ImaginaryFS

How a video game actually gets built, and why it takes longer than anyone said.

The register
Sections 01—09

09 — The Credits

Studio size

Headline headcount and the number who actually shipped are different figures.

Empty office cubicles with mesh chairs and dividers lit by dim natural light
Fig. 1

The room the schedule actually happens in.

Photo: cottonbro studio / Pexels

The number that shipped it is not the number on the press release

A studio announces a team of eighty. The credits roll on a game built, in any meaningful sense, by forty-three — the others having joined during crunch, left mid-project, worked on a cancelled feature, or contracted for six weeks in 2021 and never returned. Neither figure is dishonest. Both are real. They describe different things.

Headline headcount is a snapshot: who was employed at peak, or at announcement, or on the day a marketing brief went out. Credits reflect something closer to a process: who touched the work, across how long, and whether the studio's crediting policy captured them at all. Who is in the credits is already a contested question before studio size enters it.

Why the gap exists

Teams grow and shrink in shape rather than in orderly waves. A small core — often ten to twenty people — builds the prototype and the engine foundation. That group sets the architecture; everything downstream depends on decisions they made, usually under time pressure, before the budget was confirmed. As production scales, the headcount rises fast: artists hired to fill content, QA contractors brought in as a build nears completion, audio staff who enter the project in its final third. Then it drops again. By the time gold master is declared, a studio that "had eighty people" may be running on thirty-five.

The contracted workers complicate this further. Outsource studios supply environment assets, localisation, motion-capture cleanup, and concept work. A given game might draw on four or five such companies across its production, each contributing dozens of person-months of labour. Some of that work lands in the credits. Much of it does not, or lands only at the vendor level — one company name standing in for twenty individuals.

An empty motion-capture stage, cameras and floor marks
Fig. 2

Floor marks and camera rig. The stage day is scheduled months before the animations are needed.

Photo: Bruno Massao / Pexels

What the number actually tells you

Studio size as reported is useful for roughly estimating ambition and burn rate, but it is a poor guide to the actual production structure. A forty-person studio with a stable core and a disciplined pipeline can outship an eighty-person operation whose headcount peaked during a chaotic scaling phase and then haemorrhaged. The size that matters is not peak headcount; it is the size of the team that was present for the decisions — scope, architecture, what to cut — and who stayed to implement them.

Productions run on continuity more than on capacity. A person who spent three years on a project and left six months before ship contributed more than the meaning of their credit (or absence from it) suggests. The number that shipped the game is made up of people at different depths, for different durations. Any single figure flattens that.

Credits reflect something closer to a process: who touched the work, across how long, and whether the studio's crediting policy captured them at all.

A dim office at night with a few desks still lit
Fig. 3

What the end of a project looks like from outside: quiet, and mostly small corrections.

Photo: cottonbro studio / Pexels

Every figure in this entry is an illustrative example, not a measurement of any particular production. It is not a measurement of any particular production.