Decisions
Occupancy and travel: rooms, seats, and the booking curve
A hotel night and a flight seat perish at midnight. The forecast is pickup along a booking curve, not only yesterday’s occupancy. Price is a lever you set; remaining capacity is the constraint. Those two belong in the same conversation.
Updated Aug 26, 2026·8 min read
The product expires
A hotel night and a flight seat perish at midnight. Empty is a loss you cannot inventory away. The forecast is pickup along a booking curve, not only yesterday’s occupancy. Price is a lever you set. Remaining capacity is the constraint. Those two belong in the same conversation.
Pickup along a booking curve
The booking curve is a series
On-the-books by D-90, D-60, D-30 is a panel: each stay date has a path of pickup. Forecasting only the stay-night occupancy throws that away. Revenue management systems know this. General ML platforms do not, unless you build it.
Events and seasons
Concerts, holidays, school breaks, a conference that is still tentative. Known blocks are covariates. Speculative groups are scenarios. Do not let a tentative bus tour look like organic pickup. Calendars are geography-specific; a US holiday file on a Gulf property will lie.
- Forecast room-nights or passengers, then divide by capacity in the RMS. Occupancy % moves when you take rooms out of service.
- Price as a known-ahead driver or a what-if, not leaked from the cleared rate.
- Competing events in the city are only features if you will have them at origin.
Price in, occupancy out
Forecast occupancy or passengers in Predict.ai, with price and events as drivers. Then ask what-if on a fare or a concert weekend. The live path updates as pickup lands — the curve is not a quarterly export. That is the gap with a Canvas model that trained once on a flattened table and never saw this week’s on-the-books.
FAQ
- Should I forecast occupancy or room-nights?
- Room-nights (or passengers) if capacity changes. Occupancy % moves when you take rooms out of service. Forecast the demand; divide by capacity in the RMS.
- How do groups and contracted blocks work?
- Known blocks are a covariate. Speculative groups are a scenario. Do not let a tentative bus tour look like organic pickup.
Keep going
Guide
Price, promo, and lift
Lift is what happens to the path when you change a price or a promo you actually control. It is a scenario on a model that has seen similar moves — not a 2.4 elasticity copied from a textbook onto a fare class, a kWh tariff, or a SKU.
Guide
Holidays and calendar effects
A calendar is the cheapest accurate feature in the building. Christmas, payday, school terms, a 4-day week, Ramadan, Chinese New Year — if the model has to rediscover those from residuals, you are wasting folds on a date table.
Guide
Capacity planning
A plant, a ward, a classroom, a network link — capacity is a ceiling. The forecast is whether you will hit it, and when. Over-build is capital. Under-build is a queue, a blackout, or a denied admission.
Use case
Hotel occupancy forecasting
Forecast occupancy, arrivals, cancellations, and room revenue by property and segment.
Use case
Passenger demand forecasting
Anticipate passengers by route, station, flight, departure, and fare class.