SHA532: Forecasting and Availability Controls in Hotel Revenue Management

Transcription

1 SHA532: Forecasting and Availability Controls in Hotel Revenue Management This course includes Five self-check quizzes Two discussions Two Ask the Expert interactives One activity An action plan to apply what you learn One video transcript file Completing all of the coursework should take about five to seven hours. What you'll learn Embrace and explain the role of forecasting in hotel revenue management Create a forecast and measure its accuracy Establish or recommend room rates that maximize profitability Apply length-of-stay controls to your hotel Course Description Successful revenue management strategies hinge on the ability to forecast demand and to control room availability and length of stay. This course, produced in partnership with the Cornell School of Hotel Administration, explores the role of the forecast in a revenue management strategy and the positive impact that forecasting can also have on staff scheduling and purchasing. This course presents a step-by-step approach to creating an accurate forecast. You'll learn how to build booking curves; account for pick-up; segment demand by market, group, and

2 channel; and calculate error and account for its impact. You will develop an action plan to apply these concepts to your own situation. Sheryl Kimes Professor of Operations Management, School of Hotel Administration, Cornell University Sheryl E. Kimes is a professor of operations management at the School of Hotel Administration. From she served as interim dean of the school, and from she served as the school s director of graduate studies. Kimes teaches restaurant revenue management, yield management, and food and beverage management. She has been named the school s graduate teacher of the year three times. Her research interests include revenue management and forecasting in the restaurant, hotel, and golf industries. She has published over 50 articles in leading journals such as Interfaces, Journal of Operations Management, Journal of Service Research, Decision Sciences, and the Cornell Hospitality Quarterly. She has served as a consultant to many hospitality enterprises around the world, including Chevys FreshMex Restaurants, Walt Disney World Resorts, Ruby s Diner, Starwood Asia Pacific, and Troon Golf. Kimes earned her doctorate in Operations Management in 1987 from the University of Texas at Austin.

3 Table of Contents Module 1: Accurate Forecasts Count 1. Module Introduction: Accurate Forecasts Count 2. Watch: Why Forecast? 3. Read: A Forecast Overview 4. Watch: Measuring Demand 5. Ask the Expert: Neal Fegan on Demand Forecasting 6. What Do You Know? 7. Get Started on Your Action Plan 8. Module Wrap-up: Accurate Forecasts Count Module 2: Pickup Forecasting and the Booking Curve 1. Module Introduction: Pickup Forecasting and the Booking Curve 2. Watch: Meet the Booking Curve 3. Read: How to Create a Booking Curve 4. Watch: Introduction to Pickup Forecasting 5. Tool: How to Perform Pickup Forecasting 6. Watch: Creating a Pickup Forecast in Excel 7. Ask the Expert: Alfonso Delgado on Types of Forecasting 8. Forecast It! 9. Reflect on Pickup Forecasting and the Booking Curve 10. Module Wrap-up: Pickup Forecasting and the Booking Curve Module 3: Forecasting Groups 1. Module Introduction: Forecasting Groups 2. Watch: Forecasting Groups, Channels, and Segments 3. Create the Group Forecast 4. Forecasting Challenges 5. Reflect on Forecasting Groups 6. Module Wrap-up: Forecasting Groups

4 Module 4: Measuring Forecast Error 1. Module Introduction: Measuring Forecast Error 2. Watch: Errors in Forecasting 3. Read: Use Absolute Value 4. Watch: Calculating Error Using MAD and MAPE 5. What Is the Error? 6. Reflect on Measuring Forecast Error 7. Module Wrap-up: Measuring Forecast Error Module 5: Setting Rates Based on Demand 1. Module Introduction: Setting Rates Based on Demand 2. Watch: Setting Trigger Points 3. Watch: Demand-Control Charts 4. Read: A Guide to Rates and Demand 5. Recommend a Rate 6. Reflect on Setting Rates Based on Demand 7. Module Wrap-up: Setting Rates Based on Demand Module 6: Managing Length of Stay 1. Module Introduction: Managing Length of Stay 2. Watch: Controlling Length of Stay 3. Read: Dos and Don'ts of Length of Stay 4. Fill Your Hotel 5. Pitfalls and Benefits of Controlling Availability 6. Module Wrap-up: Managing Length of Stay 7. Complete and Submit Your Action Plan 8. Watch: Thank You and Farewell 1.

