Jorge Molina

Case Study · Jun 2025

FlySplit

I designed a two-sided platform for legally sharing private-flight costs between pilots and passengers, grounded in mixed-method research and iterative testing.

Role

UX/UI Designer & Researcher

Duration

3 months · 300 hours

Tools

Figma · Optimal Workshop · Google Forms · UXPressia

Methodology

User-Centred Design (UCD)

Executive Summary

FlySplit is a native mobile app that connects student pilots with passengers to share the direct costs of private flights, reducing time-building expenses by an estimated 30–40% while opening private flying to a wider audience.

Over 3 months and 300 hours, I led the full UX process: legal desk research, 81-participant surveys, 8 in-depth interviews, tree testing with Optimal Workshop, and iterative high-fidelity prototyping. In the final usability round, all four participants (2 pilots, 2 passengers) completed every task, with SUS scores of 97.5 for pilots and 90 for passengers — directional given the sample size, but consistent with what the think-aloud sessions showed.

The Problem

Becoming a commercial pilot in Spain costs over €70,000. "Time Building" (accumulating flight hours) accounts for an estimated 30–40% of that cost. Despite strong demand for new pilots, this economic barrier blocks many candidates from entering the profession.

The Solution

FlySplit is a native collaborative-economy app (BlaBlaCar-style) that connects pilots with passengers to equitably share direct flight costs, fuel, fees, and aircraft rental, reducing training costs for pilots and giving passengers access to private flying at a shared, transparent cost.

Research

Legal Desk Research

A thorough analysis of AESA and EASA regulations confirmed the legal basis for cost-sharing in private, non-commercial flights. Pilots may recover direct operational costs, fuel, landing fees, aircraft rental, without the activity being classified as a commercial operation, provided no profit is generated. This legal clarity de-risked the concept before any further investment.

Competitive Benchmark

Wingly (EU market leader) and Coavmi were mapped as direct competitors. The benchmark revealed recurring gaps: opaque cost-split calculations, poorly communicated cancellation policies, limited customer support, and no in-app identity or safety verification. These pain points directly shaped FlySplit's core value proposition.

FeatureWinglyCoavmiFlyshareDreamairYumping
Chat (pilot ↔ passenger)
Integrated payment
Advanced search filters
Interactive flight map
Native mobile app
License validation
Pilot substitution (cancellation)
Cost split among participants
Custom cost per member
Aircraft hangar (pilot profile)

Validation surveys

A survey of 81 participants confirmed the viability of the flight-sharing model but highlighted critical needs around trust, safety, and predictability.

Sample demographics

Passengers62%
Pilots30%
Other8%

Top AI features demand

Weather cancellation prediction80% · 59 votes
Route & flight suggestions62% · 46 votes
Autonomous support chatbot53% · 39 votes

Top demanded features

Validation of licenses

92%
88%

Validation of identity

92%
80%

Clear cancellation policy

83%
84%

Pilot evaluation

88%
84%

Passenger evaluation

88%
66%
Pilots
Passengers

In-depth interviews

I conducted 8 in-depth interviews (4 pilots, 4 passengers). Affinity mapping surfaced the finding that shaped everything after: both sides wanted to share flights, but each was held back by a different fear. Pilots feared the tax agency; passengers feared for their safety.

The Pilot's Perspective

Accumulating flight hours is extremely expensive (>€70k), so pilots are eager to share costs. However, their biggest friction is the legal and tax risk of being wrongly accused of making a profit. They also have strict operational needs, like knowing the exact weight of the passenger for safety calculations.

"The tax agency (Hacienda), even if we split costs 50/50, doesn't understand it. We need legal guarantees."

— Commercial Pilot

"It sounds politically incorrect, but we always need to ask for the passenger's weight. It's crucial for the aircraft's weight and balance calculations."

— Student Pilot

The Passenger's Perspective

Passengers are highly motivated by the unique experience and cost savings compared to traditional flights. However, their main barrier is trust. They need absolute transparency regarding the pilot's experience, active licenses, and the aircraft's maintenance before getting on board.

