Booking cancellations contribute negatively to the production of accurate forecasts, a critical tool in the hospitality industry. To lessen this influence, hotels implement rigid cancellation policies and overbooking strategies, which in turn can negatively impact revenue and the hotel’s social reputation. To tackle the uncertainty arising from booking cancellations, we combined data from eight hotels’ bookings management systems with data from other sources (weather, holidays, events, social reputation, and online prices/inventory) to develop booking cancellation prediction models. To test these models in a real production environment and assess the models impact on business, an automated machine learning system prototype was built and deployed in two hotels. In this talk, we will present how the models and the prototype were developed and deployed, as well as the results obtained.