5 Module 1: Accurate Forecasts Count 1. Module Introduction: Accurate Forecasts Count 2. Watch: Why Forecast? 3. Read: A Forecast Overview 4. Watch: Measuring Demand 5. Ask the Expert: Neal Fegan on Demand Forecasting 6. What Do You Know? 7. Get Started on Your Action Plan 8. Module Wrap-up: Accurate Forecasts Count

6 Module Introduction: Accurate Forecasts Count Accurate Forecasts Count In revenue management, your goal is to work with demand and the resources you have to maximize your revenue. Your success will depend on your understanding of the demand for your hotel services and the accuracy of your forecasts. In this module, find out why demand can be challenging to measure. Learn the subtleties of constrained and unconstrained demand and why accurate forecasting is a necessary part of any revenue management system. When you have completed this module, you will be able to relate accurate forecasting to revenue management, define unconstrained demand, and describe different ways that forecast data can be presented. Back to Module 1: Accurate Forecasts Count

7 Watch: Why Forecast? Why Forecast? How many customers will come to your hotel two weeks from tomorrow? How about two months from tomorrow? It's difficult to answer these questions without forecasts. Yet some hotels don't create forecasts. Hotel managers may say they know how many arrivals to expect, more or less, and that forecasting is too much trouble. Should you go to the trouble of creating written forecasts? Most hotels will find it difficult to reach their revenue goals without accurate forecasts. In fact, labor scheduling and purchasing are harder to do without forecasts, too. Watch this video to learn many reasons why you should forecast. Back to Module 1: Accurate Forecasts Count

8 Read: A Forecast Overview A Forecast Overview Accurate forecasts are essential to revenue management. Without a forecast, it's not possible to set availability controls wisely or to use these controls effectively to grow revenue. A good forecast is a first and necessary step in revenue management. Good forecasts are based on data. Let's take a look at some typical booking data and see how past performance provides a basis for forecast estimates. Figure 1 displays gueststays data for 100 Thursdays at a particular hotel. From this chart, you can see how actual numbers of stays vary week by week, though usually keeping close to an average value. This value can be obtained by looking over the numbers and coming up with an rough, estimated average, or it can be calculated mathematically. It can be used, along with other characteristics of the data, to forecast the number of room-nights on a Thursday in the future. Table 1 Historical Average Weekly Occupancies Forecast Week Sun 75% 48% 81% 82% 82% 81% 82% 82% Mon 76% 79% 81% 81% 83% 84% 82% 84% Tue 71% 66% 44% 61% 80% 60% 61% 61% Wed 72% 68% 68% 65% 95% 68% 65% 68% Thu 70% 58% 68% 61% 59% 71% 64% 66% Fri 71% 78% 70% 74% 75% 89% 100% 95% Sat 80% 84% 87% 90% 91% 90% 97% 94% This graph plots the actual number of guests staying on 100 Thursday nights. These data can be used to develop a probability analysis for future Thursday-night arrivals. Table 1 above contains occupancy data aggregated by day for

9 seven weeks. The manager of the hotel can use historical data like these to estimate room-nights for days and weeks in the future. The data provided here suggest something about future occupancy levels, but they also demonstrate how much these values can vary. If occupancies vary as much as they do here, how is it possible to create an accurate forecast? Forecasts are estimates. You shouldn't expect them to be 100% accurate. They tend to be more accurate when more data are available and when those data can be appropriately aggregated, or grouped. For the best results, occupancy data should always be grouped by day, but they can also be grouped by length of stay, by rate, and even by distribution channel or market segment. In addition, forecasts created for dates in the future can be made increasingly accurate as information is acquired closer to the target date. Forecast Variations Types of Forecasts Arrivals Forecast An arrivals forecast includes guests arriving on the forecast day but does not include guests arriving prior to the forecast day for multi-night stays. Room-Nights Forecast A room-night forecast includes guests arriving on the forecast day and guests arriving prior to the forecast day for multi-night stays. Levels of Forecast Detail Day of week Most forecasting is done by the day of the week, that is, a Monday forecast would be built using data from previous Mondays, a Tuesday forecast would be built using data from previous Tuesdays, and so on. Demand varies reliably by day of the week, with certain obvious exceptions (holidays, emergencies). Length of stay For length-of-stay forecasting, individual forecasts are created for each stay length (one-night, two-night, three-night, etc.). Rate class For a hotel that offers more than one rate class (for instance, rack rate, discount, or super discount), forecasts can be created for each rate class individually. Inventory type In the case of hotel-room forecasting, a forecast by inventory type might be characterized by king- or queen-size beds, private bath, or view. Aggregate Arrivals or room-nights are forecasted as one group, without regard to individual subgroups such as length of stay, rate class, or inventory type. Back to Module 1: Accurate Forecasts Count

10 Watch: Measuring Demand Measuring Demand Does your booking data present a picture of all the demand for your hotel rooms or just some of the demand? View the following video to learn how to improve your forecasts by improving your understanding of demand. As Professor Kimes explained in the video, the sum of customers booked and potential customers denied is what is known as unconstrained demand. In the context of hotel management, unconstrained demand can be defined as the number of customers the hotel could accommodate if its capacity were unlimited. It is the total demand for what the hotel has to offer in the absence of all constraints. Find out how well you understand this concept by exploring the following short scenarios. Scenario 1: The rate is closed

11 Mrs. N paid a low rate to stay at your hotel last month, and she wants to pay that rate again this month. However, demand is higher this month, and the low rate is closed. When she calls to make a reservation, she isn't happy with the rate she is quoted and she decides not to book a room. Does her call represent demand for your hotel? Yes. Your hotel has what Mrs. N desires a lower rate. Unfortunately, that rate is closed (a constraint). This event should be recorded as a denial. Scenario 2: The rate is too high Ms. S called to request a room at the senior rate at your hotel and was told that the rate was not available (a constraint). When she decided not to reserve a room, the event was recorded as a denial and Ms. S was added to the estimate of demand. Later on, Ms. S continued her search for a room using an Internet service. She happened to see your senior rate listed on your website and thought to herself "That rate is too high! I wouldn't have taken it anyway." A-ha! She should not have been counted in the estimate of demand. There really was no acceptable rate. Scenario 3: The plan is changed Mr. D called yesterday to request a room with a king-size bed, but there were no king rooms available (a constraint). He decided to book a room with a queen-size bed, instead. Your booking agent recorded this event as a room booked.

