The work that I did for General Motors is confidential. I am unable to share the specifics of the work that I did, but I can share the process that I used.
Images shared in this case study will be blurry to maintain confidentiality.
Thank you for your cooperation!
Timeline:
May 2023 - August 2023
13 weeks
Role:
I was one User Experience Researcher working alongside an Experience Strategist Intern
Why research as a designer?
I love product design, and I am a designer at heart, but when I got the opportunity to dive into research, I couldn't pass it up. Research is the foundation of design; without good research, you don't know your users well enough to design for them. I wanted to complete a UX Research internship to learn how to collect good data to inform design decisions. Conducting and analyzing User Research is an important skill for product designers to know, and that is something I wanted to practice through this internship.
Getting the problem statement and devising a plan of attack
The problem statements I have worked with in the past had a clear solution attached to them, and my job was always to figure out what features to include. The problem statement I received this summer was open-ended and exploratory. We had no idea what was causing this problem and we needed to research why this was happening and what we can do to fix it.
Defining knowns vs unknowns
The first thing we did was conduct secondary research. In order to figure out what to include in our moderator guides, we sorted all of the secondary research we collected into known and unknowns. The knowns were facts with good data already defined in previous research. We collected data from internal databases, Google, and Bain. After looking at our knowns, we identified gaps in the data and areas where we wanted to learn more about and added those questions to our unknowns.
Setting research objectives & Crafting a moderator guide
Our unknowns created a starting point for us to define our research objectives and questions in our moderator guide. We listed out a few of the unknowns that we wanted to target and listed those as research objectives. From there, we took each research objective and came up with a list of potential questions to ask users to gather more data about that topic. After lots of deleting and editing the questions, we compiled the best ones into a moderator guide.
Recruitment & Screener Questions
Based on our questions, we decided on our recruitment criteria and then wrote up some screener questions to use to recruit participants. In total, we recruited 11 participants (6 female, 5 male). We used UserTesting to recruit and conduct interviews. I learned how to create good screener questions
to prevent participants from lying. I learned to stay away from "yes or no" questions and to be as specific as possible.
Empathy Interviews: Lots of lessons learned!
After recruiting participants came the hard part: talking to them! Research was a lot harder than I thought, it was really scary to go out and talk to people you don't know. I came across many different kinds of participants-- some more reserved, and some more friendly. The biggest lesson I learned was to first build rapport with participants before diving straight into the questions. Building trust with participants made them feel more open to giving information when asked questions.
A screenshot of a virtual interview I conducted!
A big part of our questions focused on hypotheticals-- something very hard for participants to give answers about. To help them think through these ideas, part of our interview focused on walking participants through an activity where they would create their own word organizations on miro. By having them group words and talk about their reasoning, we were able to get a better look at their motivations.
While conducting interviews, we were also taking notes on a miro board so that we could create an affinity map once the interviews wrapped up.
Analysis leads to Insights
Once we finished up interviews, we did affinity mapping for synthesis where we took those sticky notes and grouped them by like themes to uncover opportunities or recommendations from the data. We created 5 key insights in the form of "How Might We" statements.
Ideation Time
Now that we understood the problem and what was causing it, it was time to think of possible solutions. To help us think of ideas, we created User Personas based on a 2 way split we noticed in all of the participants we interviewed.
We organized the personas with the How Might We statements and hosted an ideation workshop to have people come up with possible solutions from the point of view of one of the personas. We then organized all of those ideas and picked out the best ones to use as solutions.
Time to design!
I took the possible solutions and organized them into ideas and features. Ideas were more general overarching themes, and features were specific items that could fit into a theme. My team knew that I had a design background, and they let me take these ideas and create wireframes for a website that could address the problem.
Apologies for the blurriness, all content must remain confidential.
Going back to the users
After creating the wireframes, I wanted to figure out if they aligned with users' expectations, so we conducted some concept testing. We showed participants the wireframes and had them talk through ideas they liked and didn't like and things they would add or remove. These interviews helped us refine the concepts to make sure the problem was being adequately addressed.
We interviewed 7 participants for this round of testing and collected data in a miro board.
Rose, Bud, and Thorn
Since the data we collected was focused on likes, dislikes, and things that were missing, we thought Rose, Bud, and Thorn would be a good way to analyze feedback on the concepts. Roses are any positive feedback, Thorns are negative feedback, and Buds are areas of opportunity. We sorted the sticky notes into a large table that had rows for each feature we shared with participants.
2nd and 3rd iterations
After sorting through the data and taking note of common areas for improvement amongst all participants, I made changes to the wireframes and shared them with fellow designers to get more feedback. This led to the second and third iterations of the wireframes.
Next Steps
This is where my time with General Motors ended! If I had more time, I would take those wireframes and start designing them in High Fidelity on Figma.
Reflections
I truly enjoyed my time with General Motors this summer. They were the most incredible team to work with and I will forever be grateful for the support they gave me this summer. Here are some of my big takeaways from this project:
The first lesson I learned was about time management. It’s better to ask for more time to work on something and create a stronger result rather than try to cram it and have sloppy work.
The second lesson I learned is that usability is not the same thing as usefulness. In the past, I had only ever used wireframes for usability testing, so when I was going into the concept tests, I would ask more usability-focused questions instead of usefulness questions. I had to switch my mindset away from the design and more towards the idea of the concept itself.
The third takeaway I have is to always ask lots of questions. When you ask questions and get clarification often, it ensures you are on the right track, and you are making good progress. If we were steering down in the wrong direction, we asked questions and got clarification early enough that we could backtrack and rethink our next steps.
Again, I am so thankful to the team over at General Motors for making me feel like a valued member of their team.