high fidelity prototype.png

Changeit: Re-imagining the Alaska Airlines flight change experience

Human Centered Design & Engineering Undergraduate Capstone | Jan. - June 2019


Team Members: Carina Dempsey, Timnit Bekele, and Melody Xu

Duration: 6 months (January 2019 - June 2019)

Skills: Competitive analysis, User research (surveys, interviews, usability tests), paper prototyping, high fidelity prototyping (Sketch)

Role: Content strategist and research program manager


How can we empower Alaska Airline guests to feel confident in changing their own flights? To address this question, I worked in a team of four undergraduates in the UW Department of Human Centered Design & Engineering to reimagine the change experience for Alaska Airlines, a major American airline based out ofSeatac, Washington.

THe Problem: Reducing Call Center volume and empowering guests

Alaska Airlines airplane

Call center volume at Alaska Airlines has increased faster than Alaska Airlines business size. Our teamsaw this as an opportunity to empower Alaska Airlines guests to feel confident when using self-service tools, so we went through the research, ideation, and prototyping process to propose a new flight change experience that would increase guest trust.

To do this, we conducted foundational research, creating a paper prototype and tested it with guests, and created a high-fidelity prototype. My primary strengths were in establishing rapport with research participants, translating research insights into design recommendations, and telling a story about how we developed our high-fidelity mock-up with end users in mind.

Understanding the problem space: Why and when do guests change their flights?

Our guiding research question was:

How do guests prefer to change their flights, and why do they choose these particular methods?

We had the following research goals:

  • Gather quantities and qualitative data about the tools that guests use when changing their flight

  • Learn about guests’ rationale for choosing their preferred platform

  • Identify design guidelines for a reimagined change experience

Developing trust with interview participants

To begin to understand what tools Alaska Airlines guests use when changing their flight and how these preference change with context, we designed a survey on Typeform. A survey enabled us to learn the preferences from the largest user base possible, and we asked for demographic data about age, travel purpose (business or leisure), and mileage plan to see if there were significant differences across these populations.

Age versus preference graph

Age versus preference graph

One interesting finding was that guests under the age of 35 have a higher preference for the mobile app and computer, whereas guests over the age of 25 prefer the Call Center over self-service tools.

Usability test findings

From our survey, we identified 8 interview respondents and conducted 45-minute interviews with them, which comprised of a few open-ended questions and a usability test of the current system. Because of my past experience as a journalist, I wrote a set of interview questions that would enable us to develop rapport with the interview participant, learn about their past experience with changing flights, and give them the opportunity to ask questions.

During each interview, we also conducted a usability test of the current flight change system, and we had the following findings:

Flight selection in current flight change experience

Guests want more feedback: Guests want more feedback throughout the process to ensure that they’re successfully changing their flight.

“Nothing's happening. I see it shows the fly out menu, but there’s no confirmation for the keep flight.”

-Interview Participant (P7)

Flight selection experience

Redundancy is distracting: Guests were confused by the repetition of the original reservation. However, it was helpful to have some reference to their original reservation.

“Seeing repeated information about my return flight is reassuring, but also annoying because it shows the same thing over and over again.”

- Interview Participant (P5)

New selected itinerary

New selected itinerary

The current system lacks intuitive language: Many guests were confused by the meaning of terms like “credit” and “My wallet,” and often needed to Google these terms or call to learn more

“I’m unsure how the credit works. I think that's the thing that confused me the most. I wasn’t sure like how much credit I had to spend. What happens if my flight costs more than my credits?”

-Interview Participant (P6)

credit confusion.png

Change fees were an important factor when changing their flight: When looking at new reservations, guests inquired about if they would be charged a change fee. However, the current system made it unclear whether they would be charged or not.

“[There is] a separate $125 per person change fee … I don't know whether I will be actually charged that fee.”

-Interview Participant (P4)

A visit to the Call Center

In addition to learning from Alaska Airlines guests, we wanted to learn more about how and why people call. Timnit and Carina both visited the Call Center and listened in on a few calls, and we discovered a few main takeaways.

Based on our call center visit and interviews, we found that Alaska Airlines’ guests call the Call Center when:

  • A situation feels out of control: For example, guests want to solicit help from the Call Center if they are changing multiple parts of their reservation.

  • They believe they can get a better deal: In reality, Call Center employees see the same information about flight availability as guests.

