# Article on Yield Management_Function-space Revenue Management- A Case Study From Singapore

Cornell University School of Hotel AdministrationThe Scholarly Commons

Articles and Chapters School of Hotel Administration Collection

12-2001

Function-space Revenue Management: A CaseStudy from SingaporeSheryl E. KimesCornell University, [email protected]

Kelly A. McGuireRadiant Systems

Follow this and additional works at: http://scholarship.sha.cornell.edu/articlesPart of the Hospitality Administration and Management Commons

This Article or Chapter is brought to you for free and open access by the School of Hotel Administration Collection at The Scholarly Commons. It hasbeen accepted for inclusion in Articles and Chapters by an authorized administrator of The Scholarly Commons. For more information, please [email protected].

Recommended CitationKimes, S. E., & McGuire K. A. (2001). Function-space revenue management: A case study from Singapore [Electronic version].Cornell Hotel and Restaurant Administration Quarterly, 42(6), 33-46.

Function-space Revenue Management: A Case Study from Singapore

AbstractHotels can apply revenue-management systems to their function spacesand boost the revenue contributionfrom those spaces.

Keywordsrevenue management, yield management, strategies, hotel industry, Singapore

DisciplinesHospitality Administration and Management

CommentsRequired Publisher Statement Cornell University. Reprinted with permission. All rights reserved.

This article or chapter is available at The Scholarly Commons: http://scholarship.sha.cornell.edu/articles/467

REVENUE MANAGEMENT HOTEL OPERATIONS

DECEMBER 2001 Cornell Hotel and Restaurant Administration Quarterly 33

BY SHERYL E. KIMES AND KELLY A. MCGUIRE

Revenue management, also known as yield manage-ment, is commonly applied to hotels guest-room in-ventories, but the practice has not yet been systemati-cally applied to hotels function spaces.1 Hotels functionspaces provide substantial income that could be enhanced byapplying revenue-management, as we explain in this article.

As is the case in all revenue-management applications, thegoal of function-space revenue management is to maximizethe revenue contribution of each function space for each timeperiod that the space is available. The measure we apply forthis purpose is contribution per available space for a giventime (ConPAST). Hotel managers can manipulate price and

Function-spaceRevenue Management

A Case Study from Singapore

2001, CORNELL UNIVERSITY

Hotels can apply revenue-management systems to their function spacesand boost therevenue contribution from those spaces.

event duration to affect this measure. ConPAST has the fol-lowing three components: contribution, space, and time. Wesuggest that measuring contribution is preferable to measur-ing revenue because revenue alone doesnt account for thevarying profit margins arising from renting function spaces.Space refers to the number of square feet or meters in thefunction room, and time refers to the time segment for whichthe revenue contribution is measured. In this paper, we ex-plain how we developed a revenue-management strategy forthe Raffles City Convention Center at the Swisstel Stam-ford and Raffles Plaza in Singapore. (While those propertiesare now managed by Swisstel, at the time of our study thehotels were managed by Westin.)

1 Jerry Hartley and Peter Rand, Conference-sector-capacity Management,in Yield Management for the Service Industries, second edition, ed. Ingold,McMahon-Beattie, and Yeoman (London: Continuum, 2000), pp. 315338.

ventories

HOTEL OPERATIONS REVENUE MANAGEMENT

34 Cornell Hotel and Restaurant Administration Quarterly DECEMBER 2001

Revenue-management ReviewRevenue management is the application of in-formation systems and pricing strategies to matchcustomers with services at an appropriate timeand price. A common formulation of revenuemanagement is selling the right capacity (in thiscase, function space) to the right customer at theright place at the right time.2 The determinationof right in this instance entails achieving boththe most contribution possible for the hotel inrenting a function space while also delivering thegreatest value or utility to the customer for thereservation and use of that space. In practice, rev-enue management has meant setting prices ac-

cording to predicted demand levels so that price-sensitive customers who are willing to purchaseat off-peak times can do so at favorable prices,while price-insensitive customers who want topurchase at peak times will be able to do so. Theapplication of revenue management has been mosteffective when it is applied to operations that haverelatively fixed capacity, demand that is variableand uncertain, perishable inventory, appropriatecost and pricing structures, and varying customerprice sensitivity.3 Those attributes are found inthe business of renting hotels function space.

Different industries are subject to differentcombinations of duration control and variablepricing (see Exhibit 1).4 Industries traditionallyassociated with revenue management (e.g., ho-tels, airlines, car-rental firms, and cruise lines) areable to apply variable pricing for a product thathas a specified or predictable duration (Quadrant2). On the other hand, sellers of function space,movie theaters, performing-arts centers, andarenas charge a fixed price for a product of pre-dictable duration (Quadrant 1), while restaurantsand golf courses charge a fixed price but face arelatively unpredictable duration of customer use(Quadrant 3). Many health-care businesses chargevariable prices (e.g., Medicare versus private pay),but cannot predict the duration of patient use,even though some may try to control that dura-tion (Quadrant 4). The lines dividing the quad-rants are broken because in reality no fixed de-marcation point exists between any of thequadrants. Thus, a given enterprise (such as rent-ing function space) may have attributes from morethan one quadrant.

Successful revenue-management applicationsare generally found in Quadrant-2 industries,because they can manage both capacity and price.To obtain the benefits associated with revenue

Quadrant 4

Continuing careHospitals

Price

Quadrant 1

MoviesStadiums and arenasConvention centers

Hotels function space

Quadrant 2

Hotel roomsAirline seatsRental carsCruise lines

Quadrant 3

RestaurantsGolf courses

Internet-service providers

EXHIBIT 1

A Typology of Revenue Management

A version of this diagram was previously used in: Sheryl E. Kimes, Revenue Man-agement on the Links: Applying Yield Management to the Golf-course Industry,Cornell Hotel and Restaurant Administration Quarterly, Vol. 41, No. 1 (February2000), p. 127; and most recently in: Lawrence R. Weatherford, Sheryl E. Kimes,and Darren A. Scott, Forecasting for Hotel Revenue Management: Testing Aggre-gation Against Disaggregation, Cornell Hotel and Restaurant AdministrationQuarterly, Vol. 42, No. 4 (August 2001), p. 54.