12 Now your estimate of the demand for queen-size beds includes Mr. D's request. Your estimate of the demand for king-size beds does not, however but it should. Back to Module 1: Accurate Forecasts Count

13 Ask the Expert: Neal Fegan on Demand Forecasting Neal Fegan on Demand Forecasting In this video, Neal Fegan, the Executive Director of Revenue Management for Fairmont Raffles Hotels International (FRHI), discusses the forecasts they use for revenue management. As Mr. Fegan explains, they use demand forecasts to help them drive incremental revenue. Back to Module 1: Accurate Forecasts Count

14 What Do You Know? Revenue management relies on forecasts. You can apply your knowledge of forecasting and demand in the questions of this quiz. You must achieve a score of 100% on this quiz to complete the course. You may take it as many times as needed to achieve that score. Points: 4.0 Back to Module 1: Accurate Forecasts Count

15 Get Started on Your Action Plan In this course, you will see why an accurate forecast of guest arrivals at a hotel is essential to a successful revenue management strategy. You will gain the skills needed to create forecasts and measure forecast error, as well as to increase and leverage demand using room rates and length-of-stay controls. At the end of the course, you will be required to submit an action plan that outlines how you intend to apply what you have learned. What will you need in order to create accurate forecasts? How will you measure forecast error? Will you be able to recommend room rates appropriate to levels of demand and apply length-of-stay controls? While you will draft most of this action plan at the end of the last module, you will do some initial work on it now. If you are not currently associated with a hotel or business, base your action plan on a hotel or business with which you are familiar. Instructions: Download the Forecasting and Availability Controls in Hotel Revenue Management Action Plan. Review the plan document so you are familiar with its structure. In the Key Business Problem section, write an initial draft of a business problem related to forecasting and availability that you would like to address. You can modify this later, but make an initial attempt to describe a revenue management problem. In the section labelled Notes, make notes of concepts or recommendations contained in module 1 that you think may be useful to consider when you work further on your action plan. For each item you include, indicate why you think it may be relevant to your situation. Save your work. This action plan is a course requirement, but do not submit it at this time. Before you begin: While you won't submit the document now, please review the rubric (list of evaluative criteria) that will be used when you submit the document with notes and completed plan at the end of the course. Also review ecornell's policy regarding plagiarism (the presentation of someone else's work as your own without source credit). Grading Type: Not Graded Back to Module 1: Accurate Forecasts Count

16 Module Wrap-up: Accurate Forecasts Count Accurate Forecasts Count In this module, you found out why demand can be challenging to measure. You learned the subtleties of constrained and unconstrained demand and why accurate forecasting is a necessary part of any revenue management system. Now you are able to relate accurate forecasting to revenue management, define unconstrained demand, and describe different ways that forecast data can be presented. Back to Module 1: Accurate Forecasts Count

17 Module 2: Pickup Forecasting and the Booking Curve 1. Module Introduction: Pickup Forecasting and the Booking Curve 2. Watch: Meet the Booking Curve 3. Read: How to Create a Booking Curve 4. Watch: Introduction to Pickup Forecasting 5. Tool: How to Perform Pickup Forecasting 6. Watch: Creating a Pickup Forecast in Excel 7. Ask the Expert: Alfonso Delgado on Types of Forecasting 8. Forecast It! 9. Reflect on Pickup Forecasting and the Booking Curve 10. Module Wrap-up: Pickup Forecasting and the Booking Curve

18 Module Introduction: Pickup Forecasting and the Booking Curve Pickup Forecasting and the Booking Curve Forecasting begins with reservation data. In this module, you will see how reservations-on-hand (ROH) data is charted to show the accumulation of room reservations over time. You will see how charts known as booking curves differ and when reservations tend to come in and at what rate. Typical booking curves for corporate hotels differ from those of airport hotels, and both of those differ from booking curves for resort hotels. In addition, this module will show how ROH data is used to create a very common and useful forecast, the pickup forecast. When you have completed this module, you will be able to build a booking curve and create a pickup forecast. Back to Module 2: Pickup Forecasting and the Booking Curve

19 Watch: Meet the Booking Curve Meet the Booking Curve At your hotel you may have booking data going back several years. If you could get access to that data, how would you want to use it and display it? In this video, Professor Kimes presents the booking curve, a popular type of graph used for displaying and interpreting booking data. Back to Module 2: Pickup Forecasting and the Booking Curve

