Architectural Animation: What It Costs and How It Gets Made

Architectural animation is priced per second of finished footage, which is the first thing that surprises people commissioning it. A sixty-second flythrough is not one image with movement. It is roughly 1,500 individual frames, each rendered, plus the camera work, sequencing, and post-production that make them into a film.

This guide covers what actually drives the cost, how the work is produced, and when animation is the right deliverable.

Why animation is priced by duration

A still image requires one camera, one lighting setup, and one render. An animation requires a camera path, motion for anything that moves, and a render of every frame along that path. At 30 frames per second, a minute of footage is 1,800 renders.

Render time per frame on a detailed architectural scene runs from several minutes to over an hour depending on complexity, lighting method, and resolution. That compute is real and it is a genuine cost line, which is why quotes are structured around finished duration rather than around scene count.

The practical consequence is that duration is the primary budget lever available to a client. Cutting a ninety-second piece to sixty seconds removes a third of the cost. Adding thirty seconds adds it back, which is why scope discipline on run time matters more here than on any other visualization deliverable.

The production stages

Brief and storyboard. What the animation has to accomplish and for whom. An investor piece, a sales center loop, and a social media cut are three different films from the same building. The storyboard defines the sequence of shots before any production time is spent, and approving it is the cheapest moment to change direction.

Modeling. The building is built or the existing model is prepared. Animation is less forgiving than stills here, because the camera moves and sees geometry that a fixed camera never would. A model built for one hero image frequently cannot support a flythrough without additional work.

Camera and animatic. Camera paths are set and rendered as a fast, untextured preview. This is the single most important review stage. Approving camera movement on an animatic costs almost nothing; changing it after final rendering means re-rendering everything downstream of the change.

Materials and lighting. Surfaces, finishes, and the lighting scheme. Time of day is a decision with real consequence, because it determines mood and because changing it after lighting is set means redoing the pass.

Population and detail. People, vehicles, vegetation, water, and moving elements. Static entourage reads as artificial the moment the camera moves, so animated people and traffic are a meaningful cost line rather than a garnish.

Rendering. The compute-heavy stage. Frames are distributed across machines and the wall-clock time depends on how much capacity is thrown at it. This is where a compressed schedule costs money directly.

Post-production. Color grading, compositing, transitions, titles, and sound. Sound is routinely skipped and it should not be: a silent flythrough feels unfinished in a room, and music plus ambient audio is a small fraction of the total budget.

What moves the price

Duration. The largest single variable and the one most within the client's control.

Interior versus exterior. Interiors are usually more expensive per second. Every surface, fixture, and furnishing has to be modeled and lit, and interior lighting is computationally heavier than exterior daylight.

Source model quality. A clean, complete model reduces the largest labor line in the project. A partial model, or a set of drawings with no model at all, means building it, which is a separate scope with its own price.

Resolution and frame rate. HD, 4K, and higher each multiply render time. 4K at 60 frames per second is roughly four times the compute of 4K at 30, and most delivery contexts do not require it.

Amount of movement in the scene. A slow exterior orbit is far cheaper than a walkthrough passing animated people, moving traffic, water, and changing lighting.

Revision rounds. Two is standard. What counts as a revision matters more here than in stills, because a camera change after rendering is not an edit, it is a re-render.

Schedule. Rush work costs more because render capacity has to be bought or reallocated. Compression buys throughput, not additional revision rounds.

When animation is the right deliverable, and when it is not

It works in a controlled setting. A sales center, an investor meeting, a planning hearing, a trade show. Anywhere someone is seated and attending, animation communicates spatial sequence and scale in a way stills cannot.

It works for master plans and large sites. When the product is a place rather than a building, an aerial flythrough communicates relationships that no single image captures.

It underperforms as a website hero. Most visitors do not watch past the first several seconds unprompted. A strong still image does more work at the top of a website than a long animation nobody finishes.

It underperforms in social feeds unless cut for them. A sixty-second cinematic flythrough posted natively performs poorly. Short vertical cuts of the strongest moments perform far better, and those cuts should be planned during the storyboard rather than salvaged afterward.