2 Barry C. Smith, John F. Leimkuhler, and Ross M. Darrow,Yield Management at American Airlines, Interfaces,Vol. 22, No. 1 (JanuaryFebruary 1992), pp. 831.

3 See: Sheryl E. Kimes, Yield Management: A Tool forCapacity-constrained Service Firms, Journal of OperationsManagement, Vol. 8, No. 4 (November 1989), pp. 348-363; and Robert G. Cross, Revenue Management (New York:Broadway Books, 1997). See also: Robert G. Cross,Launching the Revenue Rocket: How Revenue Manage-ment Can Work for Your Business, Cornell Hotel and Res-taurant Administration Quarterly, Vol. 38, No. 2 (April1997), pp. 3243.

4 Sheryl E. Kimes and Richard B. Chase, The StrategicLevers of Yield Management, Journal of Service Research,Vol. 1, No. 2 (November 1998), pp. 156166.

Dur

atio

nU

npre

dict

able

Fixed Variable

Pred

icta

ble



management, industries outside of Quadrant 2should attempt to emulate Quadrant-2 businessesby deploying the appropriate strategic levers (forexample, by controlling duration). This impliesthat Quadrant-1 businesses, such as renting func-tion space, should concentrate on developingdemand-based pricing approaches. This is not tosay that duration management cannot be im-proved, but only that most of the benefits asso-ciated with revenue management will be obtainedfrom differential pricing.

In the next sections we review some of theoptions available for managing meeting-space du-ration, revenue, and demand.

Function-space RevenueManagementManaging function-space revenue is more diffi-cult than dealing with transient guest roomsbecause of the interaction of function space withroom sales and food and beverage demand. Whilefunction space will reap most of its revenue-management benefits from pricing, this is not todownplay the role of duration control.

Duration control. Duration can be managedin four ways: refining the definition of duration,reducing arrival uncertainty, reducing durationuncertainty, and reducing the amount of timebetween functions. 5 Each is briefly described andthe issues associated with each discussed.

Definition of duration. Most hotels definefunction-space duration by day part, but the defi-nition and number of day parts varies by hotel.Since events can span multiple day parts or useonly a fraction of a day part, we suggest that ho-tel function-space managers define duration asan hour. Most computerized systems track thisinformation, but it may be difficult to retrieve.If its impossible to track the function-roomschedule by the hour, tracking by day parts willwork, as we demonstrate below.

Arrival uncertainty. Arrival uncertainty in-volves two factors: (1) the timing and quantityof requests, and (2) the no-show and cancella-tion rates of booked events. Determining the tim-ing and number of requests from different mar-ket segments is a challenging task. Hotels still

have a great deal of difficulty with forecastingtheir demand for group rooms, and most havenot even attempted to forecast their function-space demand.6 If a hotel can obtain informa-tion on the lead time and the amount of busi-ness of major market segments, it can make betterdecisions regarding the level of demand from keycustomers, when that demand will occur, andwhen to release space to other market segments.

While hotels have a difficult time in forecast-ing function-space demand, they do a good jobof reducing cancellation and no-show ratesby requiring nonrefundable deposits and pre-payments for function space. In some high-demand cities, hotels even require prepaymentfor the requested number of guest rooms con-nected to the function-space reservation. Thispractice protects those hotels from holding roomblocks that no one picks up at the agreed-on rate.

Duration uncertainty. Since events can spanmultiple days or day parts, hotels must ensurethat the guest rooms associated with the eventdo not unnecessarily displace higher-paying tran-sient business and must also ensure that the eventlasts for the agreed-on length of time. The pre-payment that we just mentioned also helps en-sure that the contracted function space is usedfor the required length of time. However, hotelshave a much more difficult time ensuring thatthe guest rooms will be occupied for the speci-fied length of time. Many groups, particularlyconventions, suffer from high room attrition onthe last day or two of the event. If group mem-bers check out early, the hotel is usually stuckwith an empty room, unless the hotel also im-poses early departure fees or requires prepaymentfor all rooms. Then again, early departure fees androom prepayments may incur customers wrath.

Turnaround time. In high-demand periods,the time needed to turn the room around canaffect the hotels ability to meet demand. Hotelsthat try to minimize the labor costs associatedwith set-up and tear-down often end up turningdown business that could have been accommo-dated if the hotel had employed a sufficient num-ber of employees to turn over the room. For ex-

5 Ibid.

6 Sheryl E. Kimes, Group Forecasting Accuracy for Ho-tels, Journal of the Operational Research Society, Vol. 50,No. 11 (November 1999), pp. 11041110.



ample, a hotel that requires a two-hour set upand two-hour break-down for all events wouldnot be able to book a meeting that ends at 5:00in the same room with a dinner that begins at6:30. If more employees were assigned to speedthe transition between events, however, the laborcost would be more than covered by the increasedrevenue associated with the incremental event.

Pricing. Function-space pricing is complicatedby the fact that function space cannot be pricedwithout considering its effect on room sales andfood and beverage. A price that may seem toolow for the function space when considered alonemay be more than compensated for when roomand food and beverage revenue are considered.Similarly, a high function-space price may ulti-mately be unprofitable for the hotel if it occa-sions a low guest-room rate and displaces poten-tially higher-paying business from the transientrooms.

Function-space managers face two pricingdecisions: what price to charge and how to de-termine who pays which price. When analyzingwhich business to accept, the manager must as-sess the days and day parts requested, the num-ber of function rooms required, the expectednumber of guest rooms sold, and the expectedsales of food and beverage, as well as other items.Since the profit margins for different revenuestreams vary, the manager must determine theexpected contribution associated with each eventand compare that with any potential displace-ment of other business. If the expected contri-bution is higher than the potential displacement,the group should be accepted. Otherwise, itsbusiness should be declined or referred to anotherday or day part.