20 Read: How to Create a Booking Curve How to Create a Booking Curve A hotel manager reviewing her booking data notices that she has only 20 reservations on hand for a Thursday night 5 weeks in the future. She's hoping to sell 110 rooms. Should she be concerned? It's not possible to say if she should be concerned unless you are able to ask a few questions. For instance, what occupancy level has she achieved in the past on that date or day of the week is 110 rooms a realistic goal? If it is, how many of those 110 reservations should she expect to have on hand at this point, 5 weeks in advance of the arrival date? Do most reservations tend to come in many weeks in advance or during the last week? Without answers to these questions, it's very difficult to interpret the current reservation data. One tool this manager could use is a booking curve. A booking curve is a chart of reservations on hand (ROH) for a given day, recorded at regular intervals. These intervals might be months before arrival (MBA), weeks before arrival (WBA), or days before arrival (DBA). The booking curve is a picture of the accumulation of bookings in the weeks leading up to an arrival date. Based on historical data, this curve provides information about general tendencies in booking behavior that can be very useful in forecasting future bookings. You create a booking curve using ROH data of the kind found in Table 1. That is, the ROH values must be associated with specific intervals before arrival (in this case, DBAs). Because booking curves are used in building forecasts, you should use historical data for arrival days or dates similar to the day or date you are forecasting. Typically, the booking-curve data set includes data for each week before arrival and for each day before arrival in the final week, as in Table 1. (The data-gathering days are called reading days.) These data are then plotted as in Figure 1. Table 1. Reservations Data DBA ROH

21 This graph presents the number of reservations on hand at different times. Times are charted as days before arrival (DBA). DBA = -1 refers to the day after arrival, when actual numbers are finally available. Booking curves, like forecasts, can be created for specific days of the week, rate classes, or market segments, or the data can be an aggregate of all reservations on hand. If you have enough data to work with, you may want to create an average booking curve using averages of six to eight data sets like the one in Table 1. An average booking curve may provide stronger evidence of general trends than a booking curve based on just one data set because it draws on a greater number of experiences. To return to the manager with 20 reservations, should she be concerned? If the booking curve in Figure 1 is relevant to the date in question, the answer is probably not. Twenty reservations on hand 35 days before arrival is about what she should expect. It appears that she's right on track for 110 rooms reserved on the day of arrival! Back to Module 2: Pickup Forecasting and the Booking Curve

22 Watch: Introduction to Pickup Forecasting Introduction to Pickup Forecasting Pickup forecasting is one of the most accurate forecasting methods in use, and it's simple! This method builds on your knowledge of booking data and the booking curve. Learn how to create a pickup forecast in this video. Back to Module 2: Pickup Forecasting and the Booking Curve

23 Tool: How to Perform Pickup Forecasting How to Perform Pickup Forecasting Use the Guide to Pickup Forecasting to aid you in forecasting pickup The "pickup" is the number of reservations you expect to receive but have not yet received for a particular arrival date. You can estimate pickup using historical data. Download and use this guide to arrive at a forecast for a future date of arrival using pickup forecasting. These instructions will help you to calculate the expected pickup for that date. Then you'll add the expected pickup to the reservations on hand to get the forecast. Back to Module 2: Pickup Forecasting and the Booking Curve

24 Watch: Creating a Pickup Forecast in Excel Creating a Pickup Forecast in Excel In the previous video, Professor Kimes explained how to calculate a pickup forecast. Here she demonstrates how you can create this forecast using Microsoft Excel. For many, this approach will be be easier and more accurate. Back to Module 2: Pickup Forecasting and the Booking Curve

25 Ask the Expert: Alfonso Delgado on Types of Forecasting Alfonso Delgado on Types of Forecasting In this video, Alfonso Delgado, Managing Director at Solution Development Partners, talks about how different types of forecasts can be used in different industries to achieve 100% utilization. Back to Module 2: Pickup Forecasting and the Booking Curve

26 Forecast It! You must achieve a score of 100% on this quiz to complete the course. You may take it as many times as needed to achieve that score. You will be able to see your detailed results after you submit the quiz. Points: 4.0 Back to Module 2: Pickup Forecasting and the Booking Curve

27 Reflect on Pickup Forecasting and the Booking Curve In the previous module, you began work on the action plan you will submit for evaluation at the conclusion of the course. You will now perform further work on that plan. Instructions: Open your saved Action Plan and review the work you did in the previous module. In the section labelled Notes, make notes of concepts or recommendations contained in module 2 that you think may be useful to consider when you work further on your action plan. For each item you include, indicate why you think it may be relevant to your situation. Save your work. This action plan is a course requirement, but do not submit it at this time. Before you begin: While you won't submit the document now, please review the rubric (list of evaluative criteria) that will be used when you submit the document with notes and completed plan at the end of the course. Also review ecornell's policy regarding plagiarism (the presentation of someone else's work as your own without source credit). Grading Type: Not Graded Back to Module 2: Pickup Forecasting and the Booking Curve

