🎉 Festival Website Case Study

UX Research · UI Design · Cross-functional Collaboration · Analytics Planning

This repository documents the process and artifacts from my work on the CAU school festival website project.
My contributions focus on UX research, UI design, design-to-development collaboration, and analytics strategy planning.

🔗 Link to site : https://lucaus.co.kr

👭 Team Repository : https://github.com/LUCAUS2025

---

📌 Overview

This project was a university festival website built collaboratively with a development team.
I participated as a UX/UI designer, conducting research, redesigning flows and layouts, and supporting the implementation phase with structured design handoffs and visual QA.

While development & full analytics are handled by the engineering team,
I am preparing an extended analysis using GA4 and server log data as a future enhancement.

---

🧩 Project Scope

• UX Research (pain points, flows, heuristic checks)
• UI Design (Figma components, layout, visual system)
• Design-to-Development Handoff
• UX QA + iteration with developers
• Analytics Planning (GA4 + log analysis) — In Progress

---

🎨 UX & UI Contributions

1. UX Research

• Identified user needs and barriers from previous festival websites.
• Conducted lightweight user interviews & synthesized pain points.
• Created improved user flows for navigation and ticketing information.
• Performed heuristic evaluation & usability checks.

2. UI Design

• Built a complete Figma component system (colors, typography, spacing rules).
• Designed key pages: home, schedule, map, booth sections.
• Delivered responsive layout guidelines for desktop/mobile.
• Created assets for marketing and promotions as needed.

---

🤝 Collaboration With Developers

Although I did not contribute to coding, I worked closely with the dev team by:

• Preparing structured design specs & handoff files.
• Reviewing implemented pages and checking alignment with the design system.
• Communicating layout issues, spacing rules, and visual polish needs.
• Capturing before/after screenshots for reference.

---

📊 Analytical Works - Festival Website Log Analysis

Understanding User Engagement and Momentum in a Large-Scale Offline Event

This project analyzes server access logs from a university festival website to understand how users interact with a time-sensitive, high-traffic service.
The goal was not just to visualize traffic, but to identify high-value user segments, engagement patterns, and drop-off points that directly inform product and UX decisions.

By combining behavioral data with event context, this analysis reveals how momentum is created—and lost—during short-lived but intense user journeys.

Context

• Event: University Festival (multi-day, high concurrent traffic)
• Platform: Official festival website (mobile-first usage)
• Role: UX/UI Designer & Data Analyst
• Objective: Improve engagement, retention, and post-action continuity

Key Questions

• Which users demonstrate the highest engagement and intent?
• How do interaction patterns differ between casual visitors and active participants?
• Where does user momentum peak—and why does it suddenly drop?

Key Insights

1. High-Value Users Show Distinct Behavioral Patterns

Users who participated in the Stamp Tour (gamified feature) recorded
3.5× higher interaction rates than general visitors.

These users voluntarily engaged with multiple booths and features, indicating:

• High intrinsic motivation
• Strong alignment with festival goals
• Potential as a “VIP segment”

2. The “Completion Cliff” Problem

Despite their high engagement, Stamp users showed an abrupt drop-off immediately after task completion.

Finding:
The interface treated stamp completion as the end of the journey, rather than a transition point.

Interpretation:
User momentum was built successfully—but not sustained.

3. Momentum Is a Design Asset

Behavioral patterns suggest that post-completion moments are critical opportunities:

• Users are most emotionally invested
• Willing to explore additional actions
• Highly receptive to recommendations

Failing to guide users at this moment results in lost engagement potential.

Data & Methodology

• Data Source: Server access logs
• Tools: Python (Pandas, NumPy), Jupyter Notebook
• Methods:
  • Session-based behavior grouping
  • Interaction frequency comparison
  • Time-sequence analysis of user actions
• Visualization: Matplotlib / Seaborn

Design & Product Proposal

“Keep the Momentum” Strategy

Introduce contextual prompts immediately after stamp completion, such as:

• Nearby food trucks
• Popular booths
• Limited-time events

This reframes completion as a gateway, not a conclusion—extending user engagement beyond the original goal.

Why This Matters

This project demonstrates how:

• Log data can reveal latent user intent
• UX decisions directly impact behavioral continuity
• Short-lived services still benefit from long-term engagement thinking

The findings are applicable to any event-based or campaign-driven digital service.

Tech Stack

• Python
• Pandas / NumPy
• Jupyter Notebook
• Matplotlib / Seaborn

---

✨ Contact

lucyroh529@gmail.com
Portfolio site