As part of my ongoing Activity Infosite Information Architecture Redesign, I put together a three part mixed methods research plan to inform my design, as well as evaluative follow up research.

I'm a huge fan of mixed methods research. Every method has pros and cons, and in order to get a clear picture of what's actually happening, you need to choose the best methods for what you want to learn. For this project, I used ethnography, semi-structured interviews, a competitive audit, and a Kano study.

I began by conducting a mini ethnographic study, by recruiting participants who were going on a trip in the next three months. I brought these people into our usability lab, and asked them to find an activity to do on their upcoming trip. I let them use any resource that they would normally use, not limiting them to Expedia. As they explored, I asked them to speak out loud what they were seeing and thinking. I took notes on where they went, what information they looked for, and also any note taking or list making behavior. I also made observations to help me create a traveler journey map for activity shopping.

The second part of the research plan was a competitive audit, where I looked at what information our competitors and other similar e-commerce experiences showed to users at different steps in the shopping process. While it is informative to see what others in a similar space are doing, there are major limitations in what you can learn from this type of research. Mainly, you don't know why competitors are doing what they are doing; you only know what they are doing. Still, competitive audits can help you learn the current state of the field. This helped inform the third part of the research plan.

The third part of the research plan was a Kano Study, which helps to categorize customer preference for features or information. For this study, I took the information that users found important in the ethnography, as well as the information found on competitors' product in the competitive audit, and asked a pair of questions about each one. The questions were:

We received 800 responses to this Kano survey, which gave us some signifant data to back up our qualitative findings. Utilizing the survey results, we were able to categorize the different pieces of information into "delighters", "satisfiers", "must-haves", and "indifferent". Delighters are things that users like having but are not expected. Satisfiers are things that users like having and dislike not having. Must-haves are things that users expect and dislike not having.

From these three research studies, I was able to construct a potential information hierarchy for the activiy infosite page.

At a later point in the Activity Infosite Information Architecture Redesign project, I conducted evaluative research on an early Invision prototype that I created based on the information hierarchy that I had constructed. This evaluative research consisted of an intercept study at Pike Place Market. Pike Place Market turned out to be the perfect location for an intercept study about a travel product, as most people I encountered were tourists currently on a trip to Seattle. The purpose of this study was to see if there was any important information that was missing, as well as to determine whether or not users understood the pages and flow. It consisted of a simple scenario to find an activity to do and several followup questions.

This mixed methods research plan was instrumental in informing the information hierarchy and design of Expedia's new activity information pages (infosites). The findings helped generate a new traveler journey map for activities, informed us about what information is most important to travelers at different stages in the shopping process, and helped gain traction with multiple stakeholders within Expedia about the importance of research in the design process as well as the importance of redesigning our activity shopping path based on how users actually look for activities.

One major impact of this research was convincing our partners in the product organization that the new designs were the right way to go based on quality research about how customers actually look for and shop for activities. Because of this research, they were willing to back (and were enthusiastic about) A/B testing a radical new design, rather than very small incremental tests that are our normal process. They were also still supportive when the first design was slightly conversion negative, because they believed that the research and findings were solid, and that we should keep iterating on the design to get it right.

I gained more experience designing and conducting a mixed methods study, as well as conducting each type of reseach. This was also the first time I used a Kano study, so I added that skill to my repertoire through this project. I got a lot of practice explaining mixed methods research and how they all feed together to provide a more complete picture of users' behavior to others within Expedia.