A USER-CENTERED APPROACH TO TECHNOLOGY INNOVATION IN THE WINE INDUSTRY
For our capstone project we worked with a top technology consulting company to do some innovation research in the area of wine and technology. In order to turn such a nebulous ask, into a well define product idea, we started with a good deal of background research. Initial market research showed us that millennials are drinking up to 42% of the wine supply, more than any other generation. We decided to focus our research on millennials as they would be a great target audience of a technology product. Instead of trying to start designing off our pre-conceived biases we used qualitative research to understand why people drink wine. We focussed on what the Functional, Emotional, and Social needs users were using wine to fulfill.
A quick outline of our qualitative research process:
We performed 15 semi-structured interviews. These interviews focused digging at specific memories with wine and remembering what the occasion was, who the interviewee was with, and how the felt while consuming wine.
A team of three including myself did a round of 'open coding' on the interview transcripts, 5 each; during this process we outlined persistent themes from the interviews that gave insight into why people consumed wine, wanted to consume wine, and how the felt while consuming wine. We ended up with 50 themes, too many loosely defined themes to define actionable design guidelines.
We got our research team together to use Clayton Christenson's "Jobs to be done" framework to distill these into something actionable. The main 'job' that people were hiring wine was, 'to be confident in themselves.' We further broke this down into six 'sub-jobs' that people were hiring wine for. Those are:
Consumers hire wine to express class status
Consumers hire wine to relax and feel special
Young adults hire wine to feel mature
Consumers hire wine to bring people together
Consumers hire wine to have something to talk about
Enthusiasts hire wine knowledge to help them learn and grow
We synthesized these into t research deliverable for our client linked above.
Preparing to transition to Product Design
After we had established these 'jobs' consumers were hiring we went back and did a round of 'axial coding' on our interviews where we organized the insights from the interviews according how they matched up with the themes we'd decided on, going back to our data showed us that job #3 "I hire wine to feel special" was more accurately expressed as, "I hire wine to relax." This step helped us confirm our findings and hone them in.
To prepare for the product design phase, our research team broke down each job in terms of 'pushes and pulls' what's pushing the consumer towards using wine? ie: 'I can't drink beer, it makes me feel bloaty.' What's pulling the customer away from wine? ie: 'I prefer the taste of whiskey when I'm drinking.' We also took this opportunity to exam the habits consumers had, examine their anxieties associated with drinking wine, and their motivations for drinking wine.
At this point we had turned our research into a foundation on which we could create a product.
WINE SCOUT - PRODUCT DESIGN
For my Capstone Project at UCI, my team and I worked with Accenture Labs to innovate in the area of wine and technology. At this point, we had our research as a foundation for our product, for additional background on the project link to the research using the button below.
In-Person Ideation Session for Product Design for a Remote Team
At this point, we'd learned some pretty compelling reasons why people drink wine - so the next big question we asked was: what can we create that addresses these real motivations that people have around wine? I should pause here and mention that our team is remote. For most of our capstone, all our collaboration was remote. But for brainstorming and ideating, being a distributed team is kind of a problem. There's just no replacing the energy and buzz that you get when a group is talking ideas in a room together. Especially when cell phones are turned off. So, I organized a weekend-long ideation session in person. I called it...our design jam.
I modeled the design jam after the Google Design Sprint, as outlined in Google Ventures Sprint book. Our goal here was to come up with as many good ideas as we could, pick the best ones, and test them with people. Using the Push/Pull diagrams to guide us, we flared and focused to develop good ideas, and once more to sketch the strongest ideas. By the time the design jam was over, we had prototyped and tested our five favorite ideas with a few wine drinkers. Turns out that some of our ideas really resonated with people, so we took those with us. But others, we could eliminate right out the gate, with very little time wasted on them, which was also a win for us. For further information on the design Jam, see the Medium post below.
Honing the Product Vision
The in-person session took us pretty far toward defining our big "what to do make" question, but we hadn't done anything to refine our many good ideas into one, to really run with. So we started user testing. Lots and lots of testing. We even wondered if we had done too much testing. We tested prototypes (3). We did competitive testing to see how people engage some related wine websites (3). We created storyboards (14) and had people react to those... At each of these moments, we were looking to see what features resonated with people. And not just what features but what combination of features were most compelling to users. What experiences. We ended up with a long list of possible features - that ranged from "delivers wine to your door" to "picks the best wine for a group based on user preferences."