28 Module Wrap-up: Pickup Forecasting and the Booking Curve Pickup Forecasting and the Booking Curve In this module, you looked at how reservations-on-hand (ROH) data is charted to create the booking curve. You saw that booking curves differ in terms of when reservations tend to come in and at what rate. Further, you learned how ROH data is used to create a very common and useful forecast, the pickup forecast. Now you are able to build a booking curve and create a pickup forecast. Back to Module 2: Pickup Forecasting and the Booking Curve

29 Module 3: Forecasting Groups 1. Module Introduction: Forecasting Groups 2. Watch: Forecasting Groups, Channels, and Segments 3. Create the Group Forecast 4. Forecasting Challenges 5. Reflect on Forecasting Groups 6. Module Wrap-up: Forecasting Groups

30 Module Introduction: Forecasting Groups Forecasting Groups In this module, you will consider how to create forecasts for group business. For group forecasting, the level of demand is not predicted because the demand is known when the group is booked. business. When we create a group forecast, we try to answer the question, What will sales be, given the demand? Luckily, this question can be addressed with many of the same tools used for creating forecasts of transient In this module, you will see how hotels forecast groups and market segments to get the information they need to manage revenue. When you have completed this module, you will be able to explain why forecasting is done by groups and you will be able to create a group forecast. Back to Module 3: Forecasting Groups

31 Watch: Forecasting Groups, Channels, and Segments Forecasting Groups, Channels, and Segments Imagine that a professional organization has contacted your hotel to reserve rooms at a group rate for their annual conference. Imagine further that they've booked rooms with you every year for eight years. If the rate decision were up to you, would you know how to forecast this group? In this video, Professor Kimes introduces you to forecasting for groups, market segments, and channels. Back to Module 3: Forecasting Groups

32 Create the Group Forecast You must achieve a score of 100% on this quiz to complete the course. You may take it as many times as needed to achieve that score. You will be able to see your detailed results after you submit the quiz. Points: 1.0 Back to Module 3: Forecasting Groups

33 Forecasting Challenges Instructions: You are required to participate in both of the discussions in this course. Discussion topic: Forecasting is an essential ingredient of hotel revenue management. Without a forecast, revenue management techniques can't be used successfully. As a result, managers are motivated to create good forecasts whether they know how to or not. What are some of the challenges you face (or expect to face) in creating accurate forecasts? How do you and your colleagues deal with these challenges? To participate in this discussion: Click Reply to post a comment or reply to another comment. Please consider that this is a professional forum; courtesy and professional language and tone are expected. Before posting, please review ecornell's policy regarding plagiarism (the presentation of someone else's work as your own without source credit). Grading Type: Pass/Fail Points: 10.0 Back to Module 3: Forecasting Groups

34 Reflect on Forecasting Groups In the previous modules, you began work on the action plan you will submit for evaluation at the conclusion of the course. You will now perform further work on that plan. Instructions: Open your saved Action Plan and review the work you did in the previous module. In the section labelled Notes, make notes of concepts or recommendations contained in module 3 that you think may be useful to consider when you work further on your action plan. For each item you include, indicate why you think it may be relevant to your situation. Save your work. This action plan is a course requirement, but do not submit it at this time. Before you begin: While you won't submit the document now, please review the rubric (list of evaluative criteria) that will be used when you submit the document with notes and completed plan at the end of the course. Also review ecornell's policy regarding plagiarism (the presentation of someone else's work as your own without source credit). Grading Type: Not Graded Back to Module 3: Forecasting Groups

35 Module Wrap-up: Forecasting Groups Forecasting Groups In this module, you considered how to create forecasts for group business. You saw how hotels can forecast groups and market segments to get the information they need to manage revenue. Now you are able to explain why forecasting is done by group and you can create a group forecast. Back to Module 3: Forecasting Groups

36 Module 4: Measuring Forecast Error 1. Module Introduction: Measuring Forecast Error 2. Watch: Errors in Forecasting 3. Read: Use Absolute Value 4. Watch: Calculating Error Using MAD and MAPE 5. What Is the Error? 6. Reflect on Measuring Forecast Error 7. Module Wrap-up: Measuring Forecast Error

37 Module Introduction: Measuring Forecast Error Measuring Forecast Error Are your forecasts giving you the information you need in order to maximize revenue? Forecast error might be compromising your ability to use revenue management effectively. In this module, you will find out why forecast error is such an important issue. In addition, you will consider several methods to calculate forecast error. When you have completed this module, you will be able to discuss the impact of errors on forecasts and calculate error using two different methods. Back to Module 4: Measuring Forecast Error

38 Watch: Errors in Forecasting Errors in Forecasting A forecast is built far in advance of the arrival day in question and concludes when the bookings for that arrival day are complete. On the day following the arrival day (designated as -1 on the booking curve), you are able to see the number of actual arrivals and compare it to the number you forecasted. Chances are these numbers will not be exactly the same. But how much variation is too much and, furthermore, how important is the error? In the following video, Professor Kimes discusses forecast error. Back to Module 4: Measuring Forecast Error