"Mainly, I need to know they have the title that accredits them as a pilot and that they understand the airspace they are moving in."

— Potential Passenger

"I wouldn't mind, in the end you understand that in a small plane weight matters, so it's fine that they ask."

— Potential Passenger

Design action

Each fear mapped to a specific design response. The pilots' legal exposure drove the transparent cost-breakdown system, built to prove no profit is made (the legal line for cost-shared flights), plus mandatory license verification. The passengers' safety fears drove ID verification and an upfront weight-declaration step, because an accurate flight plan starts with accurate weights. None of these are generic trust features: each one answers a fear someone voiced in an interview.

Define

User Personas

Two rich User Persona cards capture biography, personality, lifestyle, skills, motivations, goals and frustrations for each key profile.

Pilot in training

PEDRO GUTIERREZ

“I need something reliable to reduce my expenses without sacrificing flight hours.”

Age

24 years

Gender

Male

Studies

PPL

Occupation

ATPL student

Biography

Pedro is in his Time Building phase, focused on strengthening his skills by investing time and resources in specialised training. With determination, he looks for opportunities to participate in flights and simulations that allow him to log hours, gain real experience, and consolidate his career path.

Skills

  • Solid aviation knowledge.
  • Follows procedures and safety norms.
  • Communicates well with passengers and instructors.

Objectives

  • Obtain commercial pilot licence.
  • Log hours cost‑efficiently.
  • Find job as professional pilot.
  • Finance training and licences.
  • Connect with other pilots.

Frustrations

  • Very high training and rental costs.
  • Few affordable flight opportunities.
  • Uncertainty about his career future.

Customer Journey Map

Five stages, from awareness to loyalty, capture actions, thoughts and emotions for each role.

Pilot in training

PEDRO GUTIERREZ

Scenario

Pedro needs 50 hours of time building to progress, but training is expensive. He uses FlySplit to share scenic or short flights so passengers help cover costs while he builds hours.

Awareness

Action

Researches ways to reduce the high costs of his time building.

Thought

Is there a way to reduce my expenses without sacrificing flight hours? I need something reliable.

Emotion

Worry

Evaluation

Action

Explores the app to verify legality (AESA regulations) and how payments are handled.

Thought

How can I make sure payments are transparent for the tax agency?

Emotion

Anticipation

Consideration

Action

Publishes his first flight and answers passenger questions via the integrated chat.

Thought

How can I make my listing more appealing? Will passengers respect weight limits and rules?

Emotion

Optimism

Use

Action

Accepts the request, completes the flight and receives the payment into his virtual wallet.

Thought

Will the platform process payments smoothly?

Emotion

Joy

Loyalty

Action

Rates passengers and tries to build a network of recurring clients.

Thought

Will my good reviews attract more passengers in the future?

Emotion

Serenity

Key Requirements

Based on the research and user journeys, 31 functional, non-functional, and legal requirements were defined. The design of the Minimum Viable Product (MVP) was prioritized around four strategic pillars:

Trust & Security

Mandatory validation of pilots' flight licenses and medical certificates, alongside ID verification for passengers. Implementation of a bidirectional rating system.

Transparent Finances

Automatic calculation and equitable breakdown of direct operational costs (fuel, airport fees, rental). Secure integrated payment gateway through the platform.

Legal Compliance

Strict compliance with AESA/EASA regulations to guarantee operations without profit motive. Clear display of mandatory aircraft insurance.

AI & Assistance

Integrated chat for direct communication and AI-powered features for weather-based cancellation predictions and route suggestions.

Ideation

Information Architecture

Based on research findings, the IA was structured using content trees for each role, then checked with a tree test before any wireframe was produced. Most key tasks were located in under 3 clicks — and the test also exposed two weak points in the navigation (detailed below), far cheaper to fix at this stage than after wireframing.