Along with determining the appropriate priceto charge, the hotel must also justify the pricescharged to different clients. Rate fences can beused to determine who pays which price. So-called physical rate fences might include thepresence of certain amenities (e.g., high-techrooms command a premium) or location (e.g.,rooms on the first floor command a premium).Non-physical rate fences might include customercharacteristics (e.g., repeat customers receive adiscount), transaction characteristics (e.g., cus-tomers who book events through the internetreceive a discount), or day of week or time of

day (e.g., customers who book events for week-ends or mornings receive a discount).

RM Program for Function SpacesTo develop a revenue-management program forfunction spaces, managers should:

(1) establish the baseline of performance,(2) understand the drivers of that performance,(3) develop a revenue-management strategy,(4) implement that strategy, and(5) monitor the strategys outcome.7

In the following sections, we will discuss how wedeveloped a revenue-management program forthe Raffles City Convention Center at theSwisstel Stamford and Raffles Plaza hotels inSingapore. We will conclude the case study witha discussion of the steps a hotel should take toimplement function-space revenue management.

The Case of the Raffles CityConvention CenterThe Raffles City Convention Center (RCCC) atthe Swisstel Stamford and Raffles Plaza inSingapore will be used to demonstrate how rev-enue management can be applied to functionspace.8 The RCCC is reasonably typical of a con-vention hotels function space. The conventioncenter occupies a 70,000-square-foot space onthe hotels fourth floor, comprising 18 functionrooms ranging in size from 20-person meetingrooms to the Raffles Ballroom, which holds over1,000 people for a sit-down dinner. The threeballrooms (Raffles Ballroom, Atrium Ballroom,and Stamford Ballroom) drive convention, wed-ding, and social revenue, while three sets of meet-ing rooms (the Executive Conference Center,Plaza Meeting Rooms, and Stamford MeetingRooms) handle demand for small meetings andconference breakout sessions. Each of the ball-rooms can be divided into three sections, andthe Plaza and Stamford Meeting Rooms also can

7 Sheryl E. Kimes, Implementation of Restaurant RevenueManagement: A Five-Step Approach, Cornell Hotel andRestaurant Administration Quarterly, Vol. 40, No. 3 (June1999), pp. 1522.

8 Much of this section comes from: Kelly A. McGuire,Function-space Revenue Management: A Case Study ofthe Raffles Convention Center at the WestinSingapore,Cornell University School of Hotel Administration, unpub-lished monograph, 2001.



be divided. Two high-tech rooms provide pro-jection and sound equipment for executive meet-ings. Foyer space is used for registration and cof-fee breaks.

The director of marketing for the hotel over-sees the four sales groups, which are dividedamong weddings, conventions and exhibitions,meetings, and dinners and dances. Each salesgroup has its own director and team of sales man-agers. Each subgroup director is responsible forsetting the sales targets for her team, and all fourdirectors work together to set prices for the nextyear. A catering-sales analyst, who reports directlyto the director of revenue management, workswith all four sales groups. Her primary responsi-bilities are to oversee the programming and ad-ministration of the Delphi catering- and sales-system software and to generate reports andanalyses.

The Raffles City Convention Center drivesguest-room sales at the Swisstel Stamford andRaffles Plaza, and the function space is frequentlysold at a reduced rate to generate room-nights.This arrangement is not atypical for a hotel ofthis size. The hotel has 2,000 rooms, divided intothree main sections: the Stamford, and the Northand South Towers of the Plaza. Compared to thePlazas towers, the Stamford is a lower-rated prop-erty in terms of dcor and amenities offered, butat 70 stories it is the tallest hotel in the world.The Plaza is more upscale than the Stamford. TheSouth Tower has recently been renovated andcommands a considerable rate premium. Planshave been made to renovate the North Tower, andeventually the Stamford itself will be upgraded.

The function-space market in Singapore is di-vided into two segments: convention centers andhotels. Singapore has two convention centers:Suntec City, located three blocks from theRCCC, and the Expo Center, which is near theairport, or about 25 kilometers from downtown.The major-hotel market includes Marriott,Hyatt, Hilton, Pan Pacific, and the Oriental, alllocated within a few kilometers of the SwisstelStamford and Raffles Plaza.

The RCCC considers its primary competitiveset to be the two convention centerschiefly,Suntec City. Until Suntec City was built, theRCCC had the largest function space in down-town Singapore. Neither Suntec City nor the

Expo Center has guest rooms of its own, how-ever, so RCCC competes by focusing on its abil-ity to provide the services of a full-service hotel,in addition to its function space. Because theyhave so many guest rooms, the Swisstel Stam-ford and Raffles Plazas secondary competitive setincludes other hotels in Singapore. However, theRCCC is nearly double the size of any of thosehotels function space, which forms a point ofdifferentiation compared to the other hotels.

Step 1: Establish a BaselineThe first step in implementing a revenue-management system for a propertys functionspace is to establish the baseline performance forthat space. Baseline-performance measuresshould include both performance characteristicsof the meeting space and demand behavior ofthe various market segments. The performancemeasurements should include both occupancyand our proposed revenue-contribution factor(i.e., contribution per available space-time).These should be applied to the entire space, eachroom type, each day of week, and each day part.Demand behavior should include information onthe demand patterns and booking patterns foreach market segment by month, by day of week,and by day part.

When analyzing space performance, hotelsshould look to the traditional measures used fortracking guest-room sales: occupancy andRevPAR. With few exceptions, hotels track oc-cupancy on a daily basis, and most hotels alsotrack their RevPAR. Very few hotels know theoccupancy of their function space and even fewerknow their ConPAST.

