Overview
GGWP is a real-time AI-powered moderation platform built for online games. With a mission to make multiplayer experiences safer and more inclusive, GGWP equips moderators with intelligent tools to detect and address harmful behavior—before it spirals.
Objectives
This case study covers my end-to-end process in designing both the product experience and the brand identity from scratch. It includes product thinking, system architecture, UI/UX design, and visual language development.
MY ROLE
Lead Designer — Strategy, Research, UX/UI, Design Systems, Branding
TEAM
1 Product Manager, 2 Designers, 8 Developers, 3 Founders
DONE WITH
Figma, Illustrator, ChatGPT, Relume, Lovable, Maze
INDUSTRY
Gaming
Direction and Challenge
The designs for GGWP focused on bold typography, creative illustrations, and a clean, minimal UI to create easily digestible content areas without overwhelming users. oxicity and harassment in online games have become rampant. While most moderation tools are reactive, GGWP needed to flip the model—using AI to proactively flag disruptive behavior and give moderators real-time context and control.
Design goals:
Craft a modern, trustworthy brand that resonates with both studios and players.
Build a moderation dashboard that balances power and simplicity.
Surface AI-powered insights with clarity, not complexity.
Make the experience feel real-time, responsive, and humane.
Research Insights
Most tools are retroactive — they rely on reports, not detection.
Moderators need better real-time context to take faster, fairer actions.
Players feel frustrated when action feels invisible or inconsistent.
There’s high distrust in “automated bans” — human review is still essential.
Research Takeaways
From interviews with moderators and devs:
Need for context (not just alerts)
Real-time action is critical
Players distrust invisible moderation
Competitive Analysis
Positioning GGWP in the Market
We analyzed competitors like TwoHat, Spectrum Labs, and bespoke in-house tools.
Key gaps in existing solutions:
Reactive systems that rely on user reports
Lack of context provided for decisions
Poor UX—outdated dashboards and manual review tools
What sets GGWP apart:
Real-time AI flagging with explainable outcomes
Rich player behavior timelines and summaries
Clean, intuitive moderation tools built for modern teams
Human-in-the-loop workflows that increase trust and transparency
Visual Design
As part of my case study, I am utilizing the “Shipment List” module. This module is a crucial element of the application, featuring six key components including filters, timelines, a map, container view, custom views, and a Prediction Score.
After creating the mockups, I remained involved in the project, overseeing QA and user testing.
Powerful Incidents Listing
The Incidents Listing acts as mission control for moderators, surfacing real-time flagged events with key details like severity, timestamp, and behavior type. Sortable columns, filters, and severity tags streamline triage, while each row expands into a detailed view with chat logs and AI-backed insights for confident decisions.
Incident Insights
AI Summarization That Builds Confidence
The Insights panel gives moderators a clear snapshot of what happened—highlighting behavior type, severity, and sentiment shifts using AI-powered analysis.
Instead of sifting through logs, moderators get a quick, readable summary that helps them take confident, fair action in less time.
Smarter Context
Positioning GGWP in the Market
GGWP's Player Detail view gives moderators a holistic view of a user's behavior—not just one isolated event. It includes prior incidents, positive behavior badges, chat frequency, and historical trends.
This level of context sets GGWP apart from reactive tools. It supports fairer, more informed decisions and builds a reputation system that rewards good players—not just penalizes bad ones.
Secure Onboarding
Fast Setup, Built-In Safeguards
GGWP’s onboarding flow is designed to get studios and moderators up and running quickly—without sacrificing control or security. Role-based permissions, SSO support, and audit logs are built-in from the start.
The experience balances speed with accountability, making it easy to scale moderation teams while keeping sensitive data and actions fully traceable.
Proactive Signals
Proactive Signals Before Problems Escalate
Player Alerts notify moderators when a user shows early signs of disruptive behavior—before it becomes a major issue. These alerts are powered by real-time pattern recognition and sentiment analysis across voice, chat, and behavior logs.
Each alert includes a summary of recent activity, confidence level, and suggested next steps—giving moderators the opportunity to intervene thoughtfully, not reactively.
All Channels in one place
Our client wanted a very interactive website that allowed his customers to submit support tickets or contact him via a live chatbot when they had issues.
Key gaps in existing solutions:
Reactive systems that rely on user reports
Rich player behavior timelines and summaries
Clean, intuitive moderation tools built for modern teams
Human-in-the-loop workflows that increase trust and transparency
Impact
After launching GGWP across two beta games, moderation speed increased by 63%, enabling teams to respond faster and more effectively. Moderator satisfaction rose significantly, driven by clearer workflows and greater control. Most notably, toxic player reports dropped by 37%, highlighting the power of proactive, real-time interventions.
The Take Away
This project taught me how to design ethical AI systems that support—not replace—humans. It also showed the importance of context in UX: the right data, at the right moment, with the right tone.
Designing GGWP meant building a tool, a system, and a voice—all aligned toward making games better for everyone.
“Working on GGWP taught me how to translate complex AI into clean, actionable UX — without overwhelming the player.”
OTHER WORK
Claim Your Free 3 Months
Get 3 free months of Framer Pro Yearly Subscription when upgrading your plan with the code partner25proyearly
Sign Up Now