Commissioning CGI Ads: What to Settle Before the Budget Is Approved

Most disappointing 3D advertising projects were decided before production started. Not by the studio's talent, but by four or five decisions made at briefing stage that nobody revisited. Getting those right is less glamorous than reviewing concepts, and it matters considerably more.

The first is ownership of the source files. A render is an output; the model, textures, rigs and scene files are the asset. Agreements that transfer only the final video leave you commissioning a new build every time you want a variant. Agreements that transfer the project files mean the second campaign costs a fraction of the first.

Scope the variants at the start

Teams routinely brief a single master spot and then discover they need nine aspect ratios, four language versions, three regional pack shots and a set of stills for retail. Each of those is cheap if the scene was built with them in mind and expensive if it was not. A scene composed tightly for a 16:9 frame does not crop to 9:16 without re-staging.

List every deliverable you can foresee, including the ones you are not certain about, before the quote is finalised. Studios producing CGI ads at volume will price a bundle far more favourably than a sequence of change requests, and the technical decisions they make early will differ depending on what they know is coming.

Understand where the cost actually sits

Clients tend to assume cost scales with duration. It scales with complexity. Thirty seconds of a product rotating against a gradient is cheaper than five seconds of the same product in a splash of liquid, because simulation, refraction and the render time they demand dominate the budget.

Characters are the other threshold. A product is an object: model it, light it, render it. A character has to be rigged, animated, and made to move convincingly, and audiences detect wrong movement instantly even when they cannot name the problem. If a character appears in the brief, the project is in a different cost bracket, and it is worth knowing that before the budget is signed off rather than at the first review.

Build the review process around the pipeline

3D production is sequential in a way photography is not. Concept, modelling, texturing, lighting, animation, rendering, compositing. Each stage builds on the one before, so a change approved late unwinds everything downstream. Approving a layout and then changing the camera angle after lighting is finished is not a tweak; it is a partial rebuild.

Agree review gates at each stage and insist that sign-off genuinely means sign-off. The studios that deliver on time are usually the ones that refuse to proceed without it, and that refusal protects your schedule more than it protects theirs.

Set the measurement before launch

Because 3D spots are visually striking, they attract attention that is easily mistaken for performance. Decide in advance what the asset is for. Thumb-stop rate and view-through are reasonable measures for an awareness piece. Cost per acquisition is the measure for a performance asset, and a beautiful render that does not convert is still a failure.

Run it against your existing control creative rather than against nothing. The honest comparison is not whether the 3D spot performed well, but whether it performed better than what you would have run instead, at the cost difference between them.