Occupancy. Function-space occupancy istricky to calculate because a single room can bedivided (or not) and the appropriate time unitmust be determined. As we indicated above, webelieve that the best time unit to use in calculat-ing function-space occupancy is one hour, butmost hotels do not have sufficiently accurate datafor a calculation at this level. Because of this, themost practical time unit is the day part. The num-ber of day parts per day varies by hotel, but it isusually related to the number of times each daythat a function space can be sold. Most hotelsuse two or three day parts per day for each func-tion space.



Function-space occupancy can be calculatedby dividing the number of day parts used by thenumber of day parts available. For example, if ahotel has ten function rooms and three day partsper day, its monthly capacity available (assum-ing a 30-day month) would be 900 day parts (10 3 30). If the hotel sold a total of 450 dayparts of function space in a month, its occupancywould be 50 percent.

With that calculation in mind, we obtaineddetailed information on functions booked atthese Singapore hotels from their Delphi systemfor November 2000 through May 2001. We ex-cluded coffee breaks in our calculation of occu-pancy and contribution figures, to reduce thepossibility of double counting room use.

We used just two day parts, AM (before 5:00PM) and PM (after 5:00 PM). On that basis, wetracked daily occupancy for each major meetingarea (Atrium Ballroom, ECC rooms, Plaza meet-ing rooms, Raffles Ballroom, Stamford Ballroom,and Stamford meeting rooms).

As mentioned previously, all of the meetingareas in our convention center can be dividedinto smaller spaces (except for those in the ECC).To calculate the rooms occupancy, we expressedeach meeting space as a percentage of the totalspace for that particular venue. For example, thehotel has four Plaza meeting rooms, so each ofthose rooms was considered as 25 percent of thetotal. The number of times a specific room was

booked was multiplied by that percentage andadded to the values for the other individual roomsto determine the total number of events bookedin that type of space. Thus, if two of the indi-vidual Plaza meeting rooms were occupied, theoccupancy for the overall Plaza meeting roomswas 50 percent, and if three of the rooms wereoccupied, the occupancy was 75 percent. Usingthis rubric, if the dividers were not in use for aparticular space, its occupancy was by definition100 percent for that particular day part. The onlyexception to this rule is how we treated the roomsin the Executive Conference Center. Since theECC rooms cannot be combined into one space,we calculated the occupancy for each of thoserooms separately.

As indicated in Exhibits 2 and 3, PM occu-pancy was much higher than AM occupancy inNovember and December. Exhibit 2 shows thatoverall (total) PM occupancy ranged from 36 per-cent in the Stamford meeting rooms to 91 per-cent in the Stamford Ballroom. Occupancy alsovaried by day of week. For example, the occu-pancy of the popular Stamford Ballroom rangedfrom 56 percent on Tuesdays to 133 percent onSaturdays (indicating that it was turned morethan once during the PM day part). As the tablein Exhibit 3 shows, overall AM occupancy rangedfrom 3 percent in the Raffles Ballroom to 48percent in the Bailey Meeting Room (one of theECC rooms). Occupancy tended to be a bithigher during the week than on weekends.

Ballroom occupancy. Since the ballroomscould be divided, we tracked the occupancy ofboth the ballroom parts and each entire ballroom.As hoteliers know only too well, when one divi-sion of a ballroom is booked, the entire spacecannot then be rentedjust the other section(s).This factor created intriguing occupancy patterns.The Stamford Ballroom, for instance, had ahigher occupancy when the entire space wasrented as a unit. On the other hand, the AtriumBallroom and Raffles Ballroom had higher week-day occupancies when they were sold as divisiblespaces on weekdays, but as the entire space onweekends.

ConPAST. As we suggest above, hotels shouldalso calculate each spaces contribution per avail-able space-time to determine that spaces contri-bution to operating profit. Again, we argue that

EXHIBIT 2

PM Occupancy for November and December 2000

Facility Sun Mon Tue Wed Thu Fri Sat Total

Atrium Ballroom 52% 38% 71% 48% 81% 89% 100% 69%

ECC: Bailey 33% 38% 50% 56% 67% 33% 33% 44%

ECC: Bras Basah 44% 63% 44% 39% 56% 56% 44% 49%

ECC: Orchard 56% 50% 54% 48% 70% 59% 33% 53%

ECC: Wilberforce 0% 63% 75% 56% 67% 56% 33% 49%

Plaza rooms 28% 53% 72% 69% 83% 81% 61% 64%

Raffles Ballroom 67% 42% 42% 44% 52% 89% 85% 61%

Stamford Ballroom 111% 63% 56% 72% 106% 89% 133% 91%

Stamford meeting room 22% 23% 42% 41% 44% 43% 39% 36%



contribution is preferable to revenue because ofthe varying profit margins of the different streamsof function-space revenue. For example, food andbeverage expenditures usually have a contribu-tion margin of 30 to 35 percent, while roomrental may have a contribution margin of 85 to95 percent, and AV rentals may have a contribu-tion margin ranging from 50 to 95 percent. Re-lying only on revenue generation (as opposed tocontribution) can lead hotel managers to developunprofitable selling patterns.

Contribution should be calculated by space(either square feet or square meters) because thismeasurement provides a standard for the com-parison of different types of rooms and roomconfigurations. For example, if a hotel has a di-visible ballroom and finds that the ConPAST ishigher when the room is divided than when it issold as one space, it should direct its marketingefforts at smaller groups or ensure that largergroups are charged a premium.

Most hotels have information on their guest-rooms competitive performance, but very fewhave information on the competitive perfor-mance of their function space. In some NorthAmerican markets, HotelRevMax provides thisinformation. For example, if a hotel has amonthly ConPAST of $5 and its competitive sethas a monthly ConPAST of $4, its yield indexwould be 1.20 ($5/$4).

The contribution per available space-time wascalculated by dividing the contribution per daypart (AM and PM) by the amount of space (num-ber of square feet) for each meeting room type(see Exhibits 4 and 5 on the next page).

In the PM day part, the Raffles Ballroom hadby far the highest overall ConPAST ($4.21), fol-lowed by the Stamford Ballroom at $0.73. TheConPAST of the Raffles Ballroom was higheston Fridays ($5.59), Saturdays ($7.71), andSundays ($4.81). Mondays were low, at $1.90.