Stills are the better buy when budget is limited. Six strong stills cover more of the marketing surface than thirty seconds of animation, and they are reused across every channel. Animation is an addition to a still package rather than a substitute for one. Both are covered under 3D walkthroughs and animation.

Getting the most from the budget

Decide duration first and hold it. Set the run time in the brief. Scope creep in this category is measured in seconds and it is expensive.

Approve the animatic carefully. This is where a client should spend their attention. Camera paths approved at the untextured stage cost nothing to change; approved after rendering they cost the render.

Commission stills from the same model. The model is already built and lit. Adding stills at that point is dramatically cheaper than producing them separately, and it guarantees the stills and the animation match.

Plan the cuts up front. The full-length piece, a thirty-second version, and vertical social cuts should be defined in the storyboard. Producing them together is a fraction of the cost of going back for them.

Do not over-specify resolution. 4K is appropriate for a large display or projection. For web and social, it is compute spent on pixels nobody sees.

Lock the design before production starts. A facade change after materials and lighting are set invalidates every rendered frame. This is the most expensive category of late change in visualization.

Timeline

A typical sixty-second architectural animation runs four to eight weeks from a complete model. Roughly one week for storyboard and camera, one to two for materials and lighting, one to two for rendering, and one for post-production, with review cycles between stages.

Where the model has to be built from drawings, add two to four weeks. Where the design is unresolved, the schedule is indeterminate, because production cannot start on a moving target without redoing work.

The most common scheduling error is commissioning animation at the same moment it is needed. Working backward from the sales center opening or the investor meeting, production should start two to three months before, not two to three weeks.

Frequently asked questions

How much does architectural animation cost?

Pricing is per second of finished footage and scoped per project, because duration, interior versus exterior, source model quality, and scene complexity move the number far more than building size does. A defined duration with a complete model produces a fixed quote.

How long should an architectural animation be?

Sixty to ninety seconds covers most marketing and investor uses. Sales center loops can run longer because the audience is seated. Social cuts should be fifteen to thirty seconds and vertical. Longer is not better, it is more expensive.

What is the difference between a flythrough and a walkthrough?

A flythrough is an exterior aerial camera moving around and over a project, used for context and massing. A walkthrough is an interior camera at eye level moving through spaces, used to communicate scale and finish. Many projects need both, and they are priced separately because the production requirements differ.

Can animation be produced before the design is final?

It can and it should not be. A design change after materials and lighting are set invalidates rendered frames rather than requiring an edit. Massing animations from an unresolved design are the appropriate deliverable at that stage.

Do we need sound?

For anything shown in a room, yes. A silent animation reads as unfinished. Music and ambient audio are a small fraction of total cost and materially change how the piece lands. Licensing matters, since unlicensed music in a public presentation is an avoidable exposure.

Should stills be commissioned at the same time?

Almost always. The model is built and lit for the animation, so producing stills from it costs a fraction of producing them independently, and it guarantees that the imagery across the campaign matches.

Choosing the shots that earn their run time

Most animations contain twenty seconds of footage doing real work and forty seconds of footage filling time. Deciding which shots earn their place is the discipline that separates a piece that holds attention from one that gets skipped.

The establishing aerial. Where the project sits and what surrounds it. This answers the location question in three seconds and it is almost always worth including.

The approach. Moving toward the building at eye level, the way a visitor would arrive. Communicates scale and street presence better than any static view, and it is the shot people remember.

The threshold moment. Passing from exterior to interior. This is technically demanding because lighting has to transition, and it is the single most persuasive movement in an architectural animation because it is the one nobody can imagine from drawings.

One or two interior sequences. The main living or working space and one differentiator, the amenity or the view. Not a tour of every room, which is where run time gets consumed without adding persuasion.

The amenity or the view. Whatever the project is actually selling. On a tower it is the view. On a community it is the shared space.

What to cut. Long slow orbits with nothing changing, corridors and circulation, and any shot that exists because the model was already built. If a shot does not answer a question a buyer is asking, it is costing money per second to be there.

