A) what are the main points in this article

B) what are the methods of measuring restaurant revenue management success?

C) What strategies does Kimes (1999) discuss ?

The goal of restaurant revenue management (RRM) is to maxi mize revenue per available seat-hour by manipulating

price and meal duration. An earlier Cornell Quarterly article discussed the theory behind RRM ,1 but the application of RRM has not been explored. In a series of articles on the application of RRM , of which this is the first, I will discuss how to develop a RRM system, demonstrate how RRM wasapplied to a pilot restaurant, and develop guidelines that restaurant operators can use to increase their revenue per available seat hour. In this paper I discuss the funda mentals of restaurant revenue man agement, examine methods of mea suring revenue-management success, explain how those measures are different from the traditional gauges of restaurants' success, and outline a five-step approach to establishing a revenue-management system. Overview of Restaurant Revenue Management Restaurant revenue management can be defined as selling the right seat to the right customer at the right price and for the right dura tion. The determination of "right" entails achieving both the most rev enue possible for the restaurant and also delivering the greatest value or utility to the customer. Without that balance, RM-type practices will in the long term alienate those cus tomers who will feel that the restau rant has taken advantage of them. Revenue management, or yield management, is commonly practiced in the hotel and airline industries. Companies implementing revenue management report increases in revenue of 2 to 5 percent over the results of prior procedures. Revenue management requires a focus on the revenue per available inventory unit. For example, hotels measure revenue per available room-night (com monly referred to as RevPAR), airlines measure revenue per avail able seat-mile (RPSM), and cruise lines measure revenue per available cabin. When restaurant operators apply revenue management to their restaurants, I recommend that they measure their results in terms of revenue per available seat hour (RevPASH). Concentrating on RevPASH has major implications for the way in which a restaurant is operated and evaluated. Many managers currently measure their restaurant's success by tallying the average check or by maintaining certain labor- and food-cost percentages. While such measures are valuable for many pur poses, they do not explicitly reflect a restaurant's revenue- (or profit-) producing performance. RevPASH, on the other hand, combines infor mation from the average check and seat use (or occupancy) to provide a measure of the flow of revenue through the system and to indicate how effectively a restaurant is using its productive capacity. Restaurant operators have two main strategic levers that they can use to manage revenue: namely, price and meal duration.2 Price is a fairly obvious target for manipula tion, and many operators already offer price-related promotions to augment or shift peak-period de mand (e.g., early bird specials, spe cial menu promotions). More so phisticated manipulations of price include day-part pricing, day-ofweek pricing, and price premiums or discounts for different types of party size, tables, and customers. Managing meal duration is a bit more complicated. On the produc tion side, managers must streamline and control their service-delivery process, as well as understand customer-arrival patterns and de termine ways of influencing meal duration.3 One of the things that makes implementing revenue man agement so difficult in restaurants is the fact that their explicit unit of sale is a meal (or event) rather than 2Kimes et al, June 1998; see also: Sheryl E. Kimes and Richard B. Chase,"The Strategic Levers of Yield Management,'"Journal of Service Research, Vol. l,N o . 2 (Nov. 1998),pp. 156166. 3 For a discussion of the effects of altering the service-cycle time, see: Christopher C. Muller, "A Simple Measure of Restaurant Efficiency," on pages 31-37 of this Cornell Quarterly; for an analysis of arrival times, see: Brian Sill and Rob ert Decker, "Applying Capacity-management Science: The Case o f Browns Restaurants," on pages 22-30 of this Cornell Quarterly. an amount of timealthough one could argue that the true measure of the restaurant's product is time.4 While one can estimate a likely mean length for that meal, the ac tual duration is not set. By compari son, implementing revenue manage ment is much easier for the hotel, airline, cruise-line, and car-rental businesses, because they sell their service for an explicitly contracted amount of time. Restaurants rarely sell tables for a fixed amount of time, and in most western cultures are reluctant to broach this topic with customers. Moreover, North American restaurateurs cannot even rely on the practice common in some countries of charging for the cover. One of the stumbling blocks to successful implementation of restau rant revenue management is the struggle that restaurant operators have in developing internal methods of managing meal duration. In the context of managing meal duration, one should not think only of reduc ing diners' average meal length. Quite often the factor interfering with revenue management is the variability in meal lengths, and not just their duration. Some of the ways in which managers can influ ence meal duration include chang ing reservation policies, redesigning menus, and pacing service proce dures and making them more effi cient. Managers can also train em ployees to respond to customers' apparent wishes regarding the length of the meal. While some customers may wish to linger over coffee, for instance, managers might be sur prised at how often the holdup in turning a table is the restaurant's own lackadaisical approach to ser vice and servers' inattention to cus tomers' needsRestaurant managers are typically evaluated by the check averages and by the food- and labor-cost per centages that they have been able to maintain. As I indicated before, none of those measures captures sufficient information about the revenue- (or profit-) generating performance of the restaurant. Some measure of the revenue-generating potential and performance