How TripHunt Finds Cheap Flights

Every 60 minutes, TripHunt automatically scans 431 flight routes from 8 UK airports to 55 destinations worldwide. Here's exactly how it works — and why it catches deals that manual searching misses.

The data source: Google Flights via SerpAPI

TripHunt pulls live price data from Google Flights via SerpAPI — the same prices you'd see if you searched manually, but automatically, across hundreds of routes simultaneously. This means prices are always real and bookable, not estimated or scraped from outdated caches.

The 90-day rolling average

For every route, TripHunt maintains a rolling 90-day price history. Every price checked is stored to build a statistical baseline for that specific origin–destination pair. When a new price is detected, it's compared to this baseline using a z-score algorithm — how many standard deviations below average is this price?

The deal scoring system

Each price receives a deal score based on how far below the baseline it is:

• 🔥 Exceptional Deal — 50%+ below the 90-day average. Possible error fare.
• ⚡ Great Deal — 35–50% below average. Genuinely cheap, worth booking.
• ✓ Cheap — 20–35% below average. A solid saving vs the typical price.

The cache architecture

To serve thousands of users without hammering the Google Flights API, TripHunt uses a four-layer cache: in-memory (L1) → Upstash Redis (L2, 1-hour TTL) → Supabase database (L3) → live API (L4, last resort). Over 95% of requests are served from cache, keeping the site fast and API costs predictable.

Price alerts

When you set a price alert on TripHunt, you're telling us your target price for a specific route. Every time the background scanner finds a price at or below your target, you get an instant push notification (Premium) or next-run email (free). Error fares — prices more than 50% below average — always trigger an immediate alert for all users watching that route.