  • Content exclusive to registered or logged-in users.
  • Content exclusive to users who have verified their identity.
  • Content exclusive to users who have validated their licenses.

Tree Test Validation & Insights

To validate the initial Information Architecture, two Tree Tests were conducted using Optimal Workshop. While core tasks achieved up to a 100% success rate, the test revealed critical navigation issues that shaped the final User Flow.

Tree testing tool

Optimal Workshop

General Success Rate

Up to 100%

Pilot Insights

The Issue

Pilots struggled significantly to find the "Cancel Flight" option and the "Wallet / Balance" section (only 33% success rate).

Action taken

The cancellation action was heavily highlighted within the "Upcoming Flights" tab. The Wallet was moved to a more accessible top-level area in the Profile to ensure immediate visibility.

Passenger Insights

The Issue

Passengers experienced confusion when trying to set up flight alert notifications and understanding the separation between booking and paying (50% success rate).

Action taken

The notification settings were grouped into a dedicated, clearer tab. The checkout flow was completely redesigned to clarify the "Book now, pay upon confirmation" model.

User Flows

Detailed flow maps were created for each role's primary path.

  • Start of flow
  • Decisions
  • Actions
  • End of flow

Design

Design Evolution

An iterative design process that transformed early paper sketches into a high-fidelity interactive prototype. All applied through a Mobile First approach and validated at every step with real users.

01. Sketches

Due to time constraints and the creation of a completely new product, the design started with a strict "Mobile First" approach. Paper sketches were created to quickly visualize the "Happy Path" for both the pilot and passenger flows.

Tap to open the viewer

02. Wireframes

Figma wireframes were tested using the "Think Aloud" technique. This revealed a critical friction: linking payments to post-flight reviews caused distrust. To solve it, I designed a real-time in-person confirmation system using a PIN code.

Tap to open the viewer

03. UI Kit & Branding

The identity reflects its purpose: Blue (#2A5C9D) symbolizes the sky and the Pilot, while Green (#4CAF50) represents the ground and the Passenger. The interface followed Material Design guidelines, applying Fitts's and Hick's laws for an accessible experience.

Tap to open the viewer

Hi-Fi Prototype: The Final Experience

The result of research and iteration. A native mobile application divided into two tailored experiences, designed so that sharing a private flight feels as clear and legitimate as booking a commercial one.

Pilot

Smart Flight Publishing

Pilots easily configure the route, aircraft, and direct costs. The application automatically calculates the fair price per passenger, strictly complying with AESA/EASA non-profit regulations.

Both Users

Secure Booking & Payment

The flow begins with the passenger booking a seat and submitting their details. Once the pilot reviews the profiles and accepts the booking, the passenger receives confirmation and proceeds to make a secure payment through the integrated platform gateway.

Both Users

Live Tracking & PIN Verification

On the day of the flight, the pilot must enter a unique 4-digit PIN provided by the passenger. This secures the physical encounter and starts the real-time flight tracking on the map.

Both Users

Virtual Wallet & Mutual Evaluation

After landing, both parties evaluate the experience. Within a maximum of 48 hours, the pilot receives the booking money in their virtual wallet, ready to be withdrawn or used for future flights.

Test

Heuristic Evaluation

Before testing with real users, an in-person roundtable was conducted with 4 Tech and UX/UI professionals. The prototype was evaluated against Nielsen's 10 usability heuristics, identifying and fixing critical frictions to improve the flow.

Nielsen PrincipleIssueSolutionChange Image

User Control and Freedom

Issue:

Users could not delete mistakenly uploaded images or tourist points during flight creation.

Solution:

Integrated clear "X" delete icons to give users complete control over their actions.

Error Prevention & Efficiency

Issue:

If the pilot accidentally closed the "Start Flight" modal, the required PIN to track the journey was lost.

Solution:

Implemented a minimized permanent overlay at the bottom of the screen to prevent critical data loss.

Consistency and Standards

Issue:

The top navigation arrow and the bottom "Back" button were redundant.

