What to Give a Renderer for a Pitch Deck

Rendering projects for pitch decks stall for the same reasons every time, and almost none of them are about rendering. They stall because nobody decided what the building is made of, nobody knows which drawing is current, or nobody asked the deck designer where the text sits.

A sponsor who arrives with those settled gets images quickly. One who arrives with a site plan and an intention gets a list of questions first, and on a live raise those questions consume time that was already short.

This guide sets out exactly what to send, what genuinely can wait, and how to answer the handful of decisions that only a sponsor can make.

The five inputs that block production

1. Drawings, and which revision is current

Plans and elevations at whatever resolution exists, plus a clear statement of which set supersedes the others. This sounds procedural and it is the single most common cause of wasted work.

Projects in motion generate several versions within weeks. Production against a superseded set is a silent error: everything looks correct until somebody compares the image to the current drawings, usually after it has circulated.

Elevations matter as much as plans. A plan establishes footprint and a plan alone leaves height, floor to floor dimensions and window positions to be inferred, and inference produces a building that is subtly not yours.

2. Site information with orientation

Boundary, north point, and levels where the ground is not flat.

Orientation decides every lighting condition in every image. Without it the choice is arbitrary, and an image lit from an impossible direction is the kind of error that a reader who knows the site notices immediately.

On sloping ground, survey data rather than estimation. Guessed levels produce entrances that sit wrong relative to the street and outdoor spaces that are flat in the image and sloped in reality.

3. Material intent

Not a specification. A decision about tier and character: masonry or panel, warm or cool, punched openings or glazed, and roughly the quality level the cost assumption implies.

This is the input sponsors most want to defer and the one that cannot be deferred, because it is what the imagery communicates. Left open, somebody in production decides it by default, and the default will not match what the sponsor meant.

The fastest way to supply it is reference imagery rather than description. Two or three buildings, with a line about what is right in each. Twenty minutes of work that removes an entire round of interpretation.

4. The slide layout

Where text sits, what aspect ratio the image occupies, and how many slides carry imagery. Obtainable from whoever is producing the deck in one message.

Without it the image arrives composed centrally and the layout crops it, which either loses the subject or produces the awkward compromise that everybody recognises and nobody mentions.

5. Agreement with the schedule

Storey count, unit count and mix exactly as they appear in the financial model, confirmed as current.

Imagery that disagrees with the numbers is the fastest way to lose a reader, and the disagreement almost always originates in a schedule that changed after the drawings were issued and before anybody told production.

What helps but does not block

An existing architectural model. If one exists it may be reusable, which is the largest single saving available. It will not be presentation ready and that is normal.

Photographs of the site and its surroundings. Genuinely useful for context accuracy, and a phone is sufficient.

Landscape intent. Frequently unresolved and easily indicated generically at this stage.

Interior layouts. Needed only if interiors are in scope, and they resolve later than exteriors almost every time.

Brand guidelines. Relevant to the deck rather than to the imagery.

What can be decided later at no cost

Camera positions and composition, which are production decisions rather than sponsor ones. Time of day, unless the argument depends on it. Whether to add further views, since views of an area already modelled are inexpensive. Final crops and export formats.

Projects are frequently held up over these while the genuinely blocking items stay open, which is worth naming because the pattern is so common.

The send list

The decisions only a sponsor can make

Four, and a supplier who makes them for you has guessed at your business.

Quality tier. What this asset competes with. It is a commercial decision that already exists implicitly inside the cost assumption.

Who the reader is. Equity, debt or a mix, because it changes which view leads and what each emphasises.

Design stage to depict. How resolved to make it look, which is a judgment about honesty rather than about production capability.

What must not be shown. A parcel under negotiation, an unannounced tenant, a phase not yet public, an access route not secured. Better established before than corrected after circulation.

How to answer the material question in twenty minutes

Since it blocks everything and sponsors resist it, a practical method helps.

Find three buildings you would be pleased to be compared with. They do not need to be in your market or your use class. Screenshots are fine.

For each, write one line about what is right and one about what is not. Warm brick and the window proportion, but not the heavy cornice. That is enough for production to author materials that read the way you intend.

Then add the caption you are comfortable with: materials indicative, subject to design development. Nothing has been committed, and the imagery communicates the correct tier.

This is faster than a specification, more precise than adjectives, and it is the single highest leverage twenty minutes a sponsor can spend on this work.

What to send when the drawings are incomplete

The common real situation, and there are workable answers rather than a requirement to wait.

If elevations do not exist, send the plan with floor to floor heights written on it and a sketch or a description of the elevational intent. That is enough to build honest massing with material read, which is the right output at that stage anyway.

If the site survey is not back, send whatever level information exists and say it is provisional. Production can build on an assumed datum and adjust, provided everybody knows that is what is happening.