of the restaurant must be developed. For a restaurant manager to con centrate only on a high average check, for instance, is equivalent to a hotel's focusing solely on a high average room rate.5 Without infor mation on the percentage of capac ity use or occupancy of the restau rant, revenue performance cannot be evaluated. A high average check may even be detrimental in times of strong demand if, for example, cus tomers linger over their meal while other parties wait for a table. Similarly, a manager's achieving specified food-cost and labor-cost percentages is laudable, but that does not tell the entire story. In particular, the margin is not a measure of prof itable use of capacity. A restaurant manager does a good job of maintaining margins and still be unprofitable. An overemphasis on margins can lead to a propensity to focus unduly on minimizing costs. Again, reducing cost is fine, but not when that causes reduced revenue due to disgruntled customers. The extent to which available seats are occupied is another com monly applied measure of success, since a busy restaurant is generally a revenue-producing restaurant. Reli ance on seat occupancy as a measure of success suffers from the same problem as reliance on hotel-room occupancy (in the absence of con sideration of ADR), because high use does not necessarily mean highrevenue. A restaurant can run at 90-percent of capacity and still not make money if menu items are sold at too low a price, for example. Because it embraces capacity use, check averages, and cost margins, revenue per available seat-hour (RevPASH) is a much better indica tor of the revenue generating per formance of a restaurant than the commonly used measures that I just discussed. RevPASH indicates the rate at which revenue is generated and captures the trade-off between average check and facility use. If occupancy percentages increase even as the average check decreases, for instance, a restaurant can still achieve the same RevPASH. Con versely, if a restaurant can increase the average check, it can maintain a similar RevPASH with a slightly lower facility use. Exhibit 1 gives a hypothetical illustration of this principle. The four restaurants in the exhibit all have the same RevPASH ($7.20), but each achieves it in a different manner. Restaurant A has a facility use of 40 percent and an average check of $18.00, while Restaurant D has a use ratio of 90 percent but an average check of $8.00. Restau rants B and C also achieved a RevPASH of $7.20, but with vary ing facility-use and average-check statistics. The easiest way to calculate RevPASH is to divide revenue (or profit) for the desired time period (e.g., day part, day, month) by the number of seat-hours available dur ing that interval. For example, as sume a 100-seat restaurant makes $1,500 on Fridays between 6:00 and 7:00 p m . Its RevPASH would be $15 ($1,500 -5- 100 seats X 1 hour). Similarly, if that same 100-seat res taurant made $5,000 over a fourhour meal period, its RevPASH would be $12.50 ($5,000 - 100 seats X 4 hours, or $5,000 -s- 400 available seat-hours).he number of turns and the length of the meal, or service cycle. As the number of turns increases and meal length decreases, the RevPASH increases. Just a one-minute reduc tion in meal time during a highdemand period can lead to an in crease in RevPASH of 1.5 to 2.0 percent. Returning to our hypo thetical 100-seat restaurant with its four-hour dinner, say that its aver age service cycle is 60 minutes. In that case, the restaurant can poten tially handle 400 customers per night. If the average check is $15, its maximum nightly revenue is $6,000, and its potential RevPASH is $15. If the meal time can be re duced to 59 minutes, the restaurant can handle an additional 6.8 cus tomers. If the average check re mains at $15, its potential nightly revenue increases to $6,102 and its potential RevPASH increases to $15.26 (a 1.7-percent increase). With the increased volume, the check average could even drop by $0.20, and revenue would still increase. Reduced meal times can be achieved by changing the service process, altering staffing levels, or altering the menu. The first few minutes of reduction are not that difficult or expensive to achieve, for example, by picking up the pace of greeting, seating, and check settle ment. Deep reductions, however,may require substantial investment for example, by adding kitchen equipment or more employees. A return-on-investment analysis that considers the effects of service-cycle changes on RevPASH can help op erators decide whether a prospective investment is worthwhile. In making their plans managers should remem ber that customer preferences and expectations limit the minimum feasible meal duration and will set a theoretical minimum acceptable RevPASH. RevPASH-based Strategies Once operators understand their RevPASH patterns, they can de velop strategies for dealing with high and low RevPASH periods. A full discussion of the RRM strate gies available will be presented in a subsequent paper of this series. Dur ing low RevPASH periods, manag ers can either try to attract more customers and increase use or rely on suggestive selling to increase the average check. During those periods with high RevPASH, operators should consider raising menu prices or try to reduce meal duration so that the restaurant can increase its turn rate. RevPASH can be used at differ ent levels of analysis and for different purposes. At the individual restau rant level, managers may choose to develop hourly or quarter-hourly RevPASH figures to help develop arevenue-management strategy best suited to their restaurant. RevPASH can also be used to evaluate the effectiveness of an operation's serv ers and managers. Say that the manager of our hy pothetical 100-seat restaurant wanted to understand her hourly RevPASH patterns for September. Obtaining data from the restaurant's POS system, the manager found that her highest RevPASH periods were on Fridays and Saturdays from 6:00 to 10:00 p m and on Sundays from 6:00 to 9:00 p m . The manager can use this information to help develop revenue-management tac tics specific to high and low RevPASH periods. For example, during high-RevPASH periods, she may focus on reducing the meal time by having table servers skip suggestions