The AM day part had an extremely lowConPAST across the board (Exhibit 5). The Ex-ecutive Conference Center had the highest over-all ConPAST ($0.24), followed by the Plazameeting rooms ($0.17).

Demand behavior. Information on the de-mand and booking behavior of each market seg-ment should be tracked and analyzed. Demandbehavior includes both functions that were

booked (constrained demand) and those thatwere turned down (unconstrained demand).Booking behavior shows how far in advance par-ticular market segments make their reservations.Information on booking behavior can helphotels better forecast sales and improve decisionson when to release space to different marketsegments.

Constrained and unconstrained demand. Aswith guest-room reservations, hotels need infor-mation on both the constrained and uncon-strained demand for function space.9 To reviewthis topic, the level of constrained demand canbe determined from actual sales, while uncon-strained demand can be estimated by declinedreservations, or the requests that could have beenaccommodated if the hotel had unlimited space.We observe that most hotels do a poor job oftracking unconstrained demand, but such infor-mation is essential for understanding the truenature of demand. Take the case of a conventionthat occupies a hotels entire function space for aweek. Had other groups wished to book spaceduring that time, they would have been turneddown due to space constraints. If function-space

EXHIBIT 3

AM Occupancy for November and December 2000


Atrium Ballroom 0% 29% 29% 30% 33% 11% 26% 22%

ECC: Bailey 33% 38% 75% 56% 67% 33% 33% 48%

ECC: Bras Basah 33% 50% 25% 44% 22% 22% 11% 30%

ECC: Orchard 56% 25% 4% 48% 41% 44% 30% 36%

ECC: Wilberforce 0% 63% 63% 56% 56% 56% 33% 46%

Plaza rooms 17% 31% 47% 58% 47% 47% 22% 39%




9 See: Eric Orkin, Wishful Thinking and Rocket Science:The Essential Matter of Calculating Unconstrained Demandfor Revenue Management, Cornell Hotel and RestaurantAdministration Quarterly, Vol. 39, No. 4 (August 1998),pp. 1519.



and the number of regrets. Unless an automatedtracking system can be developed, it is unlikelythat accurate information will be captured.

Booking behavior. Information on how farin advance various market segments make theirbookings can help hotel managers improve theirforecasting and can also help them decide whento release space to different market segments. Forexample, most hotels have a conflict betweencatering sales and food and beverage, as the twodepartments attempt to book the same space. Ifsufficient booking information were available,managers would be able to make better decisionson which type of business to accept.

The booking curve shows the booking paceof each segment. Booking-curve data can be usedto set holds and releases for function spaces andto give an indication of the comparative revenueperformance of a particular month (i.e., are webehind or ahead of typical bookings?). The book-ing curves we developed were based on averagerevenue-on-hand (RevOH) for each month be-fore arrival (MBA).

A steep booking curve indicates a fast book-ing pace while a gradual slope illustrates a slowand steady pace. Knowing the booking pace canhelp sales managers decide on release dates forspaces. Thus, if a sales manager knows that nearlyall conventions are booked by 12 months beforearrival, she can release all of the larger ballroomsthat are not booked for conventions a year aheadof time, because it is unlikely that there will beadditional requests for a convention.

A Look at DemandThe Swisstel Stamford and Raffles Plaza servefive major market segments: convention and ex-hibition, meetings, weddings, social, and tours.10

Except for tours, those market segments were rela-tively balanced in 2000. Social events had thelargest percentage of total revenues that year (28percent), while conventions were close behind at24 percent. Weddings and meetings each pro-vided 23 percent of the hotels revenues, whiletours contributed only 3 percent. Well use theconvention-and-exhibition and meetings seg-ments to illustrate the use of demand and book-ing data.

EXHIBIT 4

PM ConPASTFacility Sun Mon Tue Wed Thu Fri Sat Total

Atrium Ballroom $0.36 $0.06 $0.24 $0.15 $0.34 $0.46 $0.66 $0.33

ECC $0.36 $0.21 $0.25 $0.29 $0.30 $0.65 $0.42 $0.36

Plaza rooms $0.21 $0.32 $0.27 $0.22 $0.34 $0.38 $0.39 $0.30

Raffles Ballroom $4.81 $1.90 $2.47 $4.17 $2.35 $5.59 $7.71 $4.21

Stamford Ballroom $1.11 $0.49 $0.28 $0.48 $0.70 $0.79 $1.18 $0.73

Stamford meeting room $0.21 $0.09 $0.16 $0.15 $0.17 $0.14 $0.18 $0.16

10 Much of this section comes from: McGuire, op. cit.

EXHIBIT 5

AM ConPASTFacility Sun Mon Tue Wed Thu Fri Sat Total

Atrium Ballroom $0.00 $0.04 $0.04 $0.05 $0.06 $0.02 $0.06 $0.04

ECC $0.29 $0.17 $0.11 $0.26 $0.15 $0.47 $0.19 $0.24

Plaza rooms $0.18 $0.17 $0.19 $0.19 $0.13 $0.11 $0.19 $0.17

Raffles Ballroom $0.02 $0.21 $0.16 $0.18 $0.00 $0.01 $0.09 $0.09

Stamford Ballroom $0.17 $0.14 $0.08 $0.06 $0.04 $0.03 $0.17 $0.10

Stamford meeting room $0.20 $0.06 $0.16 $0.13 $0.17 $0.10 $0.12 $0.13

sales alone were considered in this instance, ho-tel managers would be overlooking the presenceof additional demand. They might well makebiased future allocation decisions because theylack information about the business that wasturned down.

Unconstrained demand can be divided intodenials and regrets. Denials are customers whoare turned away because of capacity constraintsor restrictions. For example, if a function spaceis not available, or if sufficient guest rooms arenot available, the hotel may be forced to declinethe request for event space. Regrets are custom-ers who are turned down because they are notinclined to pay the price of the space or a room(i.e., shoppers). Those inquiries may not be a truerepresentation of customers who intended tobook. For this reason, it is more important totrack denials than to be concerned about regrets.

