Zero to commercial success in a year

Designing an AI-powered sports analytics platform that makes complex data understandable, trustworthy, and engaging for everyday fans.

Overview

GameScript.ai helps sports fans navigate advanced analytics through AI-generated predictions and expert insights.

I led design from early concept through launch, shaping the product vision, brand, UX foundations, and visual language for a data-heavy consumer experience.

Product Vision

Design and deliver an intuitive analytics experience that empowers sports fans to:

• Understand game predictions and player performance through AI insights
• Access curated content from top sports analysts
• Explore deep data without technical expertise
• Subscribe easily and stay engaged throughout the season

Product Challenges

Problem Space

Sports analytics data is incredibly dense and complex. Fans want:

• approachable insights, not raw data
• contextual narrative around predictions
• trustworthy recommendations that explain why specific insights matter

Design challenge: Translate extensive AI predictions and human expert insights into digestible, engaging UI - without overwhelming the user.

Consumer Challenges

• Users unfamiliar with sports analytics terminology
• Need to balance simplicity for casual fans with depth for advanced users
• Subscription conversion requires clear perceived value

Design Strategy

Principles

1. Clarity through visuals
   Use visual hierarchy and clear micro-animations to make statistics approachable.
2. Narrative first
   Combine text and UI to tell why an insight matters (e.g., “AI projects this player's scoring likelihood at 78% because…”).
3. Predictive transparency
   Explain AI confidence scores so users trust rather than distrust “black-box” suggestions.
4. Engagement loops
   Push notifications and home feed highlight new insights, trending players, and expert picks.

Brand & Visual System Foundations

While the product vision was clear, the brand lacked a cohesive identity that could communicate both analytical credibility and emotional energy.

My goal was to create a visual system that felt:

• Intelligent but not overly technical
• Confident without feeling like a sportsbook
• Data-driven yet accessible to everyday fans

Before designing visual assets, I worked with product leadership to define core brand attributes:

• Insightful — grounded in real data and AI confidence
• Transparent — predictions explained, not hidden
• Dynamic — responsive to live sports momentum
• Fan-first — built for enthusiasts, not analysts

These principles guided every visual and interaction decision.

Colour Strategy

The palette was designed to balance energy and authority:

• A deep, near-black foundation to evoke sports broadcast environments
• High-contrast accent colours to represent confidence scores and performance momentum
• Semantic colour roles (confidence, risk, trending, neutral) embedded directly into data visualisation components

Colour was not decorative — it carried meaning across predictions, probability states, and outcomes.

Typography & Hierarchy

Given the density of sports data, typography became a functional tool.

I introduced:

• A bold headline style for predictions and key metrics
• A highly legible numeric treatment for odds, percentages, and confidence scores
• A structured hierarchy that reduced cognitive load when scanning data-heavy screens

The result was a system where users could scan insights quickly without feeling overwhelmed.

Illustration & Motion

To humanise the AI layer, we avoided abstract “robotic” visuals and instead focused on subtle motion and contextual illustrations.

• Micro-interactions reinforced confidence changes and status updates
• Motion was used to communicate live updates and momentum shifts
• Transitions were fast and deliberate to mirror the pace of sports

This created a product that felt responsive and intelligent — not static.

Feature Highlights & Design Decisions

Discovery Screen

Design goal: Meet users with a digestible overview of forecasts for today's games.

UX decisions:

• Top games/events carousel with quick prediction scorecards
• Trend highlights (e.g., stat momentum, hot players)
• Content cards linking to deeper projections

Impact: Users can scan at a glance what's happening today before drilling deeper.

AI Picks Interface

Design goal: Surface AI-generated picks with confidence scores and rationale.

Key elements:

• Confidence score visualized (0-100, with clear color gradients)
• Explanation layer: short text explaining why this pick is span
• Interactive stat drill-downs: filter by sport, league, or prop type

Design trade-offs: Avoid overwhelming users with too many knobs - offer sensible defaults but allow power users to dive deep.

Analyst Picks Integration

GameScript partners with social analysts (e.g., top sports handicappers) to provide curated picks.

Design approach:

• Separate tab for human picks vs AI picks
• Analyst profile cards with credibility signals (followers, win rates)
• Options to “Follow” favorite analysts

Impact: Bridges human curation with data science, increasing emotional engagement.

Mobile App Experience

Given a mobile-first audience, the app design focused on:

• Real-time alerts for new predictions
• Integrating both AI insights and hand-curated signals
• Smooth navigation between sports categories

Subscription UI note: Clear pricing tiers, trial CTA, and feature comparisons to improve conversion.

Key Learnings

1. Education fuels retention
   Sports fans adopt complex analytics only when the product explains why an insight matters.
2. Trust through transparency
   Visual explanations of AI confidence and rationale were pivotal to user belief in the data.
3. Balance narrative and data
   Users respond best when actionable insights are paired with narrative context — not just scores or numbers.

Other Case Studies