of desserts or after dinner drinks. On the other hand, at low-RevPASH times she may de cide to increase the use of suggestive selling or even reduce menu prices to boost traffic. Comparative RevPASH Regional or national managers could use RevPASH to compare performance of different restaurant units. One would want to adjust the unit RevPASH according to an area's cost of living, but a unit-by unit comparison of RevPASH would give a good indication of the relative performance of different restaurants in an area, region, or nation. Consider a city with six restau rants franchised by a particular chain, as shown in Exhibit 4. Res taurant 4 enjoyed the highest aver age check ($12.10) of the six units, but it also has the lowest RevPASH ($4.25). On the other hand, restau rant 5 has the lowest average check ($9.45), but the second-highest RevPASH ($6.25). Relative perfor mance measurements can be calcu lated by dividing one restaurant's performance by the average perfor mance of all the units. Thus, the check performance of restaurant 4 is above average (1.14), but its RevPASH performance is only 0.83 ($4.25 + $5.20). By using relativeperformance measurements like these, regional managers can better evaluate the revenue generation of the restaurants they oversee. One could also calculate and compare competitors' RevPASH to get a sense of how well a restaurant, a region, or a chain is performing. For example, the RevPASH of a particular restaurant could be com pared with the average RevPASH of the competitive set to evaluate per formance (see Exhibit 5). This type of calculation is done for hotels, for example, by Smith Travel Research reports.6 While competitive infor mation is not readily available, one may be able to develop information to perform such an analysis. The Five Steps to RRM When developing an RRM system a restaurant operator must first under stand current conditions and perfor mance. Following this, the operator must evaluate the possible drivers of that performance. This understand ing will help managers determine how to improve RevPASH statistics. Finally, the manager must monitor the impact of implemented changes on revenue performance. I describe each of these steps below. (1) Establish the baseline. Most man agers know their average check and their labor- and food-cost percentages, but few can accu rately estimate the capacity use or RevPASH of their restau rants. To develop a RRM pro gram, operators must collect detailed information on arrival patterns, meal times, RevPASH patterns, and customer prefer ences. This information can be collected from a variety of sources, including the POS system, guest checks, and me thodical observation. Once collected, the data must be analyzed to determine the mean and deviation of dining time and daily and hourly RevPASH patterns. (2) Understand the drivers. Once the baseline data have been collected, managers should analyze the factors that affect meal duration and RevPASH performance. Simple tools such as process analysis,7 service blueprints,8 and fishbone used to better understand the possible reasons for why meals last as long as they help identify the most important problems in controlling meal duration. (3) Make recommendations. After identifying the causes of the most important problems af fecting the service cycle, man agers should develop detailed recommendations on how to correct those problems. Some solutions may deal with reduc ing the overall meal duration, while others may deal with reducing variability in particu lar service steps (e.g., order taking, bussing), and still others may involve table management or customer-arrival manage ment. The manager should analyze potential return on investment for each recommen dation to ensure prudent deci sion making. A later article will present a simple model for evaluating the ROI of RevPASH improvements. (4) Implement the changes. For RRM to be successful, restaurant op erators must ensure that manag ers, servers, bussers, and other employees clearly comprehend the purpose and practice of RRM. This requires a positionspecific training program that helps employees understand their role in RRM and how RRM can benefit both the restaurant and employees. Addi tionally, operators should align any employee-incentive pro grams to coincide with the objectives of RRM. (5) Monitor outcomes. As with any business practice, the success of 9D. Daryl WyckofF,"New Tools for Achieving Service Quality," Cornell Hotel and Restaurant Administration Quarterly, Vol. 25, No. 3 (November 1984), pp. 78-91; and U. Apte and C. Reynolds, "Quality Management at Kentucky Fried Chicken," Interfaces, Vol. 25, No. 3 (1995), pp. 6-21. RRM cannot be ensured without measurement of im improvement. After establishing the baseline and implementing RRM , operators must develop a system to measure RRM performance. One should, for instance, monitor RevPASH and the average and standard deviation of dining time and compare those figures to baseline performance. Future Papers This article is the first of a series on restaurant revenue management. In future articles I will discuss spe cifically how to develop an RRM system and establish a baseline including the types of data to gather, the possible sources of data, and the analysis and interpretation of the information collected. These and the other restaurant revenue management points will be illus trated with the actual experience of a 100-seat restaurant in Ithaca, New York. The series will review methods for analyzing the possible causes of problems, including service blue prints, fishbone diagrams, and pro cess analysis, and it will examine restaurant revenue-management strategy and tactics. I will also dis cuss how to combine the informa tion obtained from the baseline and causal analysis to develop specific strategies for increasing RevPASH. Finally, the papers will examine issues surrounding RRM imple mentation and discuss methods of establishing appropriate incentive and training programs for servers, managers, and bussers, along with suggestions on how to monitor the success of the revenue-management system. By the end of the series, I hope to have drawn an outline of how to implement revenue management in your restaurant and thereby improve its revenue generating performance.