What problem are we trying to solve?

What's the cause, what data do we have to support our hypothesis?

Could we use AI to help us analyze player feedback or behaviour data?​

Goal list should inspire ideas, fix problem, increase fantasy, strengthen game vision, encourage certain player behavior or relevant to the desired elements of your audience.​​

Prioritize different goals, and make sure they are not in conflict.​​

Could we use AI to do design review?

Analyze other games, discuss or brainstorming with the team.​​

How can we use AI to learn from case studies in other games?

Design solution that solves the problem, reaches the goals, and connects to the rest of the game.​​

Consider scope, cost and time pressure to propose a feasible solution.

• Be aware of design cost, balance cost and learning cost of a solution, and seek how to mitigate it by current system.​

How to use AI to make solution more concrete and work for different players?

Use paper, game reference, engine or AI to prototype before putting more resource on dev prototype.

Could we use AI to create placeholder fast?​

Was the problem solved? Were the goals achieved?

Design test format based on your problem. Who is participant? How to collect data?

Could we use AI to confirm we have prototype good enough for testing? ​

Compare data from different tests to see if we are on the right track.

Add telemetry points and interpret them wisely.

Could we ask AI to organize and summarize feedback and consider them with data together.​

Onboarding, UI requirement, dev tool, asset list, polish.

Could we use AI do the documentation​ and final review?