Jetstar
Revamping the ‘Meals’ Experience.
The project at a glance
-
Client.
Jetstar is an Australian low-cost airline headquartered in Melbourne.
-
Product.
Jetstar.com.au is an e-Commerce website for booking international and domestic flights.
-
Team.
UX Designer , UI Designer, Business Analyst, Quality Analyst, Scrum Master, Product Owner, Front End Developer and 2 x Back End Developers (Offshore).
-
Duration.
6 weeks (3 sprints)
A/B Test (4 months) -
Users & Audience.
Australian passengers flying domestic and Long Haul International flights.
-
My Role.
UX Designer & Project Lead.
Our Challenge.
Improve the meals purchasing experience for groups on multiple flights whilst clearly communicating the value of our products.
-
Objective.
For the business:
Increased meal purchases and revenue over the next 6 months.
For customers:
Clearer understanding of the first and second meal service.
Improving the clarity of shopping cart totals.
Enabling quicker selections of meals and quantities for groups
Clearer dietary information (E.g. Vegetarian, Dairy free, Halal etc.)
Clearer error messaging.
Better imagery to increase uptake.
-
Outcome.
2 users flows for validation and improved components and features for A/B testing:
Meals page & Product cards.
Scrolling product tiles.
Product page.
Summary page.
Validation & A/B testing plan.
Results
Improved usability of key tasks for booking flow by 26%.
Increased customer satisfaction score by 28%
Increased meal uptake revenue by 23%.
Reduced bounce rate 20%
Reduced time on booking by 16%
Booking completions increased by 6%
Kickoff.
Getting the ‘basics’ right.
The majority of Jetstar’s revenue is from add-on purchases like extra baggage weight allowance, seats, and meals. In particular, Pre-flight meals account for the largest product uptake each year.
It was crucial to the business that we addressed previous usability testing issues and continually improve the purchase experience.
Objective.
Starting with the Product Owner, gain a deep understanding of the business context and strategies to create actionable next steps.
Outcome.
Insight into the work already completed allowed us to understand complexities or gaps.
Understand the product vision.
Immersion into the customer segments and analytics.
Clarity around success measures and outcomes.
Preparation for customer interviews.
Scheduled activities and developed a project plan.
Approach.
-
Research.
1 Sprint (1st week)
Kickoff session
Research recruitment
Usability testing
Landscape review
-
Synthesis.
1 Sprint (2nd week)
Research report
Current state user flow
-
Ideation.
2 Sprints (2 weeks)
Ideation Workshop
Sketching
Future state user flow
Wireframing
-
Test & Learn.
4-6 Months
A/B testing
Design iterations
Visual Design
Design system contribution
Post-build validation
Research.
Usability Testing the existing experience.
A table and legend showing PURE scores in the research report. A single step (1.a) was rated a red 3 because it was difficult for the user.
Objective.
Customer needs and pain points in finding, evaluating and selecting meals
Expectations across desktop and mobile
How Jetstar compares to other meal ordering services - and what expectations exist in the market for meal/food ordering
Opportunities offered by non-Jetstar meal ordering services
Outcome.
We tested the existing live site with 6 participants and were rated using PURE scores (Pragmatic Usability Rating by Experts) on key tasks. This was to set a benchmark to compare new designs.
Quick wins and opportunities focusing on:
Flight navigation
Selection and editing meals
Summary and Cart
Product card content
We discovered 4 main problem areas.
-
![]()
1. Flight navigation lacked contextual details.
Passengers needed information on departing times and duration to help set context for the time of day a meal was to be purchased E.g. Breakfast, dinner or a snack.
Our recommendation
Include flight departing and landing times -
![]()
2. Meal selection was not efficient and flexible.
Users were being forced to select a meal for every passenger and for every flight which made selecting meals for a group confusing and “tedious”.
Our recommendation
Allow users to select a meal and assign it to any passenger they want. -
![]()
3. Summary and Cart lacked key editing functionality.
• Customers couldn’t edit or remove specific meals whilst on the summary page. Instead, they have are navigated back to the meals selection page.