39 Read: Use Absolute Value Use Absolute Value By taking the absolute value, you convert negative numbers to positive numbers Express deviations as absolute values in these error calculations We've established that an accurate forecast is of critical importance in hotel revenue management. Now the question is, How do you measure forecast error? This module presents two methods. Let's look at a data set for a typical week at a fictional hotel. This data set provides the forecasts and the actual arrivals for each day. As shown in the chart below, sometimes the forecasted arrivals number for this hotel was higher and sometimes it was lower than the actual arrivals (rooms sold). The error in the forecast is related to the deviations between these two values. Can you get the error by simply subtracting the actual arrivals from the forecasted arrivals in each case and averaging those numbers? If you try to calculate the average error of a forecast this way, the positive and negative error values will cancel each other out, giving you an incorrect sense of how much error is present. For example, consider this set of deviations: -10, 2, -6, -2, 5, 6, 10, and -5. Averaging these numbers gives you a result of zero. But is your error really zero? We avoid this problem by expressing deviations as absolute values. Taking the absolute value of a number means considering the magnitude of the number itself and not whether it is negative or positive. In effect, taking the absolute value of a number means expressing

40 that number as a positive value. For example, the absolute value of -7 is 7. The absolute value of 7 is also 7. Absolute values are used in the calculation of mean absolute deviation (MAD) and mean absolute percentage error (MAPE), the two methods of error calculation we look at next. Back to Module 4: Measuring Forecast Error

41 Watch: Calculating Error Using MAD and MAPE Calculating Error Using MAD and MAPE Are your forecasts accurate? Accurate forecasts are critical to making good revenue management decisions. Though it's true that every forecast will contain some error, ideally the error will be small. Be sure you know how to evaluate the accuracy of your forecasts so you can address problems if they arise. Watch this video to learn how to calculate forecast error using two common methods, MAD and MAPE. Back to Module 4: Measuring Forecast Error

42 What Is the Error? You must achieve a score of 100% on this quiz to complete the course. You may take it as many times as needed to achieve that score. You will be able to see your detailed results after you submit the quiz. Points: 2.0 Back to Module 4: Measuring Forecast Error

43 Reflect on Measuring Forecast Error You will now return to your action plan and do some further reflection on how the concepts in this course may apply to your situation. Instructions: Open your saved Action Plan and review your previous work. In the section labelled Notes, make notes of concepts or recommendations contained in module 4 that you think may be useful to consider when you work further on your action plan. For each item you include, indicate why you think it may be relevant to your situation. Save your work. This action plan is a course requirement, but do not submit it at this time. Before you begin: While you won't submit the document now, please review the rubric (list of evaluative criteria) that will be used when you submit the document with notes and completed plan at the end of the course. Also review ecornell's policy regarding plagiarism (the presentation of someone else's work as your own without source credit). Grading Type: Not Graded Back to Module 4: Measuring Forecast Error

44 Module Wrap-up: Measuring Forecast Error Measuring Forecast Error In this module, you saw how forecast error can compromise your ability to use revenue management effectively. You are now able to discuss the impact of errors on forecasts and calculate error using the MAD and MAPE methods. Back to Module 4: Measuring Forecast Error

45 Module 5: Setting Rates Based on Demand 1. Module Introduction: Setting Rates Based on Demand 2. Watch: Setting Trigger Points 3. Watch: Demand-Control Charts 4. Read: A Guide to Rates and Demand 5. Recommend a Rate 6. Reflect on Setting Rates Based on Demand 7. Module Wrap-up: Setting Rates Based on Demand

46 Module Introduction: Setting Rates Based on Demand Setting Rates Based on Demand In this module, you will consider specialized guidelines for controlling various levels of demand to maximize RevPAR. Using information from forecasts and other sources, you will develop room-rate categories and learn to apply rates effectively. When you have completed this module, you will be able to create a demand-control chart and discuss how a demand-control chart is used. In addition, you will be able to develop rate recommendations based on demand patterns. Back to Module 5: Setting Rates Based on Demand

47 Watch: Setting Trigger Points Setting Trigger Points Price is one of the two key levers of revenue management. In managing price, it's important to have multiple rates and also to quote appropriate rates based on demand. Revenue managers can create rate guidelines using what are known as trigger points to define ranges of demand and the rates for each. Learn about setting trigger points in this video. Back to Module 5: Setting Rates Based on Demand

48 Watch: Demand-Control Charts Demand-Control Charts It's possible to use trigger points and forecasts to create a rate guide called a demandcontrol chart. In this video, Professor Kimes explains the demand control chart and how to use it. Back to Module 5: Setting Rates Based on Demand

