A hotel that is constantly jam-packed with guests is not automatically a profitable one. The goal of a successful hotel is not necessarily to achieve 100 percent occupancy, but rather to optimize revenue. And developing a way to more accurately forecast daily demand is a key operational challenge for hoteliers.1
Today’s hospitality landscape is characterized by increasing room supply and tight margins. In addition, stiff competition makes it imperative for hotel revenue managers to continuously adjust strategies to remain relevant. Predicting how many customers your hotel will have on any given night, anticipating the sales of products and services, and understanding how those bookings break down into business segment and respective profitability help hoteliers develop effective business growth strategies and make the most of revenue opportunities.
Unconstrained vs. Constrained Demand
Your operations team benefits from the use of a constrained demand forecast for supply purchases and planning purposes. However, to develop an effective revenue management strategy you must understand your forecasted unconstrained demand.2 Unconstrained demand is an estimate of the total demand for your hotel, ignoring rate and capacity constraints – essentially, how many rooms could be sold on a given day if your hotel had unlimited inventory. It’s only through an analysis of total demand that you can optimize your profits, capturing yield opportunities at those points when demand exceeds supply and where it becomes necessary to turn away customers that are less profitable in favor of those who may be more willing to pay higher rates.
What’s Powering Your Forecast?
By relying on science to power your demand forecasting – as opposed to using legacy methods – you leverage sophisticated processes and produce intelligent insights that reduce uncertainties and better equip managers to optimize financial results and business performance.
A major pillar supporting a demand forecast is accurate data. A good science-based forecast not only captures valuable historical data, but continuously updates real-time data as well, examining broad factors such as seasonality, market conditions, competitive landscape, length of stay patterns, pricing elasticity, geographical data, booking pace, and lead time. And it should automatically detect outliers and anomalies as well. By incorporating the right data – and avoiding “noisy data” that dilutes the reliability of a demand forecast – revenue managers can refine their approach and adjust pricing with greater precision, transforming data into accurate and actionable revenue-enhancing strategies.
Machine learning (ML) is fueling next-generation solutions, allowing revenue managers to improve forecast performance by adapting quickly to changes in the business or marketplace. Scientific models enhanced with ML have the ability to automatically learn and improve from experience without being explicitly programmed, and are especially valuable in forecasting applications. Examples of ML used in forecasting include:
- Machine learning (ML) is fueling next-generation solutions,3 allowing revenue managers to improve forecast performance by adapting quickly to changes in the business or marketplace. Scientific models enhanced with ML have the ability to automatically learn and improve from experience without being explicitly programmed, and are especially valuable in forecasting applications Examples of ML used in forecasting include: Anomaly detection mechanisms that identify and remove booking or rate observations that could skew historical statistics and degrade the accuracy of the forecast
- Advanced competitor configuration algorithms that automatically choose a best-fit competitive set based on relationships in pricing actions among a subject hotel and its competitors
- Customer segmentation analysis used to predict customer value
A good forecast is granular enough to capture the nuances of market segmentation, understanding that each customer segment reflects different preferences, booking trends and patterns, and purchase intentions. A family taking a leisure vacation will have different requirements than a business traveler attending a convention. Predicting demand and requirements for each of these customer segments helps you better target your marketing,4 budget your operational expenses, and achieve an optimal business mix – knowing at any given time which business to take and which to turn down.
Incorporating events and promotions into your demand forecasting process is crucial for accuracy in your revenue management strategy. Beyond considering major holidays, you want to add in significant events that are specific to your city or property. Events typically mean a big influx of traffic and demand to your area, and when they’re omitted from your forecast, you not only don’t have a full picture of demand, but you’ll be unable to develop an accurate pricing strategy. By separately forecasting events and promotions, you’ll determine how far in advance guests book rooms around these two things, providing you with a more accurate analysis of future demand. And you’ll also avoid letting them affect your day-in/day-out forecast.
Add Science to Group Forecasting
Group business can markedly impact hotel profits.5 And without forecasting for groups, you can’t effectively yield nor truly optimize your business. Group demand forecasting affects transient business and business from other groups, as well as ancillary revenue generators.6 It allows revenue managers to accurately assess the financial impacts7 of accepting a group during periods of high and low transient demand.
Most RMSs don’t have the capability to forecast for group business. And many hotels still utilize traditional methods that rely on transient displacement analysis. This type of analysis, however, only projects what is required in group revenue to break even with displaced transient bookings, and it doesn’t consider the significance of one group displacing another group.8 It also doesn’t factor in how to optimize a property’s meeting space.
Inaccurate group demand forecasting risks significant revenue losses. When group forecasts are too high, a property loses profits on rooms that could have been sold to high-value customers but instead lie empty. And if group business is forecasted too low, a property risks accepting transient business at suboptimal rates, or even booking to overcapacity. When forecasted accurately, hotels get a clear picture of the constraints that groups produce on sleeping rooms, providing opportunities for hotels to capture the appropriate rates from transient customers or other groups.