We eventually narrowed down our list to three things that Millennials want out of a digital wine experience: they want something that connects them to their friends, that helps them learn about wine without much effort, and that uses "algorithms" to personalize the content. No one called that machine learning, but that's what they wanted. To pull it all together, we tweaked those findings into a product vision statement that could guide what we did next: we envisioned “a platform to help you confidently discover new wines, and grow with you as you explore your palate.” After the huge open-ended ask at the beginning of this project, we were pretty excited to get this much clarity about what to make that might actually address why people drink wine. We had enough direction to start bringing the product to life.
Next up it was time to move to Interaction Design and create an actual prototype.
CASE STUDY: WINE SCOUT INTERACTION DESIGN
Once we understood user motivations about why they drink wine, and developed a strong product vision, the next step was to create a great product. We achieved this through three steps: 1) defining a goal, 2) purposeful user testing, 3) thoughtful product iterations.
For additional background on the product up to now, see the Product Design Phase.
Defining a goal
At this phase of our design we were still working on a name for our product, we knew from our product vision statement that we wanted to offer "a platform to help you confidently discover new wines, and grow with you as you explore your palate." We knew from our Research and Product phases while using our app we wanted the experience to convey specific feelings to our users, namely: 'the thrill of the hunt,' 'the wonder of the journey,' 'the joy of discovery,' and 'the excitement of exploring.'
We developed an Information Architecture (IA) that encourages users to begin their own journey with wine; to explore both wine and their own palette in order to discover new wines they like; and find out what flavors in wines they don’t like. After iterating on navigation methods, we decided on a persistent bottom nav bar that consists of four primary functions and a hamburger menu for additional features.
The 'feed' tab which offers users a place to input palette information, order a wine tasting flight to share with friends, see what others are drinking, and see wines suggested by experts, or the apps palette matching algorithm.
The 'search' tab allows users to search for wines that match their own, or their friends palette profiles. This features also allows the user to input how 'safe' or 'adventurous' they are feeling, this gives the user the option to discover more and expand their palette, or find something they know they'll like.
The 'scan' tab allows the user to use the camera to input a single wine bottle label, many labels at a time, or input a wine menu to find something that matches their palette profile.
The 'me' tab allows users to see their own palette profile visually. This keeps a running history of the wines the user has tried, as well as what the user did or didn't like about the wine. Every bottle is an adventure, and every adventure is recorded. You can also view your friends profiles from this tab, or through the hamburger menu.
Below is the final IA that we ended with and the initial wireframes that we used to get started. At this point we felt strong about the core of the product and were ready to begin testing again.
Interaction Design is where the rubber meets the road, we've defined our users and use cases, we're closely following our product vision. See below organized from case studies to tasks to in the weeds workflow decisions.
Once we had the Information Architechture (IA) sorted, we needed to verify it made sense to the user, so we brought the product back to the users. We were looking to verify two things:
Does the IA we're using lend itself to our stated goals?
Are the features we've chosen intuitive to find?
Are the features we've chosen useful to the user?
To achieve this, we gave the users very little context, we asked them two very general questions about their wine consumption, then began showing them the interactive prototype. Testers were getting hung up on the feed tab, they didn't see the difference between learn and discover, and they didn't care. Testers really liked the profile, what it contained and how it was organized, we did however, find a couple things to optimize about the way the information was presented. We took the information from testing back and iterated on our design.
Using the information from the user testing, we changed the feed page, testers were getting caught up on the feed and it wasn't meaningful to them. They didn't know or care what the difference between learn and discover was. We changed the 'feed' tab to 'home' and got rid of the discover and learn options. Through our last round of user testing we'd really honed in on what the testers felt was important to include in a search, we cut down the input as much as possible to ease the search process, and added the ability to view friends profiles directly from the profile tab. At this point we applied our visual design and went back for a final round of testing. After looking through the prototype devoid of any context, we asked the users what they thought the app would do out of three testers their responses were:
"An app to help you discover new wines and to keep track of your previous favorites and to rate the wines that you’ve tried."
"An app for helping you find out information about wine. Helps you taste the wine without opening the wine. That also has these social aspects of it. Yeah."
"Useful tool for exploring and rating wines, and figuring out what you like and don’t like, and having a really handy place to record that information."
We made some final tweaks based on the information from our final round of testing, the final prototype screens can be viewed via the link below.