If the ground floor use is undecided, say so. It is better to produce a view that treats the ground condition neutrally than to have production invent a retail frontage that later becomes a residential lobby.

The failure is not incompleteness. It is incompleteness presented as completeness, because then production makes assumptions silently and nobody discovers them until review.

How much context to ask for

A decision the sponsor should make deliberately rather than leaving to the supplier, because it moves the price more than anything else.

The useful question is what a reader needs to understand the position. On a dense urban site that usually means the immediate neighbours accurately and the wider block indicatively. On an open site it means whatever establishes the parcel relationship: the road, the adjacent development, the direction of growth.

Asking for a whole district in accurate detail is expensive and rarely earns its cost, because readers are assessing the immediate condition rather than the neighbourhood at large.

Asking for no context at all is cheaper and it removes the argument, since a building on white ground says nothing about where it is or what it is worth.

Stated in blocks or in metres, this single line removes the largest source of quote variance in the category.

Sequencing when time is short

Produce one view fully and review it before the others start.

The first image resolves everything that would otherwise be relitigated on each subsequent one: material read, lighting approach, level of resolution, how context is handled, how much population is right.

Once that is agreed the rest proceed quickly and with far less supervision. Running all views simultaneously means every comment on the first applies retroactively to all of them, which is the slowest possible arrangement and the one deadlines push people toward.

If the deadline genuinely cannot move, produce fewer views properly rather than more approximately. A document with one credible image and two clean diagrams outperforms one with four images that each raise a question.

How to review the first image

Review is where schedules are won or lost, and most sponsors review badly through no fault of their own, because nobody explains what to look at first.

The first review should cover only fundamentals, in this order. Is this the right view for the question the image exists to answer. Does the material read match the tier intended. Is the level of resolution honest for the design stage. Is the context sufficient and accurate.

Those four decide whether the image is right. Everything else is refinement, and raising refinements before the fundamentals are settled means the refinements get invalidated when a fundamental changes.

The second review is detail: the crop, the population, the vegetation, the sky, the small corrections. That is where a sponsor should be specific and demanding, because the expensive decisions are already fixed.

Mixing the two, which is what happens naturally when a group looks at an image, produces five rounds instead of two and a result nobody is happy with.

Working with the architect

A relationship worth handling deliberately, because it decides both cost and accuracy and time pressure tends to skip it.

If the architect holds a model, request it early. It is usually the cheapest possible starting point and it avoids paying to rebuild geometry that exists. It will not be presentation ready, which is normal and expected.

More important than the model is agreement about what is being shown. Imagery produced without the architect seeing it frequently depicts details the design has already ruled out, and that surfaces awkwardly when somebody compares the two.

The lightweight version is one review of the first image before the rest proceed. An hour of somebody time, and it prevents the most common category of rework.

Where no architect is appointed yet, say so in the brief. It changes what production can rely on and it changes the honest caption on the result.

What goes wrong when this is skipped

Material decided by default. The imagery communicates a tier the sponsor did not intend, and readers price the opportunity accordingly.

Superseded drawings. The most avoidable form of rework and the most irritating.

Cropping damage. An image composed without the layout, then cut to fit it.

Imagery contradicting the schedule. Usually a schedule that moved silently.

Review by committee. Four reviewers, sequential rounds, a result blander than any individual view of it.

One boundary worth stating

This guide covers what a rendering supplier needs from a sponsor. It does not cover architectural design, structuring an offering or securities compliance, which are design, financial and legal questions rather than visualization ones.

Our own terms, stated rather than implied: first visuals in 48 to 72 hours once these inputs exist, and reasonable revisions are included at no extra charge. We do not raise capital, we do not advise on securities, we do not obtain approvals and we do not guarantee them, and we do not sell the asset.

Frequently asked questions

What is the most common cause of wasted rendering work?

Producing against superseded drawings. Projects in motion generate several versions within weeks, and the error is silent: everything looks correct until somebody compares the image to the current set, usually after it has circulated.

Why is material intent required before starting?

Because it is what the imagery communicates. Left undecided, production decides it by default and the default will not match what the sponsor meant, which means readers form a view about quality tier that the sponsor did not intend.

How do you supply material intent quickly?

Reference imagery rather than description. Three buildings, with one line each about what is right and what is not, plus a caption saying materials are indicative and subject to design development. Twenty minutes, and it removes a whole round of interpretation.

Why does the renderer need the slide layout?

So the image can be composed for the frame it will occupy, with the subject in the readable half and quiet area where the text lands. Without it the image arrives centred and the layout crops it, losing either the subject or the legibility.

What should be produced when the deadline is immovable?

Fewer views properly rather than more approximately, and one view reviewed fully before the rest begin. Running all views at once means every comment on the first applies retroactively to all of them.