Maintain Control

Tools And Working Methods In O&D-Based Revenue Management Control Part 2 Of A Three-Part Series

This is part 2 of a three-part series on the methodology and background surrounding the advantages of O&D-based revenue management control. Featured in this article are tools and working methods associated with O&D control.

As stated in the first part of this three-part series (“Get Control” Ascend 2016, Issue No. 1), revenue management is a critical activity for the profitability of any airline, which leads naturally to the logical conclusion that the methods of control applied by the revenue-management department are also critical.

Many carriers have switched from traditional control methods (such as leg-based control) to O&D-based control, which represents a more holistic approach. Managing a cutover from leg-based control to O&D-based control may quite arguably be subject to certain definable risks, but the benefits justify the change.

The opportunity to absorb considerable knowledge based on the experiences of numerous carriers that have already changed to O&D-based control can quite seriously assist those yet to make the change.

Centering on learning points that some airlines have experienced, part 1 of this series examined human issues in the equation. This article, part 2, focuses on the tools and working methods of O&D-based revenue management control.

Those tools and working methods include creation, storage and reporting of O&D data in addition to leg data; O&D working methodology; demand forecasts; and revenue planning under O&D parameters.

A Successful Cutover
Coordination Is Essential

Four key aspects need to be managed in a coordinated way to ensure a successful cutover to O&D revenue-management control. The tools are fairly obvious in terms of the software, but the working methods, the need to manage O&D and leg reporting and data storage are critical. So, too, are the outputs of demand and revenue forecasts. The human factors such as organization culture, individual skills and new thinking are important. The impact of O&D on other departments should not be underestimated since O&D affects the entire airline, especially the commercial departments. Finally, the opportunity of connecting traffic must be seized and optimized.

The O&D Approach

O&D relates specifically to the journey an airline passenger makes.

Reviewing the bigger picture, if a passenger boards a flight at one airport and gets off at the next stop, that single nonstop flight (in the case of that particular passenger) is an O&D. In airline parlance, that one stop is referred to as a “leg.”

If a passenger boards a flight at one airport, lands at another airport and then flies on to a third airport, that entire passenger journey is also an O&D, and the journey has consisted of two legs.

Innumerable passengers travel on journeys that may cover many legs, and in some cases, those passengers travel on different airlines on some of the various different legs of their journeys. All of these are O&Ds.

Historically, airline revenue-management departments have used computer software to maximize their business results, the success of which has traditionally been measured based on optimization of the individual legs.

The algorithms drive the decisions to open and close availability of seats at different prices, or to maximize the flight “load factor” (a percentage derived by dividing the number of seats flown by the total seat capacity).

Logic underlying the algorithms was intended to maximize the yield (the return on each seat sold as measured in RASK, or revenue per available seat kilometer).

Therefore, the objective of the airline’s revenue-management department was to maximize the aggregate of the individual flights, meaning that airline revenue-management software has normally looked at individual flights, not passenger journeys.

Furthermore, airline revenue-management software has not maximized an airline’s route network (the network should include codeshare flights) as an entity.

In this discussion of the relative merits of O&D control as compared to leg-based control, the primary objective is to zero in on the following factors:

  • Bid price,
  • Planning and interaction,
  • Working methods,
  • Demand forecasts.
Booking Curve
Effectively Manage Availability Of Seats

A booking curve shows how total bookings are made over the life of the selling period of a flight. Effectively, a forecast of expected bookings is made and then compared to the actual bookings at specific days prior to departure, thus enabling the revenue management staff to decide how to manage the availability of seats at different prices. “A” represents the number of bookings that actually flew. “B” shows the actual aircraft capacity, and “C” reflects the overbooking (A+B) that was used to achieve the final amount of bookings.

Key Differentiators

Organization design is notably different under O&D control as opposed to leg-based control.

The primary O&D organizational-design difference revolves around the role of the demand analyst, a role which does not exist under leg-based control.

Each demand analyst is responsible for a number of O&D flows of passengers, and manages demand forecasts for these flows.

Other responsibilities of the demand analyst include:

  • Understanding the importance of forecasting (as opposed to allocation) via a new set of command levers and influences within O&D,
  • Maximizing network revenue in assigned areas to established department key performance indicator (KPI) goals,
  • Validating and maintaining demand forecasts for short- and long-term strategies for all assigned O&Ds,
  • Providing accurate proactive planning for special events and holidays in forecasting and demand adjustments,
  • Consistently reviewing reports to identify shifts in demand,
  • Communicate across the revenue-management team.

Meanwhile, flight analysts have other responsibilities, which are not in conflict with those of demand analysts but actually complement demand analysts’ duties. For example, they provide an important role by validating route-level forecasts (which is the sum of all O&D forecasts) to ensure they are realistic.

Other primary responsibilities of the flight analyst in an O&D-control scenario include:

  • Creating overbooking strategies for all legs in the market scope;
  • Implementing overbooking strategies in the revenue-management system;
  • Monitoring and adjusting no-show, group utilization and cancellation rates;
  • Monitoring spoilage and denied-boarding reports;
  • Assuring more accurate no-show and cancellation forecasts;
  • Working with airport staff to develop a common understanding of overbooking issues;
  • Monitoring forecast and route levels to ensure the sum of all O&D forecasts are reasonable and in-line with expectations.
Modifying One Of Five Options To Influence Demand
Accurately Forecasting Demand

Revenue management staff can use five different ways of modifying the demand for an O&D. Prioritizing specific markets and deciding which are significant enough to forecast is one. Similarly, prioritizing points of sale is critical as price varies by point of sale. Care must be taken to ensure several reservation booking designators are grouped into a bundle accurately. Seasonality affects demand and must be included in demand forecasts, and because outliers can cause errors in the demand forecast, outliers need to be considered and usually removed. Finally, the host market has a major impact on demand and the level of competition in the host market must be clearly understood for an accurate demand forecast.

