← All work

Kayak · Travel · Sponsored project

Bringing group travel to Kayak

Planning a trip with friends means juggling budget, dates, and location across scattered chats, and Kayak has no group travel features today. I led a team of nine UX designers through research to design them.

Role
Team leader
Team
UX designers with varying skill levels
Platform
Desktop web
Timeline
~16 weeks
kayak.com/trips

Overview

KAYAK helps travelers search for flights, stays, and transportation, but its Trips experience is designed for individuals. Planning with friends still means juggling budgets, dates, and destinations across scattered conversations.

Over 16 weeks, research into group travel behaviors revealed how friends negotiate decisions, where KAYAK's existing experience falls short, and how collaboration could fit naturally into Trips.

The result was a shared planning experience built around suggestions, voting, flexible decisions, and group coordination.

The Problem

Everyone wants a trip. Nobody can agree on one.

Planning a trip with friends means balancing different budgets, schedules, and destinations. Ideas get scattered across group chats, people search independently for the best deals, and one person usually ends up organizing everything.

Reaching a decision is often harder than finding the options.

KAYAK helps people search and organize travel, but doesn't give groups a way to make those decisions together.

Process

Milestone 1Who is the primary user?

The project began with a broad question about who to design for. Secondary research into travel trends pointed to Millennials and older Gen Z as key audiences. It also highlighted common challenges in group travel: conflicting priorities, difficult compromises, and the need for someone to take the lead.

Eighteen interviews across friends and families revealed how those challenges play out. Most participants preferred traveling with friends, while family trips typically relied on one organizer. Interview synthesis highlighted recurring patterns in communication, planning processes, priorities, and challenges. Together, these findings narrowed the focus to friends traveling together and raised the next question: how do these groups actually plan?

18interviews
9 friend groups, 9 families
9team members
2 interviews each
2user groups
friends and family
Key insight

Groups rely on one organizer but still want a voice in decisions. The design balances collaboration through suggestions and voting with one organizer having the final say.

Milestone 1 sketches

Sketching ran alongside the research throughout, so these were drawn at this stage of the project.

Sketch: Budget and availability overview
Budget and availability overview
Sketch: Group travel home page
Group travel home page
Sketch: Planning process ideation
Planning process ideation

Milestone 2How do groups travel?

With the audience defined, the next step was to understand how friends actually plan trips together. A Reddit deep dive reinforced challenges from the interviews, including scattered communication, plans falling through, and reliance on multiple planning tools. Contextual inquiry revealed a deeper pattern: planning isn't linear. Budget, dates, and location influence one another, requiring groups to revisit decisions until they find a plan that works for everyone.

Journey map from discovery to exit. Budget, dates and location are decided together during planning, with constant communication throughout.PlanningSearch tripsEveryone searches on their own, so ideas end up scattered across tabs and chats.PlanningBudgetEach person has a different number in mind, and nobody wants to ask first.PlanningDatesAvailability rarely lines up without one shared view of everyone's calendar.PlanningLocationSuggestions pile up in the chat, with no easy way to compare them.PlanningVoteThe group needs a quick way to see what most people prefer.PlanningDecideOne option gets locked in for the whole group.
Journey map: planning isn't linear, so budget, dates and location are decided in sync. Hover a step for context.
Key insight

Budget, dates, and location are decided together. Travelers often start with a destination but reconsider when the budget or timing doesn't work. The trip overview therefore connects all three through linked tabs, allowing groups to revisit decisions as plans evolve.

Milestone 2 sketches

Sketching ran alongside the research throughout, so these were drawn at this stage of the project.

Sketch: Budget sliders by category
Budget sliders by category
Sketch: Every collaborator inputs a budget
Every collaborator inputs a budget
Sketch: Group leader controls the budget
Group leader controls the budget

Milestone 3How can KAYAK improve?

With group planning behaviors understood, the next question was how KAYAK could better support them. A competitive analysis of eight travel platforms revealed collaborative features like shared calendars, voting, and itineraries that KAYAK lacked. A heuristic evaluation exposed additional friction: creating a trip was difficult to discover, and saving travel options wasn't always straightforward. These findings informed early sketches exploring shared calendars, group budgets, and location-based planning.

01Competitive analysisCompetitors had shared calendars, voting and map itineraries that KAYAK lacked.

Eight competitors across trip planning and booking were reviewed for the group features they offer. Google Drive's sharing model was also a source of inspiration for inviting collaborators.

