CarFlix

Contextual inquiry (step 1)

Our two-week contextual inquiry process resulted in an assortment of products, artifacts and documents that were shared among team members. These assets aided us in creating user groups, developing interview questions and generally gaining more knowledge about the physical drive-in space itself.

[1] First-person research (an actual visit to the context of use) — Preliminary research involved a member of the team attempting to find drive-in theaters near them to visit in person. The team member resided in Massachusetts and first used a "drive-in theaters near me" Google search to locate the nearest drive-in; however, the filtering options were sparse, and much of the legwork had to be done individually. Once there, the team member took detailed notes on the features and shortcomings of their experience at the venue. See the first set of images below for documentation of an in-person visit to the Rustic Tri-View Drive-in in North Smithfield, Rhode Island. The team additionally did a round of competitor research, where we found just a handful of existing mobile apps related to drive-ins. They all had antiquated interfaces, limited functionality, and were tailored to only one specific drive-in (rather than serving as a marketplace for all drive-ins, as we were hoping to do).

[2] Interviews with key user groups — The first step to collecting interview data was determining who we needed to interview. We referred to our original project plan where we identified our customers and stakeholders. Regardless of what features and solutions ended up in our product, we needed to form a better understanding of the overall subject matter, environment, and holistic needs of these various user groups. We decided it was important to include 4 main user groups in our contextual inquiry so that we were framing the bigger picture, and to more accurately capture the features that would drive the refined scope and target group for our prototype: [1] Owners; [2] Existing users; [3] Performers; and [4] New users.

1. Preparation: We prepared for interviews by creating a set of template questions for each user group. These templates were tailored for semi-structured interviews and consisted of roughly 10 questions each with different sub-questions to help guide the interviewer during the session to spark natural conversation. Each template explored the day-to-day activities, environment, needs, wants, and experiences that were unique to that group. We also compiled an introduction that could be used by the interviewer, as well as a general set of demographic questions that would be used for all interview types.
2. Conducting the interviews: Although all interviews during the contextual inquiry process would ideally occur within the context of use, the COVID-19 pandemic prevented us from meeting with most of the interview participants face-to-face. However, we were able to complete the interviews through phone calls, video conferences, and some in-person sessions. We ended up with interviews from 2 Owners, 7 Existing Users, 2 Performers, and 6 New Users for a collective total of 17 interviews.

Interpretation & affinity mapping (steps 2 & 3)

The group conducted an interpretation to go over the interview results, compile artifacts & products, and turn feedback into affinity notes to support the affinity diagram exercise.

The group then conducted an affinity diagram exercise with approximately 500 affinity notes via a collaborative Figma file (FigJam sadly did not exist yet) to enhance collaboration in a remote working environment. The team executed the process described in Rapid Contextual Design by Holtzblatt, Wendell, and Wood (2005) and the resulting diagram can be seen below.

Persona process (steps 4 & 5)

Next, it was time to translate our affinity mapping work into personas for our mobile app. Our persona modeling process closely followed the 8 persona development steps detailed in About Face by Cooper, Reimann, Cronin, and Noessel (2014). While I will only be giving a higher level summary here and will not go through all of the 8 steps in complete detail, you can view a more explicit look at the whole process on our project website.

We began by grouping interview subjects by their role (out of the 4 roles defined in the previous section). Each team member then brainstormed a list of distinct behavioral variables for each role. Using the variables, we derived a list of the significant behaviors that would be used to generate scales, which we then used to map the interview subjects against each variable. Next, we created a duplicate board of scales in which we color-coded the users by job role and discussed the board as a group, making adjustments where necessary to more accurately represent relative positions.

We then began the collaborative clustering process. We used circle shapes to visually represent temporary groupings on each scale of people based on clustering patterns we saw. All of our defined clusters had at least 6 variable scales that they aligned on, with the majority of them exceeding 8. As we started to define clusters, we checked that there was some sort of connection between the users, thinking back to the original interviews we conducted. Once confirmed, we removed the circle groupings and used a new set of colors to help represent the clustered groups on the scale. We ended up with 6 clustering patterns that were used for further persona synthesis.

Finally, we began to shape personas from our research. We defined activities, motivations, pain points, demographics, skills, attitudes, emotions, and their interactions with people and product. We also grabbed direct quotes to help us start framing their narrative and came up with a one-liner catch phrase that captured their essence. We checked for completeness and redundancy and then began designating persona types (primary, secondary, supplementary, and anti). As a group, we identified 5 characteristics that developed a distinct sense of identity for each persona (Tech Savvy, Social Media, Maintenance Level, Online Purchasing, and Familiarity with Drive-ins). These characteristics were also identified as ones that would be important in helping direct our prototype design direction. I’ve included three out of the six total personas we formulated below. It was clear to all of us that the primary persona we would design for was Lindsey; Many of her characteristics and goals were directly related to all of the other personas. By designing for Lindsey, the other personas would at least be acceptably satisfied. The secondary personas had a couple of key characteristics that would not be captured by Lindsey's needs. The remaining personas were deemed supplemental because they were captured by the primary and secondary personas.