Brawl AI
A UX concept for an in-game recommendation system that helps Brawl Stars players choose what to play, who to use, and how to progress with less decision fatigue.
Project overview
Brawl AI explores how personalized recommendations could be integrated directly into Brawl Stars to help players make confident gameplay and progression decisions.
The concept focuses on a Smart Play Advisor that turns player context, active quests, available Brawlers, game modes, and resources into one clear next step.
The problem
Players face an increasing number of Brawlers, game modes, upgrades, quests, events, and limited resources. The challenge is no longer finding information, but deciding what action is most useful right now.
This creates uncertainty, decision fatigue, and reliance on external communities for advice before playing or spending resources.
Choice overload
A growing roster of Brawlers, modes, quests, and events makes the next action difficult to choose.
Limited resources
Players hesitate to spend coins and upgrades because a poor choice can slow progression.
External dependency
Advice is often found on Reddit, YouTube, and Discord, interrupting the in-game experience.
UX case study
The complete case study documents the research process, player pain points, recommendation logic, design decisions, user flow, and final interface concepts.
Browse the four-page document below to explore the project from problem discovery to the final Smart Play Advisor experience.
Solution concept
The Smart Play Advisor is introduced as a native recommendations area in the main game interface, keeping guidance inside the player’s normal flow.
It offers two focused tools: a Daily Progress Plan for today’s goals and a Mode Advisor for selecting the strongest available Brawlers for a mode.
Daily Progress Plan
The plan combines active quests, unlocked Brawlers, available modes, and current resources to recommend one efficient action path.
Players see the suggested Brawler, mode, quests that can be completed together, and estimated rewards before applying the plan.
Mode Advisor
Players select a game mode and receive three ranked recommendations from their available Brawlers.
Showing the best option, an alternative, and a beginner-friendly choice keeps the recommendation useful without overwhelming the player.
Recommendation logic
The concept uses a weighted scoring model to balance short-term goals, gameplay suitability, resource value, and what is actually available to the player.
Design decisions
The interface deliberately avoids presenting every possible option. It prioritizes one recommended action and limits alternatives to the most useful choices.
Simple language, familiar game visuals, visible rewards, and immediate feedback keep the feature consistent with the existing Brawl Stars experience.
Show only the best, alternative, and beginner-friendly options.
Recommend actions that help the player progress immediately.
Refresh guidance as quests, resources, and game content change.
Final interface
The final experience guides players through a simple journey, from discovering recommendations to receiving personalized gameplay advice based on their current progress.
Project outcome
Brawl AI demonstrates how an AI-inspired recommendation experience can transform complex player data into simple, actionable guidance while preserving the pace, visual language, and fun of the game.