Funny Tourism

Funny Tourism serves two sides: B2C (international customers booking packages, day tours and transfers) and B2B (foreign agencies needing pricing and commission tracking). The needs run in parallel: multi-format package management (3/4/5-star hotel tiers vs land-only), package-independent day tours and transfers, and an end to manually re-typing partner PDF brochures.

A single package entry took 30 minutes — the biggest brake on growing a 25 destinations × 33 cities = tens-of-thousands of pages SEO surface.

A dual package model (WITH_HOTEL vs LAND_ONLY): some agencies want hotels included, some want transfers + tours only. Two architectures, one app, broader customer base. Itinerary + pricing + hotels are JSON-based — flexible and tour-customizable.

AI-powered PDF extraction (OpenAI GPT-4o): in /admin/packages/add the operator drops a partner brochure and the form auto-fills. Manual data entry: 30 min → 30 sec per package. That alone saves 30+ hours per operator each month.

Dual authentication (customer + admin token isolation — same pattern as ATP), a B2B Agent Portal (Agent + AgentLead + CommissionPayment tables), a Custom Itinerary Builder, wishlist + searched-cities tracking, a blog engine (15 SEO posts) and PDF brochure generation with download logging.

Catalog: 72 packages, 24 day tours, 22 transfers, 25 destinations, 33 Türkiye cities, 15 blog posts. Customer base: 62 registered users, 12 bookings, 24 inquiries, 9 B2B agents, 325 agent leads in the pipeline. 363 cumulative PDF downloads.

Key trend: PDF downloads grew 9× in April 2026 (March 31 → April 279). After a quiet Q1, that's a fresh-traffic signal — likely organic + AI-search citations. It's the case for tracking AI-search citations carefully.

Replicable insight: AI PDF extraction kills manual data entry. The pattern transfers to any tour operator or DMC. PDF download volume is one of the earliest user-facing signals of AI-search visibility.