49 Read: A Guide to Rates and Demand A Guide to Rates and Demand Here's an example of how to build a demand-control chart in three easy steps. 1. Create or obtain forecasts for upcoming dates. Some sample forecast arrivals and forecast occupancies for a 150-room hotel are included below. Forecast DOW Date ROH DBA Pickup Arrivals Occ% Tues 13-Sept % Wed 14-Sept % Thurs 15-Sept % Fri 16-Sept % Sat 17-Sept % Sun 18-Sept % Mon 19-Sept % 2. Next, determine the trigger points associated with hot, warm, and cold zones. Some sample trigger points are: Hot: 100% occupancy Warm: 80% occupancy Cold: 0% occupancy These trigger points indicate that the hot zone is 100% occupancy and above, the warm zone is 80% occupancy to 99% occupancy, and the cold zone is 0% occupancy to 79% occupancy. 3. Now determine the minimum rates to quote for each forecasted day. In this example, the hotel has three rates: $80, $100, and $120. An $80 rate is available if the projected occupancy is in the cold zone. The $100 and $120 rates are also available. A $100 rate is available if the projected occupancy is in the warm zone. The $120 rate is

50 also available. Only the $120 rate is available if the projected occupancy is in the hot zone. The other two rates are closed. Using this information, you can assign minimum rates. Forecast DOW Date ROH DBA Pickup Arrivals Occ% RATE Tues 13-Sept % $80 Wed 14-Sept % $80 Thurs 15-Sept % $100 Fri 16-Sept % $120 Sat 17-Sept % $120 Sun 18-Sept % $100 Mon 19-Sept % $80 The demand-control chart is complete. Back to Module 5: Setting Rates Based on Demand

51 Recommend a Rate You must achieve a score of 100% on this quiz to complete the course. You may take it as many times as needed to achieve that score. You will be able to see your detailed results after you submit the quiz. Points: 3.0 Back to Module 5: Setting Rates Based on Demand

52 Reflect on Setting Rates Based on Demand Once again, you will return to your action plan and do some further reflection on how the concepts in this course may apply to your situation. Instructions: Open your saved Action Plan and review your previous work. In the section labelled Notes, make notes of concepts or recommendations contained in module 5 that you think may be useful to consider when you work further on your action plan. For each item you include, indicate why you think it may be relevant to your situation. Save your work. This action plan is a course requirement, but do not submit it at this time. Before you begin: While you won't submit the document now, please review the rubric (list of evaluative criteria) that will be used when you submit the document with notes and completed plan at the end of the course. Also review ecornell's policy regarding plagiarism (the presentation of someone else's work as your own without source credit). Grading Type: Not Graded Back to Module 5: Setting Rates Based on Demand

53 Module Wrap-up: Setting Rates Based on Demand Setting Rates Based on Demand In this module, you considered how specialized guidelines can be used for controlling levels of demand to maximize RevPAR. Now you are able to create and use a demand-control chart. You can use information from forecasts and demand patterns to develop rate recommendations. Back to Module 5: Setting Rates Based on Demand

54 Module 6: Managing Length of Stay 1. Module Introduction: Managing Length of Stay 2. Watch: Controlling Length of Stay 3. Read: Dos and Don'ts of Length of Stay 4. Fill Your Hotel 5. Pitfalls and Benefits of Controlling Availability 6. Module Wrap-up: Managing Length of Stay 7. Complete and Submit Your Action Plan 8. Watch: Thank You and Farewell

55 Module Introduction: Managing Length of Stay Managing Length of Stay In this module, you will explore the use of length-of-stay controls and consider how to apply these controls to maximize occupancy. You will see how to develop your skills in the area of availability controls. When you have completed this module, you will be able to discuss the impact of length of stay and list several availability controls. Back to Module 6: Managing Length of Stay

56 Watch: Controlling Length of Stay Controlling Length of Stay Sometimes lower-than-expected occupancy occurs at high-demand times. Managers interested in determining why this might be the case should consider length-of-stay controls. These controls can be used to help raise occupancy during high-demand times. Learn how these controls can be implemented, and when, in this video. Back to Module 6: Managing Length of Stay

57 Read: Dos and Don'ts of Length of Stay Back to Module 6: Managing Length of Stay

58 Part of managing revenue effectively is controlling demand, not just responding to it Three approaches you can use to control demand are setting a minimum length of stay, setting a maximum length of stay, and closing a date to arrivals In hotel management, it is sometimes necessary to control demand, not just respond to it. This is part of managing revenue effectively. Here are three controls you can use during high-demand periods to maximize revenue. Minimum Length of Stay When to use Minimum length of stay can be used at any hotel where there will be a period of high demand (a string of busy nights) followed by a period of low demand. Some example scenarios include a resort hotel over a winter holiday or any hotel in the vicinity of a major national festival or conference. What to do Implement a rule to accept longer-duration reservations and reject shorter-duration reservations for arrival during a hot period. In this way, you can fine-tune demand during hot times to increase occupancy during the slow period that follows. Potential problems You may not have anybody who wants to stay longer than the minimum you have in mind. Also, your guests may decide to leave early. improving it. Requirements Be sure you have sufficient demand for longer lengths of stay. Otherwise, use of this control could have a detrimental effect on RevPAR instead of Maximum Length of Stay

