Summary
Preproduction is the stage where a game idea stops being a vague ambition and becomes a testable project. The aim is not to prove that the game will definitely succeed. The aim is to reduce uncertainty: define the hook, name the intended audience, identify the major risks, test the core assumptions, and decide whether the project is worth deeper investment.
In other words, preproduction answers a narrower question than launch planning: do we understand this idea well enough to prototype and scope it responsibly? (Bond, Introduction to Game Design, Prototyping, and Development, see source-introduction-game-design-prototyping; Kristjan, We Deserve Better Villains, see source-we-deserve-better-villains).
Key ideas
Start with the hook, not the feature pile
If a team cannot explain the game’s fantasy, emotional promise, or distinctive angle in a sentence or two, it is not ready for large-scale implementation. Kristjan’s Blue Sky phase treats the elevator pitch as the first real artefact of development, while recent GDC marketing talks argue that weak messaging often reflects weak conceptual focus rather than weak copywriting (Kristjan, We Deserve Better Villains, see source-we-deserve-better-villains; GDC marketing talks, see source-indie-marketing-gdc-talks).
Useful early questions:
- What feeling or fantasy is the player buying into?
- What makes this game meaningfully distinct from close comparables?
- Who is this for, and who is it not for?
Audience framing is part of design, not just marketing
Concept validation should include a rough model of the intended player. That does not mean creating a fake marketing persona and treating it as truth. It means being explicit about genre expectations, player literacy, session length, skill demands, and platform habits. If those assumptions stay hidden, scope and interface decisions become guesswork.
Kickstarter’s official pre-launch guidance reinforces this from a business angle: teams should identify and organise a real community before launch, not assume the audience will appear later (Kickstarter Creator Guidance, see source-kickstarter-creator-guides).
Compare honestly to existing games
Comparable titles are not there to justify copying trends. They are there to clarify expectations and risk. A useful comparable set helps a team discuss:
- what players will already assume
- where the game must meet baseline expectations
- where it plans to differentiate
- whether the scope matches the team’s actual capacity
Preproduction should surface risks explicitly
Bond’s prototyping model is risk-first: the riskiest assumption should be tested before lower-stakes work. In practical terms, that means preproduction should maintain a short list of assumptions that could invalidate the project if false: control feel, combat readability, content scale, performance, art pipeline, tutorial clarity, or monetisation fit (Bond, Introduction to Game Design, Prototyping, and Development, see source-introduction-game-design-prototyping).
Prototype before you promise slice-quality polish
The common failure mode is building polished content before the team has proved the core interaction or audience promise. A prototype asks “does this work?” A vertical-slice asks “can we present this convincingly and build it repeatedly?” Moving to slice work too early often creates expensive false confidence.
In practice
For a small Unity project, a good preproduction pass can be very lightweight:
- Write a one-paragraph pitch and a one-sentence hook.
- List three to five comparable games and what each one teaches you.
- Name your three biggest risks.
- Build one prototype per major risk, starting with the core interaction.
- Show the prototype to a few target players and record where their expectations differ from yours.
- Decide whether the project should be cut, reframed, or taken forward into vertical-slice and minimum-viable-game planning.
Do not treat preproduction as a document-writing contest. Treat it as the stage where you make uncertainty visible enough to make a responsible next decision.
Evidence
Schell’s design-lens approach encourages teams to question assumptions from multiple angles rather than treating the first concept as finished truth (Schell, The Art of Game Design, see source-art-of-game-design).
Bond argues that the riskiest assumption should be prototyped first, because that is the assumption most likely to damage the project if wrong (Bond, Introduction to Game Design, Prototyping, and Development, see source-introduction-game-design-prototyping).
Kristjan’s Blue Sky phase frames early ideation around the elevator pitch and warns against rushing past concept work before the game’s soul is clear (Kristjan, We Deserve Better Villains, see source-we-deserve-better-villains).
Recent GDC marketing talks add a modern practitioner warning: if the team cannot communicate the hook clearly in store-page or trailer form, the concept probably still needs clarification (GDC marketing talks, see source-indie-marketing-gdc-talks).
Implications
- A team that skips preproduction usually pays for it later in scope creep, muddy marketing, and unclear user experience.
- Strong concept validation improves both design and sales because it forces the team to describe what the game actually is.
- Preproduction ends when the next unknown is better answered by building a prototype or slice, not by holding another meeting.
Open questions
- How much audience validation is enough for a student project before it becomes performative market theatre?
- At what point should a small team stop refining the pitch and simply build the prototype?
- Which preproduction tasks are still worth doing for highly experimental games that intentionally reject genre expectations?
Related
overview-full-game-development-pipeline | prototyping | vertical-slice | minimum-viable-game | game-marketing-fundamentals | game-design-documentation | source-indie-marketing-gdc-talks