UX challenge: how we break free from the endless loop in flight searches.

The topic of ‘flight search’ has been on our workbench before. Back in 2015, part of us worked on the design strategy of Lufthansa Group. In 2017, airberlin became one of our first clients. Together with the team, we redesigned their digital world from scratch: flight search, booking process, homepage, and much more.

What was considered too progressive in 2016 celebrated its first successes in 2017. Six years later, in 2023, it is now being expanded as a case study.

If you, like most, have searched for a flight at some point, you are familiar with the usual song and dance involved with playing with the search criteria in order to score an optimal search result. If I change the travel date, will it be cheaper? If I depart from a different airport, will the flights be less-full? As a result, the hunt is a never-ending combination of viewing results, making further refinements, and constantly changing the search criteria. So, do you see yourself here? Twenty-nine students who took part in the study “Online Search Behaviour in the Air Travel Market: Reconsidering the Consideration Set and Customer Journey Concepts” certainly did.

According to the 2017 study, flight search algorithms cover a dynamic solution space with often more than a thousand possibilities that can change rapidly, cutting it down based on user criteria. Users, on the other hand, go through three rounds of refinement on average, filtering and refining search criteria, comparing possibilities, and making trade-off judgments. The analyzed flight searches are insufficient to support the final decision: users must still make judgments and trade-offs depending on their personal preferences and priorities. The study emphasizes the importance of improved interfaces with decision support up to the final decision in order to improve the flight search experience.

The classic flight search pattern inevitably leads to a frustrating trial-and-error loop.

The dilemma: this price-relevant information affects availability, travel time, and service. At the same time, they are factors for the traveler that can be changed depending on the result and personal preferences and flexibility. As a result, travelers develop their own user hacks to compare different search parameters and weigh the trade-off between price and convenience.

How can we give travelers a better flight search experience? Our pitch is The Balancing Act: a guided dialogue between traveler and airline. Strap in — we’re taking a deep dive.

What makes a search successful? It’s an increasingly important question in the age of global travel and its limitless possibilities. We focus on finding your personal solution. It puts the traveler, their occasion, and their budget at the center of the interaction and looks at how well the flight offer fits. To do this, we fundamentally change the tailoring of the interaction with travelers. We break down the search form into individual tasks and change the sequence of interactions. This allows a more balanced approach between friction and progress.

To redesign the interface, we need to uncover the structure of the interaction moment. For this, we use the Interaction Archetypes framework to help us align our design with the underlying usage intention — the strongest driver for user interaction.

What we currently see in the airline and flight industry space is that search now assumes that its parameters meet fixed criteria. Accordingly, a “successful” search is given if it is able to deliver a result based on the ten declarations. In interaction, however, travelers behave in a way that contradicts this assumption.

The decision-making levels through which travelers approach their destination provide information about this. They are hierarchical and causally related.

So if we create space for trade-offs through interaction, we should be able to make the flight search more targeted to the traveler’s needs. This raises three major design opportunities:

1. How might we utilize travelers’ decision-making levels to speed up the process?
2. How might we help travelers balance price and convenience to reduce search queries?
3. How might we deliver results to travelers faster to reduce friction and abandonment rates?

We say goodbye to the predominant route indication of a flight search and ask in the first step: Where do you want to travel to? We quickly realized that the “where to” question fits the mental model of travelers and can serve as a springboard for goal-oriented interaction. Only if we bring travelers closer to their destination can the airline make a relevant offer.

More Later

Further queries are refrained from in favor of the offer. All other criteria can be used to adjust the results while maintaining the flight schedule. These criteria are preselected based on the most frequent flight bookings or personal flight behavior.

• Number of adults,
• Number of children,
• Number of infants,
• Access codes to selected flights,
• Selection of class.

To make well-informed decisions, travelers need to be aware of the consequences of their choices in the flight booking process. This means they need to understand the impact of their partial decision (date, departure location, airport) on the expected outcome. The better they can do this, the easier it is for them to weigh up. Ultimately, the best flight is the result of a personal trade-off between convenience and cost.

The Best Ticket For You?

A “One-Way Flight” can be more expensive than a “Return Flight.” We consider the option “One-Way Flight” within the date selection. This is because it is an alternative to a return flight. And the associated price is an important piece of information to consider when travelers are weighing options.

Even before the flight plan has been loaded, we put all options on the table. This is how we offer maximum price transparency.

If we communicate flights and their prices prior to checking whether the flight is not yet fully booked or the prices have changed, there is a risk that the offer will have to be corrected. Usually, all of us want to make statements that can be fulfilled. But it needs a tolerance for errors in communication in order to provide volatile flight data as quickly as possible.

The following example: A flight is supposed to have a price of 300 euros, or so it was the day before. In the meantime, the prices have changed, and the flight costs 305 euros. As a result, the assumption based on the information was wrong and had to be corrected to the disadvantage of the customer. Stupid. But: One was already in a situation to give a price indication. After all, the flight before and after might cost 600 euros and is therefore even more irrelevant than a flight for 305 euros if one had assumed 300 euros.

The communication error is less important than the added value at the moment of interaction. We can only overcome technical and business constraints with the help of estimates and assumptions.

Geolocation

Geolocation data can be used to draw conclusions about the departure airport. We do not necessarily have to use the geolocation API for this. It should also be possible to achieve sufficient localization with the help of IP address search so that we can immediately create added value. Once we have identified the airports in the vicinity, we can evaluate potential connections in terms of cost and convenience.

Anyone who has ever had to search for and book a flight surely knows: it is nerve-wracking and time-consuming. Before you can even start the search, you have to enter ten details in the search form. This means that in the very first stage of the search, you’ve already had to make ten decisions. Unfortunately, and often, only one thing is certain: the destination of the journey.

Along our thought process, we have shown that the classic flight search pattern is broken, often because external factors such as technical and business constraints influence the flight search experience. However, we have shown that airlines and searches can break the pattern. This can be achieved by entering into a dialogue with the user and leading them from one decision level to another to eventually fit their specific needs and goals.

If you found this approach useful or interesting, I recommend our guide »UI Design for Humans«. It is based on our “Interaction Archetypes” framework that allows you to strategically illuminate a user’s behavioral patterns, as well as their interactions with digital interfaces. It has greatly transformed and improved our design process. We hope that it could transform your design process, too.