59 When to use This control is used when you are expecting to be able to sell out your rooms at higher rates. Using maximum length of stay, you can limit the number of rooms sold at large discounts during the high-rate time period by limiting the (discounted) multi-night stays extending into that time period. What to do Do not accept reservations at specific discounted rates for multiple-night stays extending into the soldout period. To do this, use the start of the sold-out period as a guide to determine the maximum length of stay allowable for discount customers. To accommodate guests who would like to stay at the hotel longer than the maximum length, it is possible to charge two rates: the discount rate for nights up to the maximum and the rack rate for subsequent nights. Potential problems Your guests may decide to stay longer. By law, you can't force them out of their rooms. Requirements Be sure you have high demand. Otherwise, you could decrease RevPAR instead of improving it. Closed to Arrival When to use This control can be used to restrict arrivals during a time when you expect to reach maximum occupancy through guests staying on at the hotel for multiple nights (through "stayovers" as opposed to through new arrivals). Using this control would only make sense if you believed you would achieve higher occupancy by selecting a particular set of guests (i.e. those arriving before the closed-to-arrival date). What to do Do not accept reservations for arrivals on the day in question. Allow guests staying through from previous nights only.

60 Potential problems Be very, very careful with this control, because using it will have an impact on the day you've closed to arrivals, and the day after, and the day after that. You may end up improving revenue on some days but decreasing it on others. Requirements Be sure you have extremely high demand.

61 Fill Your Hotel Grading Type: Points Points: 0.0 Submitting: External Tool Back to Module 6: Managing Length of Stay

62 Pitfalls and Benefits of Controlling Availability Instructions: You are required to participate in both of the discussions in this course. Discussion topic: Your success as a practitioner of hotel revenue management depends on your ability to manage high-demand periods wisely. This means controlling availability. Join this discussion of the promise and peril of length-of-stay controls. Do you use them or other controls like them? Have you encountered, or do you expect to encounter, any problems using these controls? Some hotels have experienced very serious consequences from having used too many length-of-stay controls at once. When you're trying to control availability, how will you determine the right number of controls to use? Share some of the experiences you and your colleagues have had, and learn from the experiences of others. If you do not have personal experience with availability controls, comment on potential issues or challenges you see with such controls and indicate how you would address them. To participate in this discussion: Click Reply to post a comment or reply to another comment. Please consider that this is a professional forum; courtesy and professional language and tone are expected. Before posting, please review ecornell's policy regarding plagiarism (the presentation of someone else's work as your own without source credit). Grading Type: Pass/Fail Points: 10.0 Back to Module 6: Managing Length of Stay

63 Module Wrap-up: Managing Length of Stay Managing Length of Stay This module explored the use of length-of-stay controls and showed you how to apply these controls to maximize occupancy. You are now able to discuss the impact of length of stay and list several availability controls. Back to Module 6: Managing Length of Stay

64 Complete and Submit Your Action Plan Throughout this course, you have been looking at how to use forecasting and availability controls as part of revenue management. In your action plan you have been taking notes relating to how these methods and strategies may be relevant in your own situation or in a hotel or business with which you are familiar. It's now time to complete your action plan. Instructions: Open your saved Action Plan document. Referring to the notes you took, complete all sections of the plan. Save your work. Submit the action plan for instructor review by uploading it below. This assignment is required for course completion. Before you begin: Please review the rubric (list of evaluative criteria) for this assignment and ecornell's policy regarding plagiarism (the presentation of someone else's work as your own without source credit). Grading Type: Pass/Fail Points: 20.0 Submitting: Online Upload Back to Module 6: Managing Length of Stay

65 Watch: Thank You and Farewell Thank You and Farewell Thank you for taking this course. Back to Module 6: Managing Length of Stay

66 Supplemental Reading List The Center for Hospitality Research provides focused whitepapers and reports based on cutting-edge research. "Discounting in the Hotel Industry: A New Approach." Hanks, Richard D., Robert G. Cross, and R. Paul Noland. Cornell Hotel and Restaurant Quarterly 43 (2002): (This article originally appeared in Cornell Hotel and Restaurant Administration Quarterly in February of 1992.) "Best-available-rate Pricing at Hotels: A Study of Customer Perceptions and Reactions." Kimes, Sheryl E. and Kristin V Rohlfs. Center for Hospitality Research Reports 5.7 (2005): "The Basics of Yield Management." Kimes, Sheryl E. Cornell Hotel and Restaurant Administration Quarterly 30 (November 1989): "Perceived Fairness of Yield Management." Kimes, Sheryl E. Cornell Hotel and Restaurant Administration Quarterly 43 (February 2002): "A Retrospective Commentary on Discounting in the Hotel Industry: A New Approach." Kimes, Sheryl E. Cornell Hotel and Restaurant Administration Quarterly 43 (August 2002): "Revenue Management: A Retrospective." Kimes, Sheryl E. Cornell Hotel and Restaurant Administration Quarterly 44 (December 2003): "Wishful Thinking and Rocket Science: The Essential Matter of Calculating Unconstrained Demand for Revenue Management." Orkin, Eric B. Cornell Hotel and Restaurant Administration Quarterly 39 (August 1998): "A '4-C' Strategy for Yield Management." Withiam, Glenn. Center for Hospitality Research Reports 1.1 (2001): Back to Course Resources