  • They want to tell their story (i.e. family emergency): One guest in particular told a story about how they spoke with the Call Center because their family member was sick, so they needed to make a last-minute change to their flight so she could stay with them longer.

Key Findings

High-fidelity prototype mock-up

Based on our research findings, we developed a design hypothesis that informed our high fidelity prototype:

We believe that by improving terminology, information architecture, and feedback, we will increase guest confidence and trust in the change experience.

Our team used Usertesting.com to test our high-fidelity prototype with 8 participants. For our user group, we decided to focus on Alaska Airlines users who may or may not have a mileage plan, but are responsible for paying a change fee. This demographic represented the largest part of our data.

Overall, participants thought our system was simple & easy to navigate.vParticipants described as “straightforward”(P7) ,“clear” (P5),  and “self-explanatory”(P8). All 8 participants ranked the task as a 1 (very easy to complete) on a scale of 1 to 5

Design guidelines

Based on our findings from surveys, interviews, and our usability test, I wrote the following design guidelines that guided the development of our high fidelity prototype I wrote design guidelines that could inform future iterations of the change process and guide the work of designers. Moreover, design principles are more durable than a mock-up alone because it can be a reference to the desires and needs of Alaska Airlines guests.

Break it down: The process of changing a flight should be broken down into manageable parts.

Flight selection page in proposed high-fidelity mock-up

Flight selection page in proposed high-fidelity mock-up

To demonstrate this value, we:

  • Separated the passenger and flight selection process into multiple pages. Each page had a discrete action, reducing information overload for the user.

  • Added breadcrumbs so guests can track their progress and know where they are in the change process. This was a feature that was used by other airlines like United, JetBlue, and American Airlines, which we discovered through our competitive analysis.

[It’s] easier than I expected, especially based on my experience with other airlines. Every step of the way, I had no doubt about what I needed to do next.

— Usertesting.com Participant (P2)

Be transparent: Guests should know how their old reservation relates to their new reservation. Change fees should be communicated up front. To demonstrate this value:

Cost breakdown in proposed high-fidelity mock-up

We created a clear cost breakdown that shows the cost differences between the  old and new reservation and change fees. We also provided an hover feature we users could see how their previous reservation applied to their new reservation, and we used language that was consistent with Alaska Airlines’ policies for changing a flight reservation.

It was really easy to change the flight and understand the cost breakdown.

— Usertesting.com Participant (P8)

Provide context: Not everyone knows the term “credit” or how they’ll receive a refund. We should provide this information in context. To address this:

Hover over feature for “My Wallet,” an Alaska-specific feature

  • The original reservation is available for reference at the top of the page when selecting a new flight.

  • We implemented hover features that enable guests can see information about terms like “credit” and “My wallet” in context. I also focused on using language that was consistent with Alaska Airline’s policies.

I love that it says Updated itinerary. Sometimes, just a few words can make all the difference in reassuring you that you’re making all the right changes.
— - Usertesting.com Participant (P5)

Lessons learnedand Next Steps

From this project, I learned that:

Alaska Airlines team photo

In a flight change experience, guests value cost transparency, guidance, and intuitive language: These values were reiterated in the call center visit, interviews, and our usability tests. Although our prototype will continue to be developed, our team believes that these values should be kept in mind when applied to future iterations of the flight change experience.

Taking a user-centered approach starts with listening: In our user research, I focused on asking open-ended questions to understand their expectations and solicit design recommendations. As a journalist, I always ask questions based on my interviwee’s responses, and I focus on listening without assumptions or judgments. This ensures that interviewees have the opportunity to express their needs and thought without the influence of my expectations, and it also establishes a sense of trust.

It’s essential to advocate for your design decisions with evidence: During our final presentation to Alaska Airlines leadership, we paired every design recommendation with a quote from a user. This ensured that people in the room had an idea of what users were saying about the current system. We also included quotes from usertesting.com to highlight strengths in our proposed design.

Opportunities for growth

If I could do anything differently, I would have spent more time on the data analysis of survey and interview findings. My team members and I each coded our interviews differently, so I had to spend a lot of extra time going through the interview transcripts to find quotes that justified my findings. Next time, I will focus on identifying a cohesive coding scheme so I can easily understand the data.

In the future, I will also take advantage of opportunities for feedback. For example, I could have spent more time talking to my professor about opportunities for improvement in our design. In my future work, I also hope to schedule feedback meetings with team members and supervisors to ensure that feedback is a continuous part of my design process.