Where animation budgets get wasted

Commissioning before the design is resolved. The most expensive error in this category. A material change after lighting is set means re-rendering, not editing, and re-rendering is the compute bill again.

Approving camera work late. Clients who skim the animatic and react to the final render are paying for the re-render of every frame downstream of the change. The animatic exists precisely to prevent this and it is routinely rushed.

Over-specifying resolution and frame rate. 4K at 60 frames per second for a piece that will play on a website is compute spent on pixels nobody perceives.

Producing one long piece with no cuts. Going back later for a thirty-second version and vertical social cuts costs more than planning all three from the start, because the edit has to be revisited and sometimes reframed.

Skipping sound and then adding it. Audio designed against a locked edit is straightforward. Audio added to a piece that then gets recut is redone.

Splitting animation and stills across vendors. Two teams build two models, light them differently, and produce imagery that visibly does not match. The client pays for the same building twice and gets an inconsistent campaign.

Real-time and interactive as an alternative

Game-engine real-time rendering has changed what is possible, and it is worth understanding where it fits rather than treating it as a replacement.

What real-time does well. Unlimited camera positions without re-rendering, live material and finish changes, and interactivity where a viewer controls the movement. For a sales center or a configurator, this is genuinely better than a fixed film.

What it costs. More upfront than a single animation, because the entire environment has to be optimized to run at frame rate rather than rendered once. The economics work when the asset will be used repeatedly and changed over time.

Where offline rendering still wins. Maximum visual quality for a hero piece. A carefully lit and rendered flythrough still exceeds what real-time produces on the same hardware, and for an investor presentation where the image quality is the argument, that difference matters.

The practical split. Offline rendering for the hero film and the marketing stills. Real-time for the interactive experiences in a sales environment. Projects with both use each where it is strongest rather than forcing one to do both jobs.

What to send a studio to get an accurate quote

Quotes vary widely for the same project, and most of the variance comes from studios pricing different assumptions rather than charging different rates. Six items eliminate almost all of it.

The model, or an honest statement that there is not one. Revit, SketchUp, 3ds Max, or the CAD set. A studio quoting without knowing whether it has to build the building is guessing, and the guess will be high or the invoice will be higher.

The intended duration. The single largest cost variable. Even an approximate range changes the quote materially.

Where it will be shown. Sales center projection, investor meeting, website, social. This determines resolution, aspect ratio, and how many cuts are needed.

Interior, exterior, or both. Interiors cost more per second, and a quote that assumes exterior-only will be revised upward when interiors appear in the brief.

Finish and material selections. If they are chosen, send them. If they are not, say so, because that determines whether the studio is producing a marketing piece or a design study.

The date it has to be delivered. Not the internal review date, the date it plays in front of an audience. Studios schedule render capacity backward from it.

A studio that returns a shot list and a stage-by-stage breakdown against that brief has read it. One that returns a single number has not, and that number will move.

Animation by project stage

The same building justifies very different animation at different moments, and matching the deliverable to the stage prevents paying marketing prices for a design study.

Concept and feasibility. A short massing animation with no materials, showing volume and site relationship. Fast, inexpensive, and appropriate because the design will change. Spending on finish quality here is discarded.

Entitlement. Context-accurate movement from vantage points the public actually occupies. Accuracy matters more than beauty, and a camera path that flatters the project from angles nobody stands in undermines credibility with a reviewing body.

Capital raise. A tight sixty-second piece with the hero exterior, the location argument, and enough interior to support the rent assumptions in the model. Shown in a room to a seated audience, which is the setting animation was made for.

Pre-sales launch. The largest production. Full flythrough, interior walkthroughs across the unit mix, amenity sequences, and the social cuts. This is also where the same model should produce the still package, because the marginal cost of stills at that point is small.

Construction and lease-up. Short updates rather than new production. As the building becomes visible, real photography progressively replaces animation, and the animation gets retired rather than refreshed.

Commissioning the pre-sales package during feasibility is the most common sequencing error, and it means paying full production cost for a building that will change before it is built.