The UX Of Flight Searches: How We Challenged Industry Standards — Smashing Magazine

Why Flight Search Feels Broken

For most travelers, booking a flight is a tense negotiation with a search form. The familiar loop — adjust a date, check another airport, refresh to see a new price — is so common that it has become an accepted part of the process. Yet the underlying structure makes this trial-and-error approach almost inevitable. A 2017 study of 29 participants in the air travel market found that users typically go through three rounds of refinement, filtering and comparing options before making a decision. The search interfaces themselves rarely support the final judgment; travelers must constantly weigh personal preferences against available trade-offs.

The pattern is so entrenched that criticism can feel futile. But when examined closely, the standard flow reveals a set of constraints — some visible, many not — that actively work against a smooth decision-making process.

What Travelers See (And What They Do About It)

Observing real booking behavior surfaces six recurring problems:

• The search itself is stressful. High prices, artificial scarcity, and a dense field of options put pressure on every query.
• Prices seem to rise with each search. Opaque pricing and fears of cookie-based tracking leave travelers suspicious that revisiting a route costs them money.
• Options are overwhelming. A single route like Frankfurt to Honolulu can present 8,777 combinations. Many users abandon the airline site entirely and turn to third-party aggregators such as Swoodoo or Google to make sense of the field.
• Waiting compounds the frustration. In our tests, a single query routinely took 10 seconds, with prices recalculated and availability rechecked on every request.
• Price dominates every decision, but its drivers are hidden. Every search parameter affects cost, yet the interface rarely communicates how.
• “From” prices set a trap. When the advertised low fare isn't available for a specific search, it creates negative anchoring — a flight that originally seemed affordable now feels overpriced.

What Happens Behind the Screen

The search experience is shaped less by usability engineering than by three systemic forces:

• Amadeus and third-party reservation systems. Most searches run on infrastructure that determines which data points are exposed and how the interface is structured. Airlines operate within these constraints and have limited room to redesign the experience.
• Dynamic pricing. Fares fluctuate in real time based on booking behavior, competitor pricing, events, and demand. This volatility is a core reason searches must repeatedly revalidate results.
• Cost per request. Each search query carries a cost for the airline. To control expenses, airlines minimize the number of queries — which discourages both proactive suggestions and iterative refinements.

Reframing the Problem

The current search model presupposes that a traveler can enter all price-relevant information — dates, airports, flexibility — before ever seeing a result. That assumption is rarely true. Availability, travel time, and service level all depend on parameters that travelers only know they want to change once they see what the trade-offs are.

The result is a dialogue between traveler and machine built on guesswork. Our counterproposal, called “The Balancing Act,” turns the search into a guided conversation. It places the traveler's occasion and budget at the center and restructures the flow so that each step informs the next, balancing friction against forward progress.

A Search Built Around Decision Points

Start With Destination

The only question with a fixed answer is where the traveler wants to go. Starting there lets the interface frame every subsequent choice — cost, convenience, timing — in terms that matter for that specific journey.

Open Up the Departure Point

Instead of assuming the closest airport is the best, the system can suggest alternative departure points when they yield better prices or connections, based on where the traveler is staying and how flexible they are.

Use Timing as a Lever, Not a Lock

Not every trip is date-fixed. Best-price calendars, expected load factors, and travel time comparisons give flexible travelers a way to see how timing affects both cost and comfort before committing to a specific day.

Delay the Administrative Details

Passenger counts, promo codes, and multi-stop configurations can wait. The traveler should first learn whether a suitable flight exists at all, then handle the logistics.

The Structure of the Interaction

To redesign the interface, we analyzed the underlying moment of interaction using the Interaction Archetypes framework, which aligns design with the user's dominant intention.

The task — finding a suitable flight — is clearly a weighing phase. Travelers compare routes, times, and prices against planning criteria and personal limits, often across multiple platforms. This maps to the “Act” usage intention: users have a specific task and want to make headway as efficiently as possible. They take a structured approach, adjusting parameters selectively to probe the boundaries of what is available.

A closer look shows that flight search is not a flat comparison but a hierarchical step process, similar to an Analytic Hierarchy Process. Decisions are made sequentially, and each level is causally linked to the next. The travel decision itself follows the same pattern: destination, timing, duration, and budget are already settled, at least tentatively, before the search begins. Travelers carry this hidden agenda, consciously or not, through every query.

The solution space that emerges points to three hypotheses:

• If the search is designed along these decision levels, travelers can make faster, more confident choices at each stage.
• If travelers can weigh options at the point of entry — balancing price against convenience before seeing results — the first query is more likely to return a usable flight, reducing resubmissions.
• If partial information appears as soon as it becomes available, travelers can scan and abandon less often, cutting friction and drop-off rates.

This restructured search doesn't remove the complexity of flight booking. It simply acknowledges that the complexity lies in the traveler's own priorities — and gives the interface a way to surface them before the search, not after the first disappointment.

Why Search Forms Fail Travelers

Most flight searches treat the traveler's input as fixed criteria. Ten fields must be completed before a single result appears, yet real-world travel decisions don't work this way. Travelers reach their destination through a series of hierarchical, causally related decisions, each one a compromise between price and convenience.