Tracking unconstrained demand implies thatsales managers must tally the number of denials



Conventions. Conventions are major revenuegenerators for the hotel, since they typically bringin a significant number of room-nights along withthe function-space revenue and additional F&Bsales. Since the events are so large and are bookedfar in advance, they form the cornerstone of thefunction-space use.

The convention-segment managers report thattheir peak month is October. The average num-ber of covers per function for conventions in 2000was 290. This average includes all breakout ses-sions and VIP luncheons, as well as the large con-vention banquets and exhibit-hall food service.Convention managers report an average lengthof stay of four days per convention. Most con-vention and exhibition events begin in the morn-ingmost frequently at 9:00 AM.

The 2000 bookings for conventions and ex-hibitions started earlier than for other marketsegments (Exhibit 6). For instance, the hotel al-ready had convention RevOH at 36 months be-fore arrival. Half of the RevOH had been bookedby 19 months before arrival and 90 percent hadbeen booked by seven months before arrival.

Since 75 percent of all functions were bookedby one year before the event, the property couldbegin to release unbooked space to other seg-ments at the one-year point. At seven monthsbefore arrival it could release all remaining space,as few convention requests are likely to come inafter that point.

Meetings. Although the percentage of totalrevenue of the meetings segment is close to thatof conventions, meetings are usually smaller insize, with an average of only 69 covers per event.Many more meetings must be booked to gener-ate revenues equivalent to those of conventionsand exhibitions. For example, in 2000, 3,190meeting events were booked, compared to 710convention and exhibition events. Meetings typi-cally last one day, occur during the week, andmainly comprise morning and afternoon sessions,each with a coffee break. Sales managers reportpeak seasons in October, November, and May.December, January, and July are off peak.

The meetings booking curve stands in con-trast to the booking curve for conventions (Ex-hibit 7). Relatively few meeting bookings oc-curred until six to eight months before arrival.So, just as convention bookings began to slow

EXHIBIT 6

Convention and exhibition booking curve

$600,000

$500,000

$400,000

$300,000

$200,000

$100,000

0 0 2 4 6 8 10 12 14 16 18 20 24 28 33 35

I I I I I I I I I I I I I I

Months before arrival

EXHIBIT 7

Meetings booking curve

$500,000

$450,000

$400,000

$350,000

$300,000

$250,000

$200,000

$150,000

$100,000

$50,000

0 I I I I I I I I I I I I I I0 2 4 6 8 10 12 14 16 22 26 30 32 34 360

Months before arrival



down, meetings bookings started to pick up.Based on that analysis, hotel managers can bereasonably sure that they will get meeting requestsuntil close to the actual event. The meetings busi-ness is a good complement to conventions, be-cause the two segments have opposite bookingpatterns. Therefore, the hotel can turn away mostmeeting requests until seven months before ar-rivalwhich is about the time that the remain-ing convention space is released.

Understanding demand and booking patternsfor each segment is important when developingan appropriate RM strategy. For example, it helpsthe hotelier determine which spaces to hold ifshe knows that weddings typically book week-end evenings, conventions book weekdays, andthat both groups are similar in size. If a conven-tion wanted to set up on a Sunday, which is apeak day for weddings, for example, the hotelcould charge a premium fee for that privilege.

Step 2: Understanding Causes ofBooking BehaviorOnce the baseline has been established, manag-ers must understand the causes behind past perfor-mance and must develop a revenue-managementstrategy to help improve performance. A thoroughunderstanding of the internal and external envi-ronment of the hotel and its sales policies is es-sential for identifying performance drivers andfor developing revenue-management strategies.Potential causes can be divided into internal andexternal factors.

Internal factors include:(1) Physical constraintsRooms may not be

large enough for certain meetings or mayhave other limiting characteristics;

(2) Labor availabilityThe hotel may nothave enough workers available to turnrooms around between events;

(3) Sales incentivesThe sales-incentive sys-tem may preclude making good revenue-management decisions;

(4) Incomplete dataThe data available maybe inaccurate or difficult to obtain;

(5) Guest-room availabilityFunction-spacerequests often hinge on whether sufficientguest rooms are available; and

(6) Prior bookingsOther business may havealready booked the space.

External factors include:(1) City-wide eventsSome city-wide events

may be so large that it precludes havingspace available for other groups; and

(2) Market conditionsIf the economic con-dition of the overall market has deterio-rated, business from certain market seg-ments may decrease.

An example of internal factor number six(above) occurred at the RCCC convention cen-ter in July 2000, when an extremely large con-vention used all of the hotels function space. Thatbooking resulted in the denial of requests fromother customers. July is normally a slow monthfor conventions, but the large convention madeJuly look like a good month.

The Atrium Ballroom may be an example ofinternal factor number one (above). That spacerecorded a much lower occupancy and ConPASTthan did either the Stamford or Raffles Ballrooms.This weak performance may have been causedby the physical characteristics of the Atrium Ball-room. It has a low ceiling, which makes it lessdesirable for weddings and social events, and thespace dividers are not soundproof, so that mul-tiple small events cannot be booked into thespacethat is, the Atrium can be rented eitheras a large space or a single small space, but not asmultiple small spaces simultaneously.

Step 3: Developing a Revenue-management StrategyEffective revenue management is predicated onthe use of demand-based pricing in which fullprice is charged during high-demand periods,while discounted rates are available during low-demand periods. To apply demand-based pricing,it is essential to identify high- and low-demandperiods for function space.

A simple approach to identifying and com-municating high- and low-demand periods isto color code them on a table. This approach hasbeen applied effectively in hotel rooms andrestaurants.11 For example, high-demand periodsare designated as hot, or red; periods of moder-

11 Sheryl E. Kimes, Deborah I. Barrash, and John E.Alexander, Developing a Restaurant Revenue-managementStrategy, Cornell Hotel and Restaurant Administration Quar-terly, No. 40, Vol. 5 (October 1999), pp. 1830.



ately high demand are designated as warm, oryellow; and low-demand periods are designatedas cold, or blue. All other demand levels are clas-sified as neutral and not colored at all.

