Co-development vs outsourcing
Compare ownership, uncertainty, communication and hand-off across the two engagement models.
Shared build integrationRepresentative editorial scene — not Kioto Gaming staff, premises or client work.
Embedded production capacity
An integrated delivery team for a defined feature stream, production problem or playable milestone.
Discuss this scopeProduction question / 01
Game co-development works when an external team joins the production system around the game, not just its ticket queue. The first scope defines the build, feature boundary, decision owners, review path and acceptance evidence. Your leads keep product and creative authority while the delivery team owns the agreed result.
See how the scope is builtCo-development differs from bounded outsourcing because uncertainty is expected inside the engagement. A design can change after playtesting, a dependency can move, or integration can expose a constraint that was invisible in the brief. The working model needs enough context and decision access to respond without hiding the impact on scope, schedule or quality.
A responsible starting point is a contained production boundary. That may be one feature, one discipline stream, a cross-disciplinary pod or a recovery plan for a milestone already under pressure. The boundary names what the external team owns, what remains with the internal studio and which shared systems need joint review.
Delivery stays connected to the same branch, build cadence and acceptance process wherever access permits. Reviewable increments show whether the collaboration is integrating before the commitment grows. The final hand-off records source, configuration, decisions, known limits and the ownership required for the next phase.
The exact team shape follows the build and milestone. These are common ownership patterns, not a fixed package.
A feature stream can be taken from behaviour and acceptance criteria through implementation, content support, QA and integrated review. Dependencies and decision owners remain visible throughout the work.
Repository flow, builds, review states, naming, tools and reporting are mapped before scale. The aim is to contribute inside the existing production truth rather than create a parallel system.
A bounded pod can combine the disciplines needed to reach one playable outcome. Roles and allocation are proposed only after the work, internal coverage and review capacity are understood.
When a stream is blocked, the first task is to separate missing work from missing decisions. The recovery boundary includes the evidence needed to accept the result and transfer ownership cleanly.
A useful proposal explains how the external team joins production, not only which roles appear on a staffing table.
The proposal names the feature, stream or milestone the external team owns, the internal dependencies that remain outside the scope and the people who can accept consequential changes. It also states what evidence closes the boundary.
Repository, build, engine, tools, source material, security rules and review cadence all affect the plan. Required access and internal review time are recorded as delivery assumptions with a clear response if they are delayed.
The first gate should produce an integrated artifact, not a status report. Both teams should know when it will be reviewed, which choices follow and how the result changes the next commitment.
Deliverables are agreed against the current build and responsibility boundary. A co-development scope can include:
Each step reduces a different integration risk before more capacity is committed.
Step 01
We identify the next milestone, current build state, missing capacity, dependencies and the decision the work must support.
Step 02
Both teams agree repository and build access, review authority, working cadence, security constraints and escalation paths.
Step 03
The first useful gate is a small result running in the correct build. It tests the pipeline, communication loop and acceptance method before the stream grows.
Step 04
Implementation and content move through playable or inspectable increments. Forecast changes and unresolved risks stay attached to the work.
Step 05
The milestone closes against agreed evidence. Source, configuration, known issues, decisions and next-step ownership are transferred with the result.
Most delivery problems begin in the working boundary, not in a missing status meeting.
Repository access does not create integration when internal owners cannot review builds, answer design questions or accept changes. The plan reserves real client-side decision time.
Expected iteration and a changed milestone are not the same. The engagement needs an agreed way to show how new evidence changes capacity, schedule or the accepted result.
A feature can cross several teams while still requiring one accountable acceptance path. Dependencies, final decisions and hand-off ownership must not disappear between organizations.
These signals justify a discovery conversation. They do not replace an assessment of the project.
Clear answers depend on the project, but these are the distinctions worth resolving before a proposal.
Co-development usually carries shared production context and outcome ownership inside an evolving feature or milestone. Outsourcing can be the better fit for a stable, well-specified batch. The right model depends on uncertainty, integration depth and who can make decisions.
That is the intended model when access and security conditions allow it. Discovery verifies engine and repository versions, build flow, review states, tools and permissions before the delivery plan assumes integration.
Team size follows the smallest ownership boundary that can reach the milestone. Adding people before dependencies, review capacity and role gaps are understood can increase coordination without increasing accepted output.
Bring the current build state, the next decision or milestone, known blockers, target timing and the internal owners who can review the work. A complete design document is useful only when it reflects the current production reality.
We will map a responsible first ownership boundary and the evidence needed before the engagement grows.