CGI vs Live Action Product Commercials: Which One Should You Book?

What actually separates the two routes?

Not realism. Modern CGI clears the realism bar for product work routinely, and the adidas ZK 2K Boost films were never trying to look filmed in the first place. What separates the routes is where the risk sits and what happens when someone changes their mind.

In live action, risk is concentrated on a single day. Everything before it is preparation and everything after it is limited by what was captured. In CGI, risk is spread across the whole schedule: nothing is unfixable, but nothing is finished either, and there is no natural moment where the film stops being negotiable.

Which factors should decide the choice?

Factor	CGI	Live action
Cost drivers	Asset build, simulation, number of unique shots, reformats	Crew, location, talent, product samples, shoot days
Product availability	Not required, CAD is enough	Physical sample required
Iteration after approval	Re-render, no new crew	Reshoot, full booking again
Impossible shots	Native capability	Needs VFX anyway
Locations	Built, no travel or permits	Scouted, booked, weather-dependent
Human performance	Weak point	Strength
Variants and colourways	Cheap after the first	A new setup each time
Formats and reframes	Re-render from the same scene	Depends on what was shot
Asset reuse next campaign	High	None
Time to air	No shoot day, front-loaded build	Shoot day plus post
Where the risk sits	Spread across the schedule	Concentrated on one day

When should you book CGI?

1. The product does not exist yet. Launch windows regularly require a finished film before the first sample ships. With CAD, that is straightforward.
2. The shot cannot be filmed. Cutaways into a product, physically impossible motion, scale changes, materials behaving in ways they never would. The PUMA Be the Spark film zooms inside a footballer's head; that is not a location scout problem.
3. You need many variants. Colourways, market-specific packaging, seasonal versions. The first one costs, the rest are adjustments.
4. The campaign is social-first. Vertical, square and horizontal from one scene, plus stills, with no crop compromises.
5. The subject is a material, not an object. BASF Design Fabrics existed because material behaviour is close to unfilmable in a way that communicates anything.
6. The asset should outlive the campaign. If the product will be marketed for years, building it once is an investment rather than an expense.

When should you book live action?

1. A person carries the film. Faces and performance remain the thing CGI is worst at, and audiences are unusually good at spotting a synthetic one.
2. A real place is the point. Documentary, provenance, craft, origin stories.
3. The scope is one simple shot of an existing product. A well-lit table-top shoot can beat CGI on both cost and speed here.
4. The tone must feel unpolished. CGI defaults to control. Fighting that costs more than embracing a camera.
5. The brand's visual language is photographic and consistency matters more than spectacle.

Is the hybrid route a compromise?

No, and it is increasingly the default. Filming the environment and building the product digitally gets you real light, real texture and a real place, with a product that can be changed at any point. The Samsung Galaxy Z Flip 7 campaign works exactly that way: filmed space, tracked camera, digital product, composited so it reads as one image.

The catch is that hybrid work requires a shoot planned for it. Lens data, clean plates, lighting reference and a camera move that can be solved. Adding CGI to a plate that was shot without any of that is possible and expensive.

How do the cost structures actually differ?

Live action cost is dominated by the shoot day and everything that has to align on it. CGI cost is dominated by how much has to be built and how many times it is iterated. The practical consequence: a live action budget is mostly fixed before shooting and mostly spent afterwards, while a CGI budget is elastic in both directions and expands with feedback rounds.

That is why the useful question in a CGI brief is not "how long is the film" but "how many unique shots, how much simulation, how many formats". The breakdown sits in what drives the cost of a CGI commercial.

What about time to air?

Live action needs lead time before the shoot for casting, location and permits, then post. CGI needs no shoot but front-loads asset build, and its schedule is more sensitive to approval latency than to render capacity. In my experience the two routes land in a similar window for a comparable film, with one difference: a CGI schedule can absorb a late product change, a live action schedule usually cannot.

Which route ranks better for reuse across markets?

CGI, by a distance. Localised packaging, market-specific text, different colourways and regional formats are re-renders of an existing scene. In live action each of those is a setup, and often a reshoot.

How should I brief either route?

Same discipline both ways: state what the film has to achieve, name the air date, name the approval chain, and bring three references with a sentence each about what you like. The rest is scoping. If you want the CGI route scoped concretely, see CGI commercials or 3D product animation, or just send the brief.