• Sergio Mendoza

Segmentation Obsolescence and the Need for re-Optimization of Airline Fare Structures

Updated: Feb 25

Author: Sergio Mendoza, Ph.D., Cofounder & CEO Airnguru, Sr Advisor Pricinguru


#priceoptimization #revenuemanagement #segmentation #rebound #farestructures #faremanagement #consumerbehavior #pricing


Abstract: The COVID pandemic disrupted the airline industry, substantially changing consumers' purchase and travel patterns. Here we briefly discuss the impacts of such changes on Revenue Management processes, the urgent need for refreshing our understanding of consumer behavior and suggest strategies to adapt to a quickly evolving new scenario.


1. What to expect for the rebound


Government-enforced border restrictions, fear of infection when traveling, quarantine requirements, and other mobility restrictions enforced by law or self-imposed across society as a whole, are each enough to explain why air travel was almost extinguished during the COVID pandemic.


The general expectation though is that freeing travelers and airlines from the severe restrictions that have been imposed around the world would bring back travel volumes to pre-pandemic ones. In fact, a brief survey over your list of family and friends will most likely reveal that a relevant proportion of leisure travelers is eager to be able to travel again, go to beaches, visit relatives or friends abroad, leave home for a while, and take a real vacation. People are tired; many are burned out. Some have benefited from home-office by improving their quality of life -for example, moving to a more convenient or more comfortable location to live and work remotely-, however, home-office with long days in front of computer screens has taken a toll on many.


We should also consider that lifting pandemic restrictions will incentivize touristic destinations and surrounding industries to invest and bet on the recovery of the demand for leisure travel. For many countries, the travel & tourism industry represented a substantial employment generator and a significant fraction of their GDP (typically between 5% and 20% for the largest economies), so we expect those countries to proactively favor its recovery.


On the other hand, COVID restrictions forced corporations to move from traveling for in-person meetings and industry conferences to performing all meetings online. Most corporations and small companies learned and adapted to remote working. Virtual meeting technology provided tremendous support for this profound, forced transformation of labor and human interactions during the pandemic. Some of this behavior will most likely survive after COVID (for the sake of corporate efficiency and productivity!). Still, meeting face-to-face to start and develop productive relationships with customers and partners will hardly be equaled by online meetings, so those businesses that resume key face-to-face meetings will have a commercial advantage over those that don’t.


In summary, it seems reasonable to expect a fast post-pandemic recovery of leisure travel demand even beyond pre-pandemic levels and a slower, partial recovery of business travel demand. By "post-pandemic," we mean the first three years after governments lift most restrictions that limit travel. After the post-pandemic period, assuming no other extraordinary circumstances, the travel industry's growth should primarily depend on overall economic growth again.


2. Evolving demand behavior


In this highly restricted, risky, uncertain context, travelers feel, reason, and prioritize very differently than they did before the pandemic. Consequently, the pandemic has substantially changed consumers' purchasing and traveling behaviors.


Airlines, hotels, and travel agencies have observed substantial fluctuations in booking anticipations, lengths of stay at destination, and customers' needs for flexible change conditions for booking dates, destination, and ticket or booking cancellations.


But the pandemic also affected consumers' shopping behaviors, the consumers' willingness-to-pay (and therefore the demand price-elasticity), the ratios of group bookings over individual bookings and of business over leisure bookings, the preferences for ancillaries, and more.


Consumers' feelings and behaviors will likely keep evolving quickly through the rebound and post-pandemic periods. And, it wouldn't be surprising to observe some residual behavioral changes after the post-pandemic period.


3. Segmentation obsolescence


Airlines' Revenue Management strategies rely on good demand segmentation. Airlines perform demand segmentation on consumers' purchase and travel behaviors and preferences. Demand segmentation is considered part of the Pricing function. Demand segmentation is a key input to product and service segmentation, which defines the set of offers that airlines may present to their customers. Basically, product segmentation mirrors demand segmentation. Fare structures and their components, the fare fences, and fare rules, and the way they are organized in fare families or brands with associated value attributes and price levels, are at the foundation of airlines' product segmentation and aim at maximizing revenue share by capturing consumers’ willingness-to-pay.



If the segmentation were “perfect,” the airline would get its “fair revenue share,” that is, its “fair market share” out of every demand segment from the highest willingness to pay down the demand curve till it fills its capacity. If the airline brand were also “perfect,” i.e., the airline’s brand was the preferred one by every consumer, the airline would fill its capacity capturing every consumer with the highest willingness to pay down the demand curve till completing its seat capacity.


A core assumption made by airlines’ Revenue Management teams is that aggregate travelers’ behaviors are relatively stable, so demand segmentation could be considered relatively static over time. This assumption is broken now due to the profound disruption caused by the pandemic. And this is not a minor problem: the contribution of product segmentation represents a substantial fraction of the total contribution of Revenue Management, especially when load factors are low and consequently, the relative benefit of Yield Management diminishes.


So, though second to the dramatic effect of overall demand contraction, the quick evolution of consumer preferences and behaviors mentioned above, if not incorporated timely in the demand segmentation and, therefore, if not timely recognized in the airline’s product segmentation and price optimization, may also be having a significant detriment on revenue.


The damage caused by the effect of segmentation obsolescence is reflected as a series of gaps in the process: lack of product granularity, dilution caused by overly relaxed fare fences and/or too low prices, spill caused by excessively restrictive fare fences and/or too high prices, etc.


Another significant headache for Revenue Management practitioners is the obsolescence of demand forecasts, on which Yield Management heavily relies for the purpose of inventory optimization. With unprecedented demand contraction and volatility, on top of evolving consumer behaviors, error rates of demand forecasts have substantially increased, leaving them useless. Thus, the pandemic disrupted the legacy inventory optimization process. A weakened Yield Management lever reinforces the need for a more robust Pricing lever. (See our previous discussion on forecasting in highly uncertain contexts: Simu-casting: around and beyond forecasting.)


4. How to tackle these challenges


Under evolving consumer behavior and preferences, airlines should revisit the legacy segmentation process. A first reasonable approach is to increase the frequency of the segmentation process enough to re-capture the behavioral evolution on a regular basis. At the same time, airlines should search for and evaluate new potential segmentation variables that were disregarded or unavailable in the past. Such variables may now be relevant in understanding consumers’ preferences and willingness to pay. Data gathering and processing and unsupervised Machine Learning techniques for clustering purposes are now much easier to implement and frequently execute than they used to be not long ago, thanks to the fast evolution of cloud services.


Willingness-to-pay should also be re-estimated on a more frequent basis, as airlines need to understand price-elasticities associated with the new demand segments, and these elasticities will keep evolving through and beyond the pandemic as well.


Product segmentation shall be revisited consequently with the same frequency as demand segmentation, aiming at improving fare structures, including fare rules and fences, fare levels, and brand attributes, aligning them to the updated demand segmentation and re-estimated willingness-to-pay.


The additional efforts to set up this new process would be increasingly profitable for airlines as traffic comes back to pre-pandemic volumes. This machinery would not only become a powerful tool for RASK growth over the years to come, but it may also become an essential building block for future Offer Management systems

(see Dynamic pricing of airline offers.)


5. How to move forward


If you want to explore and discuss ideas on updating the segmentation process and optimizing fare structures, and the tools required to support such tasks, we would be very happy to share with you our vision and experience in the airline industry and beyond.

Leave us your contact details in order to coordinate a conversation (no strings attached!):

https://www.airnguru.com/services/

or write us here:

contact@airnguru.com


137 views0 comments

Recent Posts

See All