Solution:

Removed the redundancy and standardized a "Cancel" button that triggers a data-loss warning to prevent accidental exits.

Visibility of System Status

Issue:

Tourist points listed in the flight preview lacked clear correspondence with the map markers.

Solution:

Added numerical indicators to the list matching the exact pins on the map for quick identification.

Usability Testing

Four real users (2 pilots and 2 passengers) completed end-to-end task scenarios using the "Think Aloud" technique via Google Meet. The goal was to validate the "Happy Path" flow and detect any remaining frictions.

100%

Task success rate · all 4 participants, every scenario

97.5/100

SUS score, Pilots · "Excellent" range · n=2

90/100

SUS score, Passengers · "Good-Excellent" range · n=2

Design Iterations: Listening to the users

Cost Terminology

Feedback

Pilots found the breakdown of "landing", "parking", and "ramp" fees confusing and non-standard.

Applied Solution

Terminology was consolidated into standard industry terms: "Airport Fees" and "Handling Costs", simplifying the publication process.

Tap to open the viewer

Security & Flight Plans

Feedback

Pilots noted they needed the ID (DNI/Passport) of all passengers prior to the flight to legally complete the flight plan.

Applied Solution

The booking flow was updated to make the companion's ID field mandatory before a request can be submitted.

Tap to open the viewer

Pilot-Passenger Communication

Feedback

Pilots mentioned they usually send the same repetitive instructions (meeting points, rules) to all passengers.

Applied Solution

A system of quick message templates (including AI-generated options) was added to the flight confirmation screen.

Tap to open the viewer

Scheduling Flexibility

Feedback

Pilots stated they often book planes for specific dates, making the "recurring weekly schedule" feature insufficient.

Applied Solution

A "Specific Date" mode with an integrated calendar was added to the flight publication flow.

Tap to open the viewer

Interactive Prototype

Two profiles, one seamless experience. Explore the final high-fidelity design and use the toggle to live the "Happy Path" from both perspectives.

Pilot

Conclusions

Key takeaways

FlySplit validated a legally sound collaborative model under EU Regulation 965/2012: cost-sharing is legal as long as direct costs are split among a maximum of six occupants with no profit motive. That legal clarity wasn't a footnote, it became a core design constraint.

Beyond compliance, the economics are the point: time-building is one of the biggest barriers to entering the profession, and splitting direct costs would cut it by an estimated 30–40%. That figure is an estimate built from the cost structure surfaced in research, not a measured result — production data would be the next step.

Usability testing closed the loop: all four participants (2 pilots, 2 passengers) completed every scenario, with SUS scores of 97.5 for pilots and 90 for passengers. With a sample that small I read those scores as direction rather than proof; the stronger signal was watching both roles complete the full flow unaided.

What I learned

The most important insight wasn't visual, it was structural. Pilots and passengers share the same platform but need opposite things: pilots need legal and financial transparency to protect themselves from tax risk; passengers need trust signals to feel safe boarding with a stranger. That dichotomy shaped every design decision, from the onboarding flow to the booking confirmation screen.

I also learned that designing for a regulated niche means absorbing its language. Iterating from generic terms to "airport fees" and "handling costs" wasn't a copywriting choice, it was a legal and trust issue. The mandatory companion ID field didn't come from a design principle; it came from a real pilot explaining what happens if a flight plan is incomplete.

Finally, the 4-digit PIN to verify in-person encounters before tracking begins remains one of my favorite solutions in this project, a small interaction that solves a real-world trust problem with almost no friction.

Next Steps

Unhappy Paths & Edge Cases

Mapping and designing error flows, cancellation management, and refund protocols to guarantee user trust during unforeseen events.

Cross-Platform Scaling

Evolving the current Mobile First approach to Tablet and Desktop formats, offering an immersive experience for detailed route planning on larger screens.

AI Integration

Implementing highly demanded features from our surveys: optimized route suggestions, virtual assistants for FAQs, and real-time weather prediction.