Each individual decision is the result of a trade-off between price and convenience. A successful search is, therefore, the smallest compromise.

By designing for trade-offs through the interaction itself, the search can align more closely with how travelers actually plan. That reframing opens up three design problems worth solving:

1. How can the search mirror a traveler's decision-making levels to make the process faster?
2. How can balancing price with convenience reduce the number of queries a traveler makes?
3. How can results arrive sooner to prevent abandonment?

A Sequential Path To Flight Selection

Instead of the standard route form, the search can begin with a single question: Where do you want to go? This matches the traveler's mental model and gives the conversation a clear goal — an airline can only make a relevant offer once the destination is known.

Adopting this approach means removing the search form entirely and leading the traveler through a dialogue following their natural decision levels. Four pieces of information, collected in order, are enough to build a suitable flight plan:

1. Destination.
2. Origin.
3. Departure time.
4. Return time.

Concentrating on each field individually lowers cognitive load and frees up space for content, even on small screens — provided each input costs less effort than the value it returns. Structuring the questions this way lets the interface orchestrate information along the traveler's decision-making hierarchy, creating a series of partial successes:

1. Can we fly to the destination?
2. Can we fly from a suitable departure point?
3. Can we fly out at the appropriate time?
4. Can we fly back at the right time?

Everything else is deferred until after the flight plan is shown. Criteria like traveler count or cabin class are applied to adjust the results,preselected based on the most frequent bookings or the user's history:

• Number of adults.
• Number of children.
• Number of infants.
• Access codes to selected flights.
• Selected travel class.

Guiding Each Decision Rather Than Trying To Guess It

Travelers can only weigh options intelligently when they grasp the impact their partial choices — date or departure point — have on the eventual outcome. The better they understand that trade-off between convenience and cost, the more successful the result.

The Most Suitable Departure Airport

Transport is flexible. In western Germany, for example, a 90-minute radius includes both Frankfurt and Düsseldorf, two substantial hubs. That latitude brings questions about choice, preference, mobility and price — questions the interface can help answer. Using geolocation data and the airline's route network, the search can rank nearby airports by a mix of comfort and price, considering travel time and airline as well.

That ranking also gives airlines room to work. By placing targeted offers, they could steer travelers toward options that balance demand across their network, potentially using discounts to influence decisions.

The Right Travel Dates

Travel dates obscure the decision because they have the biggest effect on price. Shifting the outbound flight by a day can change the fare by hundreds of euros, which is why the date selection accompanies price information — additional indicators show prices for outbound and return journeys for each day. That way, travelers don't have to redo the search later if their assumptions turn out to be wrong.

Availibility also signals early. Days on which the destination isn't served or that show exceptional load factors get flagged, letting travelers factor supply problems into their plan before they become a painful surprise.

Considering A One-Way Trip

Prices don't behave intuitively — a one-way flight can easily exceed the fare for a round trip. Because this option is a real alternative in the decision, the price should still appear during date selection, providing context whether or not the trip plan includes a return flight. Placing all options on the table before the flight plan loads offers maximum price transparency.

Note: This case study does not consider multi-stop flights.

Trading Certainty For Speed Where It Counts

Precision has a price. Real-time flight availability and live fares come with processing times and per-call costs. Sometimes maintaining a definitive answer prevents a fast one. Rather than insisting on perfect certainty, the search must work with a degree of tolerance for error so volatile flight data can be shared quickly.

A fare of 300 euros shown in last night's cache becomes 305 euros at booking. Slightly wrong, yet still vastly more useful than no information at all — especially when those near misses matter to a traveler weighing their next step. The value at the moment of interaction outweighs the precision loss; estimates and assumptions are the price of overcoming technical and business constraints.

Caching Prices

True price transparency depends on caching. Price estimates are taken from previous searches and used to display up-to-the-minute approximations without triggering a new, costly calculation. These estimates can serve until the flight is confirmed. Fares may shift once the final flights are selected, but that specific accuracy is sacrificed for fast answers and better selection criteria. Price is too essential to every trade-off to leave out for want of a few moments.

Running Availability Checks Off To One Side

Put explicit checks of live availability at the end of the interaction. Route data and travel times make the flight plan available immediately; its verification either happens downstream or in parallel. This allows the systems to communicate without getting in the way.

Reasoning From Location

Location data offers a reasonable inference about the home airport, not necessarily through the precise geolocation API — even an IP-based lookup can provide clues about location. Once the nearby airports are known, the relative cost and convenience options can be evaluated.

Redesigning The Search Around Trade-Offs

The default assumption in airline search is rare and often unjustified: a flight deal only comes into view if all ten fields equal a fixed, predefined truth. Yet frequently, only the destination is certain to coincide.

The standard ten-field form fumbles this reality. However, the balance between convenience and price can be respected more effectively once the industry acknowledges the flight search can be remodeled. Approaching a search as an evolving dialogue — moving users from one decision level to the next — that reflects their individual interests and constraints is a workable path for results that genuinely match the traveler.