As a result of the different roles under O&D, there is greater need for close cooperation among flight analysts, demand analysts and pricing executives. Actions and accountabilities are spread fairly evenly across the three roles.

Under leg-based control, the flight analyst was responsible for both the demand forecast (in this case, essentially a booking curve) and the opening and closing of availability at the individual-selling-class level.

Under O&D, the demand analyst is responsible for the demand forecast of O&Ds, which include one leg that is the responsibility of the flight analyst (there are many O&Ds across most legs). Thus, flight analysts will be obligated to work with several demand analysts, and demand analysts will have to work with several flight analysts.

Pricing executives under O&D are usually responsible for several O&Ds (this is not tremendously dissimilar to a pricing executive’s role under leg-based control).

The pricing executive is responsible for setting price points and restrictions for all O&Ds, and also has responsibilities for filing fares, and thus for the fares utilized in the O&D system, which applies an average of relevant fares using a hierarchy which, when demand exceeds supply, creates a threshold price called a bid price.

Bookings with fares at or above the bid price will get access to the seats and be able to complete the booking, whereas bookings with fares below the bid price will be denied access, and will therefore be unable to complete the booking.

A highly important takeaway here is that the quality of the teamwork across the group of staff will inevitably determine how successful the O&D-based control will prove to be.

It’s frequently noted that under leg-based control, a flight analyst could theoretically work in isolation (although that still isn’t good practice, and is definitely not recommended), whereas under O&D control, working in isolation is impossible and would undoubtedly lead to serious loss of revenue.

Another key learning point from O&D cutovers is the fact that the O&D-based control environment has a lot more effect on flights other than the one being planned for.

Again, any one flight will have several O&Ds flowing across it, so actions intended to boost one O&D may impact other O&Ds that have a common leg, the example being a price decrease intended to boost the volume of demand in one country or on one local O&D.

However, if the balance of volume is enough greater, this could also reduce the bid price. It would therefore, in effect, reduce the bid price and open availability for other O&Ds on another flight leg to lower-priced bookings, thus sub-optimizing the second flight, and it may additionally change the demand forecast for other O&Ds that share the common leg. It does not change the demand forecast for other O&Ds; however, it will change the bid price, which could change the availability in other O&Ds since the threshold is lower, and demand forecasts in other O&Ds remains unchanged.

This is very undesirable. It is also illustrative of a risk of O&D-based control unless strong teamwork between the demand analyst and the flight analyst enables corrective action to be taken to offset the original action on the impacted O&Ds and flights.

The working methods under O&D-based control require greater planning across all commercial departments (sales, revenue management and network management) than under leg-based control. This, in turn, requires closer monitoring of actual performance versus the plan, as well as closer monitoring of actual performance versus budget.

Working methods under O&D are quite different from those under leg-based control.

So when managing a cutover from leg-based control to O&D-based control, the revenue-management department should require its staff to work through numerous examples of real business situations, such that all individuals in the department understand not only the new working methods, but also how to use the new tools in the O&D software (finding a level of intervention that works without impacting other areas).

Also, enabling an accurate demand forecast is a fundamental issue and requires practice, as well as training to master, somewhat like learning to ride a bicycle or safely drive a car.

Nuances Of Manual Interventions

Manual interventions, if they are made too often, can reduce the accuracy of the demand forecast.

The O&D system is designed to have a certain tolerance with regard to slight changes in price or demand, and can therefore manage most cases.

To get the best from the software demand forecast, manual intervention should be kept to a minimum as a general rule of thumb.

Differentiating System Signals

There is a key difference between how the leg-based control system works (using a booking curve, which is a demand forecast at flight level, and the opening and closing of individual selling classes as the two main “levers” for the flight analyst to control results) and O&D software, which uses an O&D demand forecast and bid price.

Under leg-based control, a booking curve is the demand forecast. The booking curve tracks how the bookings (reservations) come in over the “booking horizon” (the time during which the flight is available in the reservations system).

A booking curve is a graphical representation of passenger-booking behavior and is produced for each selling class (and for each flight) every day.

The key difference under O&D-based control is that the demand forecasts are made for each O&D, not each flight. Rather than opening or closing selling classes to control availability of seats at a given price, availability under O&D is driven by the combination of the demand forecast and the resulting bid price.

Onward With O&D Control

When managing a cutover from leg-based revenue-management control to O&D-based control, there are several key lessons to be learned in relation to several business aspects, including:

  • Bid price,
  • Planning and interaction,
  • Working methods,
  • Demand forecasts.

As stated, the process of changing to O&D-based control carries risks, but experts agree that the benefits justify the change.

Tools and working methods are critical parts of that change, and managing these factors well is imperative to success.

This is part 2 of a three-part series. Watch for part 3 in the next issue of Ascend, which will discuss connections, codeshares and responsibilities of other departments in O&D revenue management control. View part 1, titled “Get Control,” from the previous issue of Ascend.