Each demand level (or color) should have adifferent set of prices associated with it, based onqualified discounts with rate fences. The highestprices should always be available regardless of thedemand level, but a variety of qualified discountsshould be offered during other periods. A quali-fied discount implies that customers must meetsome restriction (or rate fence) to receive thelower price. Companies use rate fences in an at-tempt to prevent price-insensitive customers fromobtaining discounts that are not necessary to en-courage their business. The rate fences are gen-erally designed to screen out customers who arenot eligible for discounts. Rate-fence rules includephysical attributes of the room (e.g., high tech,good view), transaction characteristics (e.g., cus-tomers who prepay, those who book over a yearahead of time), customer characteristics (e.g., fre-quent purchasers, type of organization), or con-trolled availability (e.g., coupons or direct-mailoffers).

The definition of demand levels is subjectiveand can be based either on occupancy orConPAST. For example, at our Singapore prop-erties, we designated occupancy rates over 80 per-cent as hot; those between 65 and 80 percent aswarm; those less than 20 percent as cold; and therange of 20 to 64 percent as neutral. Alterna-tively, demand-level classifications can be basedon ConPAST, but since the contribution fromeach function space varies, occupancy may be abetter and more comparable measure.

AM Pricing StrategyA key part of the RCCC managements revenue-management strategy was to grapple with theproblem presented by the weak AM day part.Riddled with cold periods, this slow day part hadno hot periods at all and just two warm periods(see Exhibit 8). The brightest spot was the BaileyRoom, which displayed fairly high demand onTuesdays and Thursdays. Given that this is theonly strong point for this day part, the hotelshould quote rack rate for this room. However,if the rooms high occupancy comes from frequentcustomers or standing meetings, those custom-

EXHIBIT 8

AM demand levels


Atrium Ballroom 0 29% 29% 30% 33% 11% 26% 22%

ECC: Bailey 33% 38% 75% 56% 67% 33% 33% 48%

ECC: Bras Basah 33% 50% 25% 44% 22% 22% 11% 30%

ECC: Orchard 56% 25% 4% 48% 41% 44% 30% 36%

ECC: Wilberforce 0 63% 63% 56% 56% 56% 33% 46%

Plaza meeting rooms 17% 31% 47% 58% 47% 47% 22% 39%

Raffles Ballroom 0 4% 4% 0 0 11% 4% 3%



EXHIBIT 9

PM demand levelsFacility Sun Mon Tue Wed Thu Fri Sat Total

Atrium Ballroom 52% 38% 71% 48% 81% 89% 100% 69%

ECC: Bailey 33% 38% 50% 56% 67% 33% 33% 44%

ECC: Bras Basah 44% 63% 44% 39% 56% 56% 44% 49%

ECC: Orchard 56% 50% 54% 48% 70% 59% 33% 53%

ECC: Wilberforce 0% 63% 75% 56% 67% 56% 33% 49%

Plaza meeting rooms 28% 53% 72% 69% 83% 81% 61% 64%




Key for the two tables above: hot warm low



ers would likely qualify for a discount. Customersinterested in booking the Bailey Room who areextremely price-sensitive should either be steeredtowards another room or another day of the week.

During cold periods, managers should focuson filling the rooms by applying any appropriatediscounts for groups requesting space. In addi-tion, sales managers should use the informationon low-demand periods to develop a marketingplan for generating business for those periods.

PM Pricing StrategyEvenings were much busier at our Singapore con-ference center. This day part enjoyed several hotperiods, most notably in the ballrooms (see Ex-hibit 9, on the previous page). The Atrium andStamford Ballrooms, for example, were hot on

Thursdays, Fridays, and Saturdays, and the Stam-ford Ballroom was hot on Sundays as well. ThePlaza meeting rooms and the Raffles Ballroomwere hot on Thursdays and Fridays. The hotelshould be quoting only rack rate for those roomsduring the hot periods. If customers are pricesensitive and unwilling to pay full price, theyshould either be directed to another night of theweek, to the AM day part, or to a cooler func-tion room.

This analysis can highlight not only when tooffer discounts, but when to boost rack rates. Forexample, we suggest that the extremely high oc-cupancy of the Stamford Ballroom (sometimeseven over 100 percent, indicating two turns dur-ing the day part) may indicate that the rates be-ing charged for the hot periods are too low. Simi-larly, the Atrium Ballroom recorded a relativelylow ConPAST, despite its high occupancy rateson Thursdays through Saturdays. Again, this out-come may indicate that the hotel is charging ratesthat are too low. Even though the Atrium Ball-room is due for renovation, it remains extremelybusy.

Step 4: ImplementationFor revenue management to be successful, hotelmanagers must ensure that the director of sales,sales managers, the director of catering, and thedirector of food and beverage clearly understandthe purpose and practice of revenue management.This requires a position-specific training programthat can help employees understand their role inrevenue management and how revenue manage-ment can benefit both the hotel and employees.Additionally, managers should align employee-incentive and -evaluation programs to coincidewith the objectives of revenue management.

Sales-incentive systems that reward achievingvolume or revenue goals can lead to suboptimalprofitability. When volume quotas are used, salesmanagers may book business regardless of its con-tribution. On the other hand, if revenue quotasare the target, sales managers may pad expectedrevenue with unprofitable add-ons. In addition,volume- and revenue-quota systems often leadto unproductive competitions between the salesmanagers of different market segments. The useof contribution quotas can help alleviate theabove conflicts.

Some hotels have experimented with the con-cept of rewarding sales managers for bookingbusiness during low-demand periods. For ex-ample, business brought in during cold periodsmay count for three times as much as the busi-ness brought in during hot times, while businessbrought in during warm periods may count fortwice as much as hot-period business.

