Transparency Note
Nexorev is a solo-founder, pre-revenue venture in pilot stage. This article is not a customer case study and does not describe a deployed hotel result. It explains how Mustafa Bilgic tests the pricing logic before PMS-integrated pilots.
What Is Being Tested
The current backtest uses public North Italy tourism and economic signals, synthetic PMS-like booking curves, and a local hotel-market fixture stored as nexorev_north_italy_hotel_market_data.json. The goal is to test whether the model can forecast occupancy and recommend rate changes without pretending that Nexorev already has customer traction.
Model Inputs
- Demand: seasonal arrivals, local tourism trends, day-of-week effects, holidays, and event pressure.
- Hotel context: occupancy, lead time, booking pace, room type, cancellation risk, and length of stay.
- Market context: competitor set price bands, macroeconomic indicators, weather, and regional tourism data.
Current Backtest Metrics
- Occupancy forecast MAPE: 9.8% on the public-data fixture.
- Occupancy forecast RMSE: 6.4 percentage points.
- RevPAR lift simulation: +7.6% versus a static-rule baseline.
These figures are useful for diligence, but they are not production hotel outcomes. Real performance will depend on PMS integration quality, rate approval workflow, market volatility, and how each hotel chooses to apply recommendations.
What Comes Next
The next step is founder-led pilots with independent hotels in North Italy. Each pilot should start with a baseline period, a documented approval workflow, and post-pilot reporting that separates model recommendations from confirmed commercial outcomes.