Flight Search UX

They fill it in, wait, get a wall of near-identical results, change one field and wait again. Two tabs later they are on Google Flights.

Every one of the world's 50 largest airlines uses the same search pattern. That agreement is easy to read as evidence that the pattern works.

What This Module Is

The other eight modules each took one domain and asked what it changes. This one takes a single flow and runs the whole method on it: diagnose what people experience, find the constraints underneath, redesign inside them, and name what you gave up.

The redesign at the center of it is real work with a name. Robert Goesch, co-founder of the agency DUMBO, published it in Smashing Magazine in July 2023 as the Balancing Act. It is a fictitious case study, developed on the team's own initiative and never launched, and the article says so.

That changes how to read it rather than whether to. The reasoning is worth studying closely and the screens are not a spec to copy, because nothing here has been tested against live booking data.

Anatomy of an Itinerary

Two rows in a result list both say 312 euros and eight hours. One is a single aircraft; the other is two airlines on separate tickets with a 55-minute connection and no checked bag.

They are different products at the same price, and the interface presented them as comparable. Understanding why requires the vocabulary the industry actually uses.

The Five Terms

The segment.One takeoff and one landing on one flight number. The atom of the whole system.

The leg.One direction of the journey, made of one or more segments. A return trip has two legs; the outbound leg with a stop in Paris has two segments.

The layover.The gap between two segments. The number that matters is the minimum connection time at that airport. It varies by terminal and by whether the flights are international, and a connection under it will not be sold.

The fare basis.The rules attached to the price: whether it can be changed, refunded, what baggage is included, whether a seat can be chosen. Two seats on the same aircraft can carry different fare bases and differ in price by a factor of three.

The ancillaries.Everything charged separately: bag, seat, meal, priority, change fee. On low-cost carriers these routinely exceed the fare itself.

Why the List Is Hard to Compare

Three properties make a results list hard to read, and none of them is a layout problem.

The advertised price is not the product.A fare without a bag is a different offer from one with a bag, and the sort puts them next to each other. This is the late shipping cost from E-commerce UX, arriving earlier and with more variables.

Duration hides risk.Eight hours with a 55-minute connection and eight hours with three hours in Frankfurt carry different odds of arriving. The list shows the same number.

Separate tickets are not one journey.Two segments booked on one ticket make the airline responsible for a missed connection. On separate tickets nobody is, and the interface almost never says which one it just sold.

What a Result Row Owes

The anatomy tells you what a row has to carry to be comparable at all.

The total with bags for the traveler's actual case. The door-to-door duration including the layover. The number of segments and where. The connection margin, flagged when it is close to the minimum. And whether it is one ticket or two.

Five facts. Most result lists show two of them, which is why the refinement rounds happen.

The Two Questions Behind Everything

Underneath all of it, somebody is trading money against time, and certainty against both.

Every term above feeds one of those three quantities. A design that surfaces the five facts lets people make that trade on the screen. One that shows a price and a duration is asking them to make it in another tab.

Pro tip: Take the top five results from any flight search and write out the five facts for each by hand. The time it takes you is the work your interface is currently handing to the traveler.