Step 5: System EvaluationAs with any business practice, the hotel must ap-ply a reliable measure for the performance of therevenue-management strategy. The success of arevenue-management system can be measured inthree ways: ConPAST, the yield index, and therevenue-opportunity model. ConPAST can bemonitored for growth, but may not be a reliablemeasure because of the complications arisingfrom external market conditions. The yield in-dex and the revenue-opportunity model can re-duce the effects of market conditions on themodel and give a clearer indication of perfor-mance.

North American hotels have long used theyield index to measure the success of their rev-

We believe that revenue managementcan help improve the profitability ofhotel function space.



enue management of guest rooms, and couldapply the same concept to function space. Asdescribed previously, the yield index is simply theConPAST for an individual hotel divided by theConPAST of its competitive set. As long as ahotels ConPAST remains above one, it is out-performing its competition, regardless of marketconditions. Competitive information is becom-ing available in many North American citiesthrough the HotelFlash banquet reports availablefrom HotelRevMax (www.hotelrevmax.com).The reports provide information on banquet-related food, beverage, and room rental per avail-able room, per occupied room, per group room,and per square foot for the subject hotel and forits competitive set. However, hotels located incities not covered by reports of that kind mayfind accurate competitive data difficult to obtain.

For a hotel to use the revenue-opportunitymodel, it must track the timing of both con-strained and unconstrained demand. Therevenue-opportunity model measures the follow-ing three variables: (1) the amount of contribu-tion that would have been achieved if requestsfor space were accepted on a first-come-first-served basis, (2) the amount of contribution thatcould have been achieved if perfect informationwere available, and (3) the amount of contribu-tion that was actually obtained.12 The differencebetween the first-come-first-served revenue andthe perfect revenue is referred to as the revenueopportunity. The performance of a revenue-management system is measured as a percentageof that ideal revenue opportunity that the hotelactually obtained. For example, assume a hotelwould have made $20,000 if it had accepted func-tion-space requests on a first-come-first-servedbasis, but would have made $25,000 if it hadperfect information. Its revenue opportunity inthis instance would be $5,000. If the revenue-management system led to a contribution of$22,000, the hotel would have earned 40 per-cent of its revenue opportunity ($2,000/$5,000).

A Five-step ApproachFunction-space revenue-management is still inits infancy, but we believe that revenue manage-

ment can help improve the profitability of hotelfunction space. In this paper, we attempted todemonstrate how that might occur by proposinga five-step approach for developing a function-space revenue-management strategy and apply-ing that approach to the Raffles City Conven-tion Center in Singapore. As function-spacerevenue management matures, more sophisti-cated forecasting and optimization approachessimilar to those currently used for guest roomswill be developed, and hotel operators will be-come accustomed to measuring the occupancyand ConPAST of function space, just as they nowrefer to the occupancy and RevPAR of guestrooms.

Hotels that wish to develop a function-spacerevenue-management program should followthese five steps:

(1) Establish a baseline. Hotels must first de-termine the baseline performance of theirfunction space. Baseline performance in-cludes information on demand patterns,room-occupancy percentages, averagerent per available square foot per day, andcontribution per available square foot perday part.

(2) Understand the causes. After establishingthe baseline performance, hotels must de-termine the possible causes of the perfor-mance. Understanding the causes of per-formance is important because it can helpmanagers determine why certain thingshave occurred.

(3) Develop a strategy. We demonstrated asimple way to identify peak and non-peakdays and day parts. Once that is in place,a hotels managers can develop appropri-ate pricing strategies for each day part. Inaddition, the hotel can consider impos-ing length-of-stay controls for certaintimes, and determining the pricing andreasoning for the pricing by time of year,day of week, and time of day.

(4) Implement the strategy. Implementation isthe most difficult part of revenue man-agement. With function-space revenuemanagement, managers must ensure thatsufficient training of appropriate person-nel is completed, that incentive and re-ward systems are aligned to match the12 Smith et al., pp. 831.



Sheryl E. Kimes, Ph.D. (above, left), is a professor at the CornellUniversity School of Hotel Administration ([email protected]).Kelly A. McGuire (above, right), who holds a Master of Manage-ment in Hospitality (M.M.H.) degree from Cornell University, is aconsultant in the Hospitality Division at Radiant Systems ([email protected]). The authors would like to thank Jeannette Ho,revenue manager of Singapores Raffles Plaza and Stamford hotels,for her support and assistance with this project.

2001, Cornell University;refereed paper: submission received on October 11, 2001;revisions requested on November 20, 2001; accepted onDecember 2, 2001.

goals of revenue management, and thatinformation systems are positioned toprovide the data necessary for decisionmaking and performance measurement.

(5) Monitor success. Once revenue manage-ment has been implemented, hotels mustmonitor performance and compare it tothe baseline performance. Several perfor-mance measurements exist, includingthe increase in revenue per availablesquare foot, increase in yield index, andthe revenue-opportunity gauge.

The intent of this paper was to discussfunction-space revenue management by using theexperience of a major operation in Singapore asan example to illustrate how a hotel can imple-ment revenue management. We believe this ar-ticle is just the beginning of what promises to bea successful and profitable approach to manag-ing function space. Future developments willmost likely include the sophisticated forecasting,optimization, and duration-control methodscommonly found in revenue management forguest rooms. While more sophisticated methodshold great promise, a hotel will not be able totake full advantage of such methods unless itviews its function space as a resource that can beprofitably managed.

Cornell University School of Hotel AdministrationThe Scholarly Commons12-2001

Function-space Revenue Management: A Case Study from SingaporeSheryl E. KimesKelly A. McGuireRecommended Citation

Function-space Revenue Management: A Case Study from SingaporeAbstractKeywordsDisciplinesComments

# Article on Yield Management_Function-space Revenue Management- A Case Study From Singapore

Documents

Transcript of # Article on Yield Management_Function-space Revenue Management- A Case Study From Singapore