Revenue Management
Demand Forecasting
Predicting future demand by segment and date to drive pricing, restrictions, and staffing.
Academic definition
Demand forecasting is the practice of predicting future room demand by date and segment to inform pricing, inventory restrictions, distribution, and staffing decisions. Forecasts draw on historical booking patterns, current pace and pickup, the booking window, known events, seasonality, market conditions, and competitive intelligence. They are typically produced at the segment level — transient, corporate, group, and channel — because each behaves differently and is priced and controlled differently. A forecast of unconstrained demand, the demand that would exist absent capacity limits, is particularly valuable for pricing scarcity on high-demand dates. Forecast accuracy is measured against actuals and refined continuously as a date approaches and more booking data accumulates. Demand forecasting underpins yield management and dynamic pricing: without a credible view of future demand, rate and restriction decisions become reactive. Traditionally the forecast targets demand volume and revenue rather than the profitability of the demand predicted.
Worked example
A hotel forecasts a near sell-out for a conference week; segment-level detail shows the demand skews to high-commission OTA channels, prompting early direct-rate pushes to reshape the mix before the leaks lock in.
The ProfitPAR™ lens
A demand forecast is only as useful as the decision it drives, and for decades it has driven the wrong one: forecasting how many rooms will sell rather than how much profit they will keep. Two demand scenarios with identical room counts can carry completely different margins depending on the channel and segment mix behind them. ProfitPAR™ extends forecasting from volume to profit — forecasting not just that a date will fill, but through which channels and at what retained margin, so pricing and restriction decisions optimise the kept rupee. A forecast that predicts a sell-out via high-commission OTAs and one that predicts the same sell-out direct should trigger very different actions. Forecast the profit, not just the heads.
Forecasting how many rooms will fill is worth little without forecasting what they will keep.
Measure what each room actually keeps.
ProfitPAR is offered as a private revenue practice, by invitation. Tell us about your hotel to begin.
Request an InvitationWorked examples use illustrative figures in Indian Rupee (₹) with lakh/crore conventions. They are composite and representative, not drawn from any single property. ProfitPAR™ is a trademark of Sibasish Misra.