Demand Forecasting Basics

What Is Demand Forecasting?

Demand forecasting is the process of predicting how many guest rooms a hotel will sell on each future night. An accurate short-term forecast — typically covering the next 7 to 30 days — lets a revenue manager set competitive rates, schedule the right number of staff, and avoid the twin mistakes of selling out too cheaply or leaving rooms empty. Every department in a hotel relies on the forecast: housekeeping uses it to plan room assignments, food and beverage uses it to order supplies, and the front office uses it to prepare for arrivals and potential walk-ins.

Three Inputs Every Forecast Needs

1. Historical Occupancy Data

The most reliable starting point is what has already happened. By looking at the same days in prior weeks, months, or years, you establish a baseline. The most common calculation is the average occupancy by day of week — for example, averaging every Tuesday over the past 30 to 90 days to predict next Tuesday. Hotels also track year-over-year trends to spot seasonal patterns: a beach resort sees high summer occupancy; an urban business hotel may peak mid-week and soften on weekends.

2. Booking Pace (Pickup)

Booking pace describes how quickly reservations are accumulating for a future date compared with a reference period. A pickup report shows reservations on hand today versus the same number of days before arrival in a prior period. If next Friday already has 80 rooms booked 14 days out but the historical norm at this point was only 55 rooms, the pace is running ahead — a signal of strong demand. If pace is running behind the norm, that is a soft period and may call for promotional rates or marketing action. Reading pace trends daily is one of the most valuable habits a revenue manager can develop.

3. Demand Generators

External events shift demand away from historical patterns. Demand generators include:

• Local events — conventions, sporting events, concerts, festivals, graduations, and government meetings can fill a market overnight.
• Corporate travel patterns — a factory opening or a company headquarters nearby creates predictable mid-week demand.
• Competitor activity — if a nearby hotel closes for renovation, its displaced guests need somewhere to go. If a new hotel opens, supply increases and your share of demand may shrink.
• Holidays and school calendars — leisure travel rises around public holidays and school breaks.

Always cross-reference your historical data with a community events calendar before finalizing any forecast number.

Reading a Booking Pace Report

A typical 14-day booking pace report lists each arrival date in a row. Columns show: rooms on the books today, rooms on the books at the same lead time in a prior comparable period, the pickup difference, and sometimes an unconstrained demand estimate. When you scan the report, look for:

• Dates with pace running 10% or more ahead of prior period — consider holding rates or reducing discounts.
• Dates with pace running noticeably behind — investigate why. Is there a competing event in a nearby city drawing travelers away? Has a corporate account gone quiet?
• Unusual spikes — a sudden jump in single-night bookings may indicate a local event you had not yet identified.

Calculating Average Occupancy by Day of Week

To build your baseline, pull at least 30 days of historical data — more is better. Organize figures by day of week and calculate the mean for each day. This smooths out one-off anomalies and gives you a reliable starting point. You will then adjust upward or downward based on pace and known demand generators.

Drafting a 7-Day Forecast

A practical 7-day forecast follows this sequence:

1. Enter the historical average occupancy for each day of the week.

2. Note the current rooms on the books for each date from the booking pace report.

3. Estimate remaining pickup — how many more reservations are likely to arrive before the date based on historical booking patterns.

4. Add or subtract based on identified demand generators (events, competitor closures, holidays).

5. Arrive at a forecasted occupied rooms number. Divide by total available rooms to express as a forecasted occupancy percentage.

6. State your assumptions in writing so anyone reading the forecast understands your reasoning.

Defending Your Forecast

A forecast is only as good as the logic behind it. When presenting to peers or managers, be prepared to explain every adjustment. If you raised a Thursday forecast above historical average, name the specific event driving that decision. If you held a weekend forecast flat despite a nearby festival, explain why you believe the hotel's location or segment mix limits its benefit. Clear, data-backed reasoning builds credibility and makes it easier to update the forecast as conditions change.

A Note on Accuracy and Integrity

Forecasts affect staffing levels and labor costs — decisions that directly impact other workers' livelihoods. Never inflate a forecast to meet a manager's preference or downplay a soft period to avoid difficult conversations. Honest, well-documented forecasts support sound business decisions and are a mark of professional integrity in revenue management.