GAK Ticket Tracking Homepage Improvements

Today I made significant improvements to the GAK ticket tracking dashboard. This system monitors ticket sales for Grazer AK 1902 football matches and generates a dashboard that updates every 5 minutes via cron.

The gak-parsers repository contains scripts that:

• Fetch ticket data from GAK's ticket API
• Store historical data in SQLite
• Generate visualizations and statistics
• Run automatically every 5 minutes

New Features Added

Sales Velocity Tracking

Added real-time sales velocity showing:

• Tickets sold in the last day
• Tickets sold in the last hour
• Tickets sold in the last 10 minutes

This helps identify sales surges and patterns leading up to match day.

Capacity Percentage Display

The dashboard now shows stadium capacity utilization based on a 15,000 seat capacity, accounting for:

• Regular tickets
• Season tickets
• VIP and sponsor allocations

Past Matches Section

Implemented a comprehensive past matches view showing:

• All matches from the current season (July onwards)
• Mini sparkline graphs for each match showing sales over time
• Top 3 performers highlighted with badges and red borders
• Rankings based on final ticket sales
• Season summary with match count and average attendance

UI/UX Improvements

• Responsive grid layout: 4→3→2→1 columns based on screen width
• Cleaner titles: Show only away team names (removes "GAK 1902 :" prefix)
• Uniform card heights: Flexbox layout ensures consistent sizing
• Last updated timestamp: Shows when data was last fetched
• Team color scheme: Uses GAK's red (#d9534f) throughout

Technical Implementation

Modular Architecture

The codebase was refactored into a clean modular structure:

tickets/
├── lib/
│   ├── db.py      # Database operations
│   ├── api.py     # API fetching
│   └── graph.py   # Matplotlib graph generation
├── templates/
│   └── ticket-html.tmpl  # Jinja2 template
└── ticket-fetch.py        # Main cron script

Key Technologies

• Python 3 with virtual environment
• SQLite for historical data storage
• Matplotlib for graph generation (Agg backend for cron)
• Jinja2 for HTML templating
• python-dateutil for timezone-aware datetime handling
• requests for API calls

Data Corrections

Added a helper function to apply corrections for known ticket reporting issues in the API data, ensuring accurate historical comparisons and visualizations.

Challenges

The most complex part was implementing fair comparisons between matches at equivalent time points (e.g., "how did this match sell 48 hours before kickoff compared to others"). This required careful timezone handling and database queries to find comparable data points.

Ultimately, the prediction feature was removed due to data quality issues, but the historical comparisons and visualizations remain valuable for tracking sales patterns over the season.

Result

The dashboard now provides a comprehensive view of ticket sales trends, helping identify which matches generate the most interest and how sales patterns evolve throughout the season. All while maintaining the no-JS philosophy and updating automatically every 5 minutes.

Source code available at: https://github.com/Crap-Solutions/gak-parsers