Alaska Airlines: new feature prioritizations and mobile app adoption

How I contributed to the team, specifically:

Each team member took a part in every stage of research to not miss any opportunity of learning during the whole process. Knowing that my strong suit is in the qualitative research methods, however, I was more active in offering my perspectives and stance when designing the user interview protocols, compared to the other methods. In addition, I also played the role of a cheerleader on the team and kept up the positive vibes, as I always do in any team setting :)

‍

Context & Goals

While having the basic functionalities of a mobile app, the Alaska Airline mobile app did not have a strong position among its competitors and the user base had been dwindling. Our team was tasked with helping the stakeholders understand the guest experience within the app holistically.

Through a few stakeholder interviews, we understood that the fundamental business need is to increase app adoption. Although there is much room to better the user experience from a design perspective, the stakeholders wanted our team to focus on the prioritization of additional new features to be implemented, as the current app only offered limited core features.

As a first step, we turned business needs into research questions to guide our studies. The goals were to eventually provide insights that would help meet the business goals by answering the research questions that stemmed from them:

‍

Process & Rationale

Research question 1 is a "why" question, which would require rich and deep qualitative data from the right users. Breaking down the question further, it's clear that we needed to diagnose the problems behind decreased usage of the mobile app evidenced by existing data, and we needed to generate some hypotheses that would in turn have implications around the sample of users we'd recruit for the study.

Research question 2, on the other hand, is a more clear-cut, straightforward "what" question. We would need to provide a list of features, categorized and prioritized based on importance and urgency, which will be defined by a specific metric. This would require a quantitative method.

All things considered, we crafted a 3-step research plan:

‍

Methods & Workflows

Due to NDA, I am unable to share the details and the meaty results from each method. However, you can find how the three methods work interwoven together.

‍

Impact

We presented our findings to a team of stakeholders, including the VP of product, the product manager, the UX designer, the Visual Designers, and other UX researchers on the team for the mobile app. Our results were well-received for two reasons:

1) the 15 new features categorized and prioritized with individual scores provided specific, actionable recommendations that can be immediately taken into account by the development team; and 2) we also shared more long-term, future-oriented product recommendations based on the qualitative user insights, to which the VP of product responded:‍

"That really confirmed my hypothesis, so this validation is really important and great for our long term strategy".

‍

Learnings & Reflections

The art of mixing qual- and quant methodologies

I applied the Kano Model as a quantitative research method for the first time! Through understanding the logic behind the model, experiencing the detailed executions that go into crafting a rigorous Kano survey, and getting results from careful calculations, I truly learned the art of using quantitative data to complement the qualitative. The individual scores received by each feature granted much more depth to the qualitative insights we already gathered from user interviews, and vice versa. In addition, the storytelling as a whole was also greatly enhanced by having both types of data supplementing each other during the final presentation to the stakeholders.

The tradeoffs between focus and broadness

One thing I would have done differently if I were to do this project again, is to have a more narrowed focus on the user group. Although we uncovered a range of different insights by including a diverse set of participants (by travel frequency and app usage), I believe we could have achieved more depth in our qualitative data had we focused on one type of user. Of course, no methodology is perfect and our approach worked for this project for its emphasis on feature prioritizations, but I would keep in mind the inevitable tradeoff when I approach future studies.