The Situation
One of our travel clients operates a booking platform where travelers choose their seats using a graphical cabin diagram. Seats are displayed as small clickable shapes, with color used to communicate price and availability.
For mouse users, the experience felt quick and intuitive. But seat selection was also the last major step before payment, which meant a failure here could stop the entire booking.
The Challenge
The client was seeing some travelers struggle to complete seat selection independently.
When we reviewed the experience, we found that the seat shapes were built from non-interactive elements with click handlers. They could not be reached with the Tab key and did not have meaningful accessible names.
Availability and price were also communicated primarily through color.
As a result, screen reader users received little useful information about individual seats. Keyboard users could not select a seat at all, and voice control users had difficulty identifying and activating the seat they wanted.
For travelers with low vision, zooming into the cabin map also made it difficult to maintain context while navigating the diagram.
A visually polished seat map had become a barrier at one of the most important points in the booking journey.
What We Found
Our manual review identified several accessibility issues:
- Seats were missing meaningful roles, names, and states, affecting SC 4.1.2, Name, Role, Value.
- Availability and pricing relied too heavily on color, affecting SC 1.4.1, Use of Color.
- Seats could not be operated with a keyboard, affecting SC 2.1.1, Keyboard.
- Seat selections and related updates were not consistently communicated to assistive technologies, creating issues with status feedback.
- Some seat targets were too small, creating difficulties for users with limited dexterity and raising concerns under SC 2.5.8, Target Size (Minimum).
Individually, these looked like component-level problems.
Together, they meant some travelers could reach the final stages of booking and still be unable to choose a seat independently.
How We Addressed It
We kept the graphical seat map because it remained useful for many travelers, but rebuilt the interaction underneath it.
Each seat was changed into a meaningful interactive control with an accessible name that communicated the information needed to make a decision, such as row, seat number, location, availability, and relevant seat features.
Seat availability and pricing were no longer communicated through color alone. Text labels and a clearer legend provided the same information without relying solely on visual differences.
We also introduced a list view of the available seats, allowing travelers to browse and filter seats by type and price without using the graphical cabin map.
Seat selections and relevant price changes were communicated to assistive technologies, while focus visibility and target sizes were improved across the interface.
Finally, we retested the complete seat-selection journey using keyboard navigation, screen readers, and voice control rather than reviewing the seat map in isolation.
The Outcome
After remediation, the client saw a measurable improvement in travelers' ability to complete seat selection independently.
In follow-up usability testing, successful seat selection among keyboard, screen reader, and voice control users increased from 63% to 79%.
Accessibility-related support requests connected to seat selection also decreased by 14% during the comparison period.
Travelers were better able to identify available seats, understand pricing and seat features, make a selection, and continue to payment without relying on a mouse or assistance from support.
For the client, the improvements reduced friction at one of the final stages of the booking journey while preserving the graphical seat-selection experience.
What We Learned
The seat map itself did not need to be removed.
The problem was that its visual design had been built without an equivalent interaction model for people using other input methods and assistive technologies.
Once the seat information, keyboard behavior, selection states, and alternatives were considered together, the experience became easier to use without changing the core booking flow.
For this client, accessibility meant keeping the visual experience people already liked while making sure it was not the only way to successfully choose a seat.
Closing
A graphical picker can look complete while still missing the semantics and interaction patterns some travelers need to use it.
If your product includes a seat map, room selector, or similar visual picker, talk to an accessibility expert about testing the complete journey before those barriers reach your customers.
.png&w=3840&q=75)
(1).png&w=3840&q=75)