Wanderlog Troupe Google Flights Expedia MakeMyTrip Goibibo Airbnb Vrbo
BudgetTrack who has paid
LocationMap with a day-by-day itinerary
DatesShared availability calendar
CollaborationVoting to settle disagreements
02Heuristic evaluationGroup planning was buried, and hotels could not be saved to a trip.
Annotated KAYAK screenshots from the evaluation: an ad for a hotel that cannot be saved to a trip, a hard-to-find Create Trip button, and a complex booking flow.
Nine evaluators annotated KAYAK while planning a trip as first-time users.
  1. Hidden entry point"Create Trip" sat in a corner menu.
  2. Limited itinerary searchActivities searched by location only, with no dates or times.
  3. Dead-end adsHotel ads could not be saved to a trip.
03Feature sketchingSketches for a shared calendar, group budget and location map.
Sketch of a calendar feature.
Sketch of a budget interface.
Sketch of location exploration with a map.
CalendarEveryone adds availability in one place
BudgetSliders that roll into a group budget
LocationMap for exploring stays and activities
Key insight

KAYAK lacked the collaboration features found in competing tools, while existing planning features were difficult to discover. This shaped a more accessible Trips experience centered on shared budgets, suggestions, and voting.

Milestone 4Design and test

Research findings shaped early sketches and prototypes, which were evaluated through concept testing. Testers preferred the linked-tab trip overview, which made different parts of the trip easier to navigate. Testing also revealed usability issues, including unclear collaborator roles and permissions. These findings guided refinements to the experience, making group planning more intuitive and easier to manage.

Tab overview over a gridTesters preferred seeing every part of the trip as linked tabs.
Fixed
Clearer rolesAdmins and viewers were hard to tell apart, so role indicators were added.

Milestone 4 sketches

Sketching continued through design and testing, so these were drawn at this stage of the project.

Sketch: Budget problems and trip flow
Budget problems and trip flow
Sketch: Entry points for budget planning
Entry points for budget planning
Sketch: Alternate budget layout
Alternate budget layout

The Solution

The final design brings group planning into KAYAK's existing Trips experience. Friends can suggest destinations, compare availability and budgets, vote on options, and build an itinerary together. Because plans constantly evolve, budget, dates, and location remain flexible until finalized. Clear collaborator roles give everyone a voice while keeping one organizer responsible for final decisions.

Here's how it works.

Compare my original screens with a refined, kayak.com-style pass.
01

Start a trip

Choose a group or solo trip up front, so planning together is the default rather than an afterthought.

02

Invite

Invite friends as editors or viewers, so everyone knows who can change the plan.

03

Overview

Every part of the trip sits in one linked view with a clear status, and any part can be changed at any time.

04

Location

Members suggest destinations on a map, then the group votes and the organizer finalizes.

05

Dates

A shared calendar shows each member in their own color, so overlapping availability is easy to spot.

06

Budget

Each person sets their own budget by category, then sees how it compares across the group.

07

Itinerary

The trip laid out day by day, alongside the map.

08

Add activity

Anyone can add food or an attraction to a day, without leaving the itinerary.

Try the interactive prototype in Figma ↗

Impact & learnings

My impact

LeadershipLed a mixed-skill team

Set the plan for each milestone and used mentorship and regular critiques to keep UX designers of varying experience moving together.

StakeholdersKept KAYAK in the loop

Ran stakeholder meetings to share progress, gather feedback and keep the direction aligned with the product.

Research to designTurned findings into features

Interviews, a Reddit deep dive, a contextual inquiry and a competitive analysis shaped suggestions and voting, a shared group budget and a linked-tab overview.

TestingPlanned and ran concept testing

Testing confirmed the overview and exposed problems, like unclear collaborator roles, that I redesigned.

What I learned

Research can overturn assumptions.

I started with planning as a linear flow. The contextual inquiry showed that budget, dates and location shape each other, and the design changed to match.

Designing inside an existing product takes restraint.

New collaboration features had to feel native to KAYAK, so consistent patterns and visual language mattered as much as the ideas.

Leading is mostly clarity.

Clear goals, shared feedback habits and room for everyone to contribute got more from a mixed-skill team than tighter control would have.

Users and business both count.

Prioritizing the changes that helped both most kept a short timeline focused.

Limitations & next steps

Limitations
Next steps

The timeline limited scope. Not every planned feature could be designed in the time available.

1Design the remaining features that didn't fit the timeline.

Voting was under-tested. Concept testing centered on the overview and collaboration flow.

2Test voting on stays, flights and transportation, within KAYAK's design system.

Not every traveler was covered. Kids and pets weren't part of the design.

3Extend to more groups, including families with kids and trips with pets.

KAYAK is a metasearch. With no direct booking, a group can plan together but not book in one place.

4Explore handoff to booking for a finalized group plan.