to be depicted. The group spends its attention on the parts of the image that carry no information and skips the parts that do.

A plainer image keeps the attention where it belongs, because there is nothing else to look at. That is not a limitation of a cheap image, it is a design property of a correct one.

The commitment problem

The expensive consequence of over resolution arrives months later and it is usually blamed on the design rather than on the image.

A concept render showed a particular facade rhythm and a warm material. Everybody saw it, several people liked it, and one of them remembers it clearly. The developed design, produced properly against a real budget and real constraints, looks different.

At that point the conversation is about why the building got worse, which is the wrong conversation, because it never was that building. The concept image invented a facade and the invention became the baseline.

Nobody intended it and the mechanism is entirely predictable, which is why the defence has to be structural rather than a matter of good intentions.

What legibility genuinely requires

There is a floor as well as a ceiling, and going below it produces images that are honest and useless.

A non technical viewer needs enough to judge size, which means scale reference: figures at realistic density, vehicles, and the surrounding buildings at correct height.

They need enough to judge position, which means the street, the ground and the immediate neighbours.

They need enough to judge character in broad terms, which means the material category and the proportion of solid to glazed.

And they need a viewpoint that corresponds to how a person would encounter the building, because an aerial answers arrangement and communicates nothing about experience.

That floor is not expensive. It is mostly context, and context is shared across options and built once.

The specific case of glass

Glazing causes more trouble at this stage than any other material, because it is unavoidable, expensive to represent honestly, and enormously persuasive when represented flatteringly.

A concept image with dark reflective glass reading a warm sky looks like a finished building and commits to a facade treatment that has not been chosen. The same building with neutral flat glazing reads as a proposal.

The workable approach is to show glazed areas as areas: correct proportion, correct position, neutral tone, no reflections doing dramatic work.

It looks less impressive and it is honest about the one thing least likely to survive to construction unchanged.

Landscape, which is almost entirely fiction at this stage

Mature trees at full canopy are a landscape decision, a budget commitment and a fifteen year assumption, and concept images routinely include all three without anybody noticing.

At concept stage what is usually known is roughly where planting will go and roughly how much space it has. What is not known is species, size at installation, or whether the budget survives.

Showing establishment scale planting rather than mature canopy is both more honest and more useful, because it shows the condition at handover rather than the condition somebody will see if they stay for a decade.

It also avoids the specific failure of a scheme appearing to depend on planting that later gets cut.

People, and the density question

Figures belong in a concept image for scale and for nothing else, and the density chosen quietly makes a claim.

A handful of people, placed where people would plausibly be, does the scale job completely and claims nothing.

A busy plaza claims an activation nobody has demonstrated, and it is one of the most common ways a concept image sells an outcome rather than describing a building.

The test is whether the number of people shown could be justified if somebody asked where they came from. Frequently it could not.

When more detail is genuinely warranted

There are cases where resolution serves the decision rather than distorting it, and they are identifiable in advance.

When the thing being decided is the facade, which happens on institutional and civic projects where the expression is the question rather than the massing.

When a specific relationship matters at close range, such as how a new element meets an existing building at a junction that has to be resolved.

When the audience will see the material without anybody present, and a certain amount of finish is what makes it comprehensible rather than persuasive.

In each of those, the additional detail is answering a question rather than filling a gap, which is the distinction that matters.

Different levels for different images in the same package

It is legitimate, and frequently better, to vary detail within a package provided the variation is between images rather than between options.

A distant view establishing the scheme in its context needs less. A close view addressing a specific junction needs more. Both can appear in the same package.

What must not vary is the detail between options in the same view, because that is the comparison and any difference in treatment decides it.

Stating that structure explicitly in the brief prevents the common misunderstanding where a client asking for more detail on one image receives more detail on everything.

Why the pressure always runs toward more

It is worth understanding why concept images drift toward over resolution, because knowing the mechanism is what allows a team to resist it.

The client asks for more because the plain version looks unfinished in a meeting and nobody wants to present something that appears unfinished. That is a reasonable instinct about how they will be perceived rather than about what the image needs to do.

The studio supplies more because finished imagery is what their portfolio is built from, what their team is skilled at, and what demonstrates capability. Producing something deliberately plain feels like underperforming.

Neither party is doing anything wrong and the combined effect is a package that costs more and informs less. The only defence is deciding the level in the brief, in writing, before either instinct has a chance to operate.

Presenting a plain package without apologising for it

The practical difficulty with plain concept imagery is not producing it. It is standing in front of a group with it, and that is worth preparing for rather than discovering.

The framing that works is stating the purpose before showing anything: these images are deliberately neutral so that the only visible difference between the options is the difference being decided, and anything not shown is something that has not been decided.

Two sentences, delivered before the first image, convert what would read as unfinished work into what reads as rigour. Delivered afterwards, in response to a question, the same explanation reads as a defence.

It also sets up the second round correctly, because the group now understands that resolution comes after selection rather than being withheld arbitrarily.

The one thing worth over showing

There is a single exception to the general rule of restraint, and it runs the other way: the surrounding context deserves more attention than the proposal does.

The neighbours, the street, the ground and anything that will remain are all decided already, because they exist. Nothing about them is provisional, so showing them accurately carries none of the commitment risk that showing a facade does.

They also do most of the explanatory work. A viewer judges the height of a proposal against the building beside it and its bulk against the width of the street, which means the accuracy of those references determines whether the judgement is sound.

So the sensible distribution of effort at concept stage is inverted from what most packages do: comparatively more on the things that already exist, comparatively less on the thing being proposed.

How to ask for the right level

Describe what has been decided rather than describing a look, and let the studio infer the level from that.

Say that massing, height and position are fixed while facade, materials and landscape are open. A studio that understands this stage will produce the correct level from that sentence alone.

Ask explicitly for neutral light and categorical materials, because the professional default is toward attractive and a studio will supply it unless told otherwise.

And ask them to flag anything they had to invent in order to make an image work, because there is always something and knowing what it is prevents it from silently becoming a commitment.

The summary in one paragraph

Show the volume, the height, the position, the openings as areas, the material as a category, the context at correct height, figures for scale, and neutral light.

Do not show the facade design, the specific material, the entrance detail, the landscape at maturity, the signage or the atmosphere.

If somebody can point at an element and ask when it was decided, and nobody can answer, remove it. That single test produces the right image far more reliably than any description of a style, and it costs less to produce than the alternative.

Frequently asked questions

What is the test for how much detail to include?

Point at any element and ask when it was decided. If somebody can name the meeting or drawing, it belongs. If nobody has decided it, showing it as though they had creates a commitment that the developed design will later appear to break.

How should undecided materials be shown?

As categories rather than selections. Predominantly glazed rather than a specific curtain wall, a masonry mass rather than a named brick, openings as proportioned areas rather than resolved windows. It also survives value engineering because nothing specific was promised.

Why does more resolution make the decision worse?

Because it moves attention away from what is being decided. Viewers stop comparing volumes and react to facade, entrance and atmosphere, all of which are placeholders that will change, so the group spends its attention on the parts carrying no information.

What is the minimum a concept image needs?

Enough to judge size, position and broad character: figures at realistic density, surrounding buildings at correct height, the street and ground, material as a category, and a viewpoint corresponding to how a person would encounter the building.

When is more detail actually justified?

When the facade is the thing being decided, when a specific junction with an existing building has to be resolved, or when the audience will see the material without anybody present to explain it. In each case detail answers a question rather than filling a gap.