By choosing an RMS that incorporates science into group demand forecasting, you have the sophistication and flexibility you need to handle the complexities of predicting demand and establishing pricing for group business. Next-generation group forecasting provides a comprehensive daily analysis of allocation and pricing for both sleeping rooms and function space. It also incorporates the differences between transient and groups in terms of booking windows – understanding that groups book earlier in the booking horizon, while transient customers book closer in. A scientific solution can also handle the inherent fluctuations involved in group demand – preventing potential revenue losses by dynamically updating rates across market segments as it evaluates whether anticipated demand from RFPs actually occur.
Demand forecasting is germane for revenue management in the hospitality industry. An RMS with demand forecasting capabilities backed by science significantly improves accuracy – leveraging complex algorithms and extensive data sets that guide hoteliers in making fact-based decisions that lead to substantially higher profits. By integrating pricing with market demand intelligence, you can achieve the best business mix across both transient and group segments. Demand forecasting provides a strong foundation for an effective growth strategy, and it leads to a number of downstream positive effects on metrics and bottom-line income.
1Ampountolas, Apostolos. “Forecasting Hotel Demand Uncertainty Using Time Series Bayesian VAR Models – Apostolos Ampountolas, 2018.” SAGE Journals, 4 Oct. 2018, journals.sagepub.com/doi/abs/10.1177/1354816618801741.
2Thompson, Gary. “Revenue Management Forecasting Aggregation Analysis Tool.” Cornell Hotel of School of Administration, Center for Hospitality Research Tools, Sept. 2009, scholarship.sha.cornell.edu/chrtools/21/.
3Boulton, Clint. “Starwood Taps Machine Learning to Dynamically Price Hotel Rooms.” CIO, CIO, 13 May 2016, www.cio.com/article/3070384/starwood-taps-machine-learning-to-dynamically-price-hotel-rooms.html.
4Lynn, Michael. “Segmenting and Targeting Your Market: Strategies and Limitations – Lynn – 2012 – Wiley Online Books.” Wiley Online Library, The Cornell School of Hotel Administration on Hospitality: Cutting Edge Thinking and Practice, 2 Jan. 2012, onlinelibrary.wiley.com/doi/abs/10.1002/9781119200901.ch23.
5White Paper: Innovative Forecasting & Demand Growth Strategies for Fueling Hotel Bookings. Cvent, Social Tables, Rainmaker, cdn2.hubspot.net/hubfs/3226550/eBook%20Fueling%20Hotel%20Bookings%20-%20Rainmaker%20and%20Social%20Tables.pdf.
6Morse SC, Beckman E (2016) A Decision Model for Hotel Revenue Management Displacement Analysis for Transient Room Demand vs. Group Room Demand. Journal of Hotel & Business Management 5: 141. doi: 10.4172/2169-0286.1000141, https://www.longdom.org/open-access/a-decision-model-for-hotel-revenue-management-displacement-analysisfor-transient-room-demand-vs-group-room-demand-2169-0286-1000141.pdf.
7Willmore, Simon. “12 Advantages of Revenue Management Technology.” Travel Daily, 71 Sept. 2017, www.traveldailymedia.com/12-advantages-revenue-management-technology/.
8“Groups & Total Hotel Revenue Management.” HSMAI, 2016, www.hsmai.org/knowledge/summary.cfm?ItemNumber=24710.
About the Author
Dan Skodol is Vice President of Revenue Analytics at Rainmaker. Dan came to Rainmaker with over ten years of revenue management experience in gaming, hotels, multifamily real estate and airlines. He is responsible for researching and designing enhancements and innovations within Rainmaker’s hospitality product suite.
Dan previously held Director of Revenue Management roles for two casino organizations in Atlantic City and Archstone Communities. He holds a BA from Yale University and a Master of Management in Hospitality from Cornell. Dan and his wife reside in Denver, CO with their four-year-old son and enjoy skiing, hiking and travel.
The Rainmaker Group, a Cendyn company, is the premier provider of revenue and profit optimization solutions to the hospitality industry. Rainmaker’s intelligent profit platform helps hotels, resorts and casinos optimize revenue, drive increased profitability, save valuable time & outperform competitors. As part of Cendyn, Rainmaker offers a complete set of software services for the industry, aligning marketing, sales and revenue teams to optimize their strategies and drive performance and loyalty across their business units. To learn more about Cendyn and its suite of hotel revenue management and profit optimization solutions, visit LetItRain.com or www.cendyn.com.
Cendyn is the leading innovative cloud software and services provider for the hospitality industry. With a focus on integrated hotel CRM, hotel sales, and revenue strategy technology platforms, Cendyn drives sales, marketing and revenue performance for tens of thousands of hotels across the globe. The Cendyn Hospitality Cloud offers a complete set of software services for the industry, aligning marketing, sales and revenue teams to optimize their strategies and drive performance and loyalty across their business units. With offices in Boca Raton, Atlanta, Boston, San Diego, London, Munich, Singapore, Sydney, Bangkok and Tokyo, Cendyn proudly serves clients in 143 countries, delivering over 1.5 billion data-driven, personalized communications on behalf of their customers every year. For more information on Cendyn, visitwww.cendyn.com.