DEPLOYMENT PLAN
Deployment & Community Impact
How HargaHantu is deployed and how we measure community benefit.
Deployment Architecture
FrontendNext.js on Vercel — Global Edge Network, automatic SSL, preview deployments
DatabaseSupabase Postgres — Managed database with Row Level Security, automatic backups
StorageSupabase Storage — Private bucket for receipt images with signed URLs
AIOpenAI API — Vision-based receipt extraction (pay-per-use)
Deploy Steps:
- Create Supabase project, run SQL migration
- Push to GitHub, import in Vercel
- Set environment variables in Vercel dashboard
- Deploy — live in under 5 minutes
Community Benefit Metrics
HargaHantu measures community impact through these key metrics:
Price Transparency Index
Number of unique item-area combinations with 5+ price points. Target: 100+ within first month. This means reliable price data for 100+ everyday items across Malaysian cities.
Savings Potential
Average price difference between cheapest and most expensive merchant for each item. Measures how much money users can save by choosing the cheapest option.
Contributor Growth
Number of unique users submitting receipts per week. Target: 50+ active contributors in month one.
Data Freshness
Percentage of tracked items with price data from the last 7 days. Target: 80%+ freshness for top 50 staple items.
Scalability
HargaHantu is designed to scale with the community:
- Vercel auto-scales with traffic (serverless)
- Supabase Postgres handles millions of price points
- Trigram search indexes ensure fast queries at any scale
- PWA with service worker caching reduces server load
- AI costs are per-extraction, scaling linearly with usage
Performance Targets
90+
Lighthouse Performance
90+
Lighthouse Accessibility
90+
Lighthouse Best Practices
90+
Lighthouse SEO