• None of the participants clicked the Cart to confirm their selection because there was no visual feedback when adding a meal.Our recommendation
Include an ‘Edit’ button next to each meal item rather than at the passenger level. -
![]()
4. Product content and images weren’t appealing.
Participants felt the images were too small and not appealing. There were also usability issues where the ‘More info’ accordion expanded underneath the card and wasn’t seen.
Our recommendation
• Prioritise solving bigger usability issues
• Request an update on image assets
Landscape Review.
Comparing to competitors, meal ordering and delivery apps.
Screenshot of Landscape Review presented in Miro. Competitors were grouped by components under columns for easy comparison.
Objective.
Evaluate, compare and find new opportunities.
I compared against direct airline competitors, food ordering apps, and meal delivery services to gauge if they were solving similar problems in a better way.
Outcome.
We discovered better solutions to improve the following components:
Value proposition, Search & filtering, Imagery used, Product content, Navigation controls, Selection experience, Summary pages, and Editing options.
Ideation Workshop.
Vision alignment & collaboration.
A preview of some ideas generated by team members remotely on Miro.
Objective.
Align on what we’re doing & why, and generate diverse ideas.
Having a development team from 3 different timezones posed challenges previously with team misalignment and lack of buy-in on new features. Not only did we want to generate diverse ideas from different perspectives we wanted everyone to feel part of the process.
Outcome.
As a more engaged team we created more robust concepts and agreed on a direction forward.
Overall we achieved:
Team member introductions
Reviewed customer & business objectives
Shared inspiration
Aligned on our experience vision
Dot voted on principles to guide ideation
Reviewed research findings
2 rounds of Ideation
Presented refined ideas back to the team
Voted on ideas that best align with our objectives & experience principles
Agreed on next steps
User flows & Wire flows.
Focusing on how passengers select and assign meals
Before getting into any design, this step helped us understand where the potential pain points, opportunities and key decisions would be without investing too much time.
We wanted to visualise the customer’s steps and validate two different purchase flows. Therefore we used user flows and wire flows as a talking point with the client and team to validate whether customers prefer to:
‘Select a Meal’ first > Then assign to a passenger? Or do they,
‘Select a passenger’ first > Then ‘Assign a meal’?
Wireframes for Multi-Variate Testing
-
![]()
1. More useful information on meal Product Cards.
Because customers were unsure which meals were appropriate for their dietary requirements, if we included dietary symbols and ingredients then they would have more confidence in selecting a meal that’s right for them. Also using better food imagery would increase appeal.
-
![]()
2. Context is key for Flight Navigation.
I included flight departing and duration times to help customers decide if purchasing a meal was appropriate or not. Another interaction improvement to increase the visibility of the flight navigation was changing it to a scrollable tab rather than clicking a natural language dropdown link.
-
![]()
3. Selecting a meal for a flight then a passenger.
I wanted to validate whether customers select a meal first or select a flight first. It was important to not introduce contradicting interaction patterns and keep users’ cognitive load low. There were also business rules that constrained this experience that I had to validate with developers.
-
![]()
4. Adding 'Editing' functionality in the Meals Summary.
Customers can now edit an individual meal they selected instead of having to edit every meal under that flight. They can also reassign that meal to someone else. Again, this was validated internally first by the development team to see if it was feasible.
Planning for Multi-Variate Testing
A testing plan for PO approval.
Process.
Multiple metrics will be measured at once to understand how it affects other components, pages or products in the same flow.
I had to ensure that a data-driven and non-prescriptive hypothesis has been written with clear metrics to define its success.
After designs have been reviewed by a PO and the optimisation team, they will be sent live with monitoring/analysis completed by an external agency over 4 months.
Outcome.
A report will detail how the new designs compared to the original variate. Metrics are based on the percentage of uptake revenue, bounce rate, time on booking, booking completions and PURE scores.
Once results have been analysed and released, the hypothesis may need a further round of iteration and testing, or it may be approved for commitment to code after a final design review. The component would then be added to the Jetstar Design System.
Results
Improved usability of key tasks for booking flow by 26%.
Increased customer satisfaction score by 28%
Increased meal uptake revenue by 23%.
Reduced bounce rate 20%
Reduced time on booking by 16%
Booking completions increased by 6%