UX Design by Rebecca LeVine
Wayfair: Customer Messaging Portal
In 2021, Wayfair introduced the Supplier Transfer Program: a means of connecting customers directly with third party manufacturers in lieu of always resolving customer service issues in-house.
These exchanges took place over email and weren’t a perfect customer experience:
-
Manufacturer messages sometimes went to spam.
-
Emails didn’t thread, making it hard to track conversations.
-
Customers weren’t clear when they were communicating with Wayfair and when they were communicating with the manufacturer.
-
Customers felt, overall, like the company they’d made their purchase from—Wayfair—didn’t care about resolving their issue.
DISCOVER
Before beginning iteration, I did a quick comparative analysis. How did other companies handle direct communication between customers and third party vendors? What did their UI look like; what tools did they provide customers in terms of escalation; how did they manage customer expectations?
I focused on Amazon, Etsy, and Airbnb, noting specifically if and how they supported conversations between three parties: themselves, the vendor, and the customer. Airbnb, for instance, kept a single thread between the customer and the “host,” but included guidance from Airbnb in the form of in-thread banners.
Airbnb’s message platform between the third party vendor (i.e. host) and the customer. “Messages” from Airbnb are shown as text outside of the chat bubbles.
ITERATION
Armed with this information, I assembled a list of features the app should have in order to address users’ most pressing needs. From there, I started a combination wireframe and lo-fi prototype, focusing on three processes: onboarding; the “intervention” (the point at which the app redirects the user from their social media browsing); and the mindfulness activity. I also started thinking about a dashboard/home screen, and what features it should include.
I started with these lo-fi screens, thinking about what kind of data the user would need to input, and what choices they should be offered.
Once I’d worked out some basic flows, I started experimenting with UI designs. I wanted the app to feel clean and modern, but still warm and approachable, so I settled on a lilac and navy blue color palate with pink accents, and a neumorphic design approach.
A few iterations of the first screen in the "activity" flow, requiring the user to tap the button that best describes their mood after their allotted time spent on social media. I removed some of the options to make the process less overwhelming; added a personalized greeting, and removed the app's name from the header, while keeping the amorphous shape but in a contrasting color.
It quickly became clear that the dashboard would be the most challenging piece of the project: I needed to decide which elements it should and shouldn’t include, then figure out how to present them in an easily comprehensible but still visually pleasing way. I also wanted the user to be able to start a mindfulness exercise directly from the dash, but it took several iterations to find the clearest way to do that.
The first draft of the dashboard featured a labeled "Start" button, but its placement didn't make much sense, and it didn't jump out enough. A revised version featured a "start" button that wasn't clear enough as a CTA, and was crowded with too many data visualizations. The third is more recognizable as a CTA, with a clear placement and an eye-drawing color.
TESTING + REVISION
After iterating a few times on the prototype, it was time to start testing. I conducted three interviews over Zoom, in which respondents received the prototype and shared their screens with me. They completed various tasks, answered questions, and provided feedback. With their permission, I recorded the interviews so I could review them in depth.
I was pleased to learn, from the tests, that all three respondents found the onboarding, intervention, and activity processes quite intuitive. I was concerned that they would chafe at the “intervention” moment—that is, when they are automatically removed from a social media app after the designated amount of time; in the testing prototype, this moment takes place automatically, without a push notification or alert—but while none of the users predicted this turn of events, they all said they liked it. “If there were a notification, I’d probably just ignore it,” one respondent said. “And that would sort of defeat the purpose.”
After the user's allotted amount of time on their selected social media apps is up, UnScroll automatically opens, without a notification.
A few concerns did come up across the interviews, all of which were helpful in my constructing a revisions roadmap. One issue all three respondents mentioned was where their data—the emotions they selected, the notes they recorded—were going. This was both a privacy question, and a technical one: would the app have access to, or sell their data? And shouldn’t they be able to access their notes?
To address these concerns, I added language to the onboarding screens explicating the privacy policy, then added a field to the dashboard that would let users view their notes.
Some other issues came up around the dashboard as well. Two respondents had questions about the graphs I’d initially included; they weren’t clear on units of measure, or what the graphs were really showing. I realized the data visualization I’d put in place looked nice, but wasn’t communicating enough actual information. I needed to rethink the graphs, make them clearer and more communicative.
The third respondent didn’t take issue with how to interpret the graphs, but she was troubled by their presence there in the first place. She pointed out that having reached the dashboard after completing a meditation exercise, the last thing she wanted to be presented with was her “stats”—she wanted to stay in the moment, not fret about how many times in the past week she’d been anxious, how many times she’d been envious, etc. Some users want to see this information, others don’t. It should be easy to find, but require an action to view it.
Users weren't clear on how to interpret the graphs on this version of the dashboard. One user didn't want to see this information at all.
Tapping any field expands it, showing more information. For instance, to see more about the feelings they've recorded, the user just taps the button listing their "main feeling."
To view the notes they've made, the user just taps the relevant button. This addresses the issue, flagged by several interviewees, of how the user can access the notes they've entered before or after mindfulness exercises.
The updated dashboard reflects the user feedback by offering a more succinct summary of its contents: instead of being confronted by graphs and charts right away, users can tap a particular field and expand it to see more detailed information. They can also tap to view the notes they’ve entered while using the app.
SUMMARY
Most of us spend too much time on social media, and it’s not just a question of it interfering with our productivity: it’s interfering with our mental health, too. I wanted to design an app that would not only limit the user’s time on particularly addictive apps and sites, but also redirect them toward a task that would improve their mood and sense of wellbeing. The usability tests I conducted were invaluable in improving the design of UnScroll, and toward the creation of a useful, usable productivity and mindfulness mobile app.
All Projects
Our team had no ability to affect the policies that likely informed some of this dissatisfaction.
What we could do, however, was improve the experience: we could add clarity, increase ease of use, and minimize friction. Thus, we began to explore a Customer Messaging Portal.