AI microdrama studio vs in-house: which should you run?

Cost splits three ways, and none is simply cheapest. An AI studio quotes per finished minute or per episode, a premium that buys their pipeline and their people so you never touch a model. Freelancers cost less per hour but the real bill is coordination: a writer, an editor, a colourist and a localizer each booked, briefed and chased. Running it in-house on a platform is a single plan plus a small team, so after the first episode pays for the setup, the marginal cost of each additional episode is low. For a 60-episode season, the studio route is predictable and expensive, freelancers are variable, and in-house trades a learning curve for the lowest per-episode cost.

Speed depends on where the waiting happens. An AI studio is the fastest to hand off, you brief once and step back, but delivery sits in their production queue and every change costs a round-trip. Freelancers move at the speed of whoever is free that week, so a season stretches across everyone's calendar. In-house removes the hand-off entirely: a shot is drafted, reviewed and re-rendered in the same sitting, and a locked episode structure repeats for the next one. The studio wins the first week; in-house wins the next fifty-nine.

Control is the clearest divide. With a studio, your influence is the quality of your notes and how many rounds the contract allows; the pipeline is theirs. Freelancers give direct control, but it is fragmented, the editor cannot fix what the generator produced, and the colourist inherits both. In-house keeps the whole chain in one place: the script, the shots, the cut and the grade all answer to the same person in the same session, so a small change does not become a chain of hand-offs. If the look matters shot by shot, in-house gives the most say.

Ownership of outsourced work is exactly what the agreement says, no more. With a studio or a freelancer, negotiate a full assignment or work-for-hire so the scripts, the voices and the footage are yours and not licensed back to you; read the deliverables clause and the usage terms before you commit a season to it. Running it in-house changes the default: the scripts, the saved character and location references and the generated footage live in your own workspace, so there is nothing to assign and nothing to claw back. Whichever route you choose, confirm the ownership terms of any platform or vendor in writing before a 60-episode commitment.

A 60-episode season is a scale problem, and each route scales differently. A studio scales with budget: more episodes, more invoice, and their capacity is the ceiling. Freelancers scale with people and the coordination between them, which grows faster than the headcount. In-house scales with process: once the cast references are saved and a per-episode sequence is captured as a reusable workflow, each new episode is duplicate, swap the script, run, so the eightieth episode costs the same effort as the eighth. For a long season, a repeatable in-house process is the one that does not get more expensive per episode as it grows.

The team you need is smallest at the two ends and largest in the middle. A studio needs only a producer to brief and approve. Freelancers need that producer plus a hire for every craft, which is the heaviest coordination load of the three. In-house on a platform collapses the crafts into one tool: a writer to own the story and a producer-operator to generate, cut and assemble means a two-person team can carry a season. It is not zero-skill, someone has to learn the workspace and keep the quality bar, but the headcount is a fraction of a traditional crew.