Preference-Driven Air Shopping With Targeted One-to-One Shopping Responses

Travelers won’t spend long periods of time on a single website searching through long lists of itineraries to find the perfect flight and fare. They will visit many websites, shopping and comparing, until they find that perfect itinerary. A new shopping algorithm that will soon be available to airlines will change the way travelers shop, and it can help convert shoppers into customers.

Typically, a traveler plans a trip by visiting a travel website such as an online travel agent (OTA) or airline.com and submits a shopping request to evaluate available itinerary options.

During the shopping process, travelers look for a good balance among schedule and fare attributes they consider relevant for their trip. Due to the large number of schedule attributes that can define an itinerary — brand, departure time, arrival time, number of stops, codeshare, interline, connection quality, elapsed time, etc. — today’s websites generate a large number of itineraries and leave the problem of selecting a single itinerary from the long list to the traveler.

For example, OTAs strike a balance between low-fare efficacy and itinerary diversity to maximize conversion rates and generate anywhere from 700 to 1,600 itineraries for a single shopping request.

Most airline websites, with the notable exception of a few mega carriers, display an exhaustive list of itineraries for each shopping request. With the growing adoption of mobile and tablet devices for travel, displaying a large number of itineraries is cumbersome, it negatively impacts the user experience and it leaves the task of selecting the right itinerary to the traveler, which negatively impacts conversion rates.

Quite to the contrary, when a traveler makes a request to a travel agent, the customer typically states his or her preferences, and the agent proposes a single itinerary based on the stated criteria. The itinerary proposed by the travel agent is usually the best itinerary.

However, the shopping experience for the airline customer is about to change, and airlines stand to convert more customers as a result.

Coming soon, Preference-Driven Air Shopping for airlines is a new shopping display algorithm based on a traveler’s relative importance of schedule and fare attributes to determine the best itinerary. Travel website filters cannot solve this problem since a filter would exclude an itinerary based on one attribute, even though it would have been outweighed by the goodness in the other attributes.

Schedule And Fare Attributes

Preference-Driven Air Shopping enables airline travelers to explicitly expose schedule and fare attributes as preferences. A travel shopper can further customize the relative importance of each attribute and also select specific attribute value(s) where ever applicable.

The new algorithm is based on TOPSIS, a Technique for Ordering Preferences by Similarity to Ideal Solution, and ranks itineraries based on a traveler’s relative trade-off between schedule and fare attributes. It scores all itineraries in the shopping response based on the trade-off evaluation using relative weights (importance) of the schedule and fare attributes specified by the traveler and displays the itineraries in order of the scores.

The application requires a shopping cache with pre-stored itineraries by market, departure date and length of stay to provide the shopping response. The display algorithm can be applied to a regular OTA, where the shopping cache will comprise multi-airline itineraries or to an airline.

Travel Dates As Preferences

Using the Preference-Driven Air Shopping algorithm, travel dates are modelled as preferences. As such, travelers can specify multiple travel dates and rank them relative to the other selected preferences.

For example, a traveler can now specify the preferred departure-time window (such as 9 a.m. to 11 a.m.) and willingness to pay a little more for this convenience. What differentiates this from the typical sort/filter is that if there is a relatively inexpensive flight that departs at 8:50 am, it would still score high on the priority scale and get displayed higher on the list.

Moreover, the service model includes both travel dates and lengths of stay as additional preferences, so the traveler can now specify multiple travel dates and lengths of stay and rank each of these relative to the other selected preferences. In effect, the traveler can pose the request: “Find me the lowest priced itinerary for a weekend trip in July.”

Three Top Itineraries Displayed

Of the 500 itineraries returned in a specific shopping query based on a traveler’s selected and prioritized preferences, only the top three itineraries (outbound and inbound) will be served to the shopper, with an option to display all itineraries.

The application has added intelligence that tracks the shopper’s preferences and implicitly uses them for subsequent shopping. It displays a single itinerary most relevant to the customer, enabling the traveler to quickly view the itinerary being offered and opt to change by specifying the reason.

For instance, the traveler can request to find an itinerary with a different airline. The application would then use and track this preference and try and find a more suitable itinerary. Out of the 500 itineraries returned, only the top itinerary would be displayed. Therefore, the traveler would arrive at the most suitable itinerary more easily than searching through a long list of candidates.

If the displayed itinerary is not satisfactory, the traveler can ask for a different itinerary by selecting the “Change” button. The traveler can also ask for a cheaper fare, an earlier or later flight, a shorter connection, etc.

The iPad version of the application provides explicit controls for the shopper to specify preferences and their relative importance and displays the three most pertinent itineraries, with an option to view all itineraries. The traveler’s preferences are stored so his or her next shopping experience will receive more relevant results.

Today’s travelers are both time and cost sensitive. When shopping for flights, they generally won’t spend a lot of time on a single website arriving at their desired results. Rather, they are quick to jump from website to website until they find the most ideal flight for the most reasonable fare. Preference-Driven Air Shopping will make shopping for air travel easy and satisfying. The solution is yet another mechanism airlines can use to convert shoppers to customers.

Top Shopping Results On iPhone

Using Preference-Driven Air Shopping from an iPhone, the top three favorable results would appear first, with the ability to scroll to see more or all itineraries that were served up based on the traveler’s preference criteria. If the displayed itinerary is not satisfactory, the traveler can ask for a different itinerary by tapping the “Change” button.

Mobile Device Stores Preferences

The iPad application provides explicit controls for the shopper to specify preferences and their relative importance. The preferences are then stored in the traveler’s mobile device so he or she can receive more relevant results for future trips.

Table of Contents