Treehouse was a dream job for me. I hit the ground running and over my short tenure designed, coded, and shipped several features. I worked on A11y changes to the Design System as well as conversion and retention projects on the Techdegree team.
40% increase in student winback rate + 25% lift in office hours attendance.
1 Product Designer, 1 Engineer, 1 PM
Research, Customer data, PRD creation, Icon Design, Product Design, cross-functional communication, Presenting work to leadership.
During my time at Treehouse, I was lucky enough to work on the Techdegree (TD) squad. Our squad was a cross-functional team focused on acquisition and conversion. I worked with my PM and engineering partners to define and frame in-scope problems. My role within the team was as a co-facilitator and design lead. In this project we were able to increase the number of students attending weekly office hours, win back previous Techdegree students, and ship some really helpful quality of life features to our students!
The Techdegree was a relatively new product. Data for our programs were pointing to a need to optimize the existing experience. Through exit interviews that teachers carried out, we learned that students felt that support needs were not met. Teachers and success representatives felt like students weren't taking advantage of existing resources. For example, weekly office hours were being held for each of the offered disciplines but attendance hovered around 10-20%.
Building features that are useful across the user-base (from trial to paid) is a first-tier goal at Treehouse. With this project the challenge really centered around how to make something as banal as using Slack feel human and personal. In an educational situation, it's even more pressing. A student has to complete specific units and assignments in a timely manner or the program becomes less valuable. As a platform, we have to deliver that value even when students are not in optimal brain space.
Some other challenges we identified were:
These were all real concerns we kept in mind during these processes. It became a sort of tenet of this work to "Minimize anxiety," as best we could.
Everything, for me, starts with people. Our team found common qualitative and quantitative actions in what we referred to as Bright Spot Students. On their path to success, these students had engaged with our Slack groups early and often. Engaging in conversation, creating camaraderie, and asking questions related to coursework. We found that students who were most engaged with the community as a support system tended to be more successful.
Digging into some of the qualitative data that teachers had been collecting was helpful. Correlating it with quantitative data inside our products told the full story. We found that many of our students who were not successful were lacking a few of the key Bright Spot Student traits. These unsuccessful students were doing the work. Often the main differentiator was that they were not engaged in the community.
I started by assessing our student's in-app lifecycle. Bright spot student traits became a sort of north star for us. Our TD students lived in three main areas of the product. The Dashboard, the learning/assessment views, and Workspaces. Workspaces held a lot of legacy code and were not in the scope of the current project. Each of these spaces has a specific set of actions that it prioritizes.
The Dashboard provides guidance on the student's progress, success, and next steps. The student lifecycle allowed us to optimize for certain actions inside a goven space and gave us more information to go on when prioritizing tasks for this project. The missing piece was finding the help the student needed when she needed it. We started this project by working through a lot of rough ideas. These earlier ideas focused on things like early Slack adoption and even course retargeting.
It was at this point that we circled back to the idea around the office hours for students. Our Bright Spot Students were often active participants in this program. Taking it upon themselves to field the emailed list of weekly dates and times and attending the Slack/Zoom meetings.
This was interesting for 2 reasons:
Identifying the challenges of the students and our success reps early on were integral to this project. They were reporting 2 sides of the same problem! We started sorting out how we might better serve both students and Success reps.
When we started working on this project a few things became clear about the dashboard:
We removed the mentoring section and cleaned up the Slack CTA a bit as a first iteration. That was a big win - removing experience debt and giving students a clear call to action was helpful. We knew there was more we could do. Our next iteration was focused on bringing those office hours into the app for visibility and clarity. We worked with our Success team to get office hours schedules ahead of time and in addition to their regular email we hardcoded the office hours schedule into the app. This was a trade-off we made to lessen the impact on success reps while we tested this feature over the next several weeks.
We started testing Mockups and prototypes straight away. Teachers were using the above version of the dashboard in upcoming exit interviews. Through those exit interviews, we found that 4 of 5 students were either unaware or routinely forgot about office hours. That felt like a big win. There was still work to do to improve this feature before we shipped it though.
Updating the layout of the office hours schedule to feel more like an event was the first order of business. A common layout with the necessary information so it stands out as easy to read and understand. The next piece was creating a more helpful call to action for Slack. We felt like outlining the real-world benefits of using slack for the TD program was crucial.
We again put this new UI in front of students and stakeholders and the feedback was positive. Building a simple backend for the office hours UI was going to streamline the processes of the Student Success team. Students would be able to see when office hours, and eventually, other events were taking place. There was a lot to be excited about here. One piece of feedback kept nagging at me though.
- A Former student in an exit interview
It was impersonal. Beyond initial barriers, there was still a lack of familiarity and human interaction. It's hard to feel a human connection in a product. One of the ways that we can do that though, is with faces. Avatars create a mixed-reality world where you can connect with someone with a sense of familiarity. We wanted to find ways to use this to our advantage in the dashboard. This went through several iterations.
In a Techdegree program, our Student success reps are the first points of contact. Those reps were most likely to be holding office hours and fielding technical questions in Slack. Since we already had these points of contact, it was a no-brainer to use those people as a student-facing lifeline in the dashboard. Most TD's had one lead success rep. We added that person to the dashboard in a new section called, "Your Techdegree Support Center."
After doing the front-end work for this project I had the opportunity to again reach out to some students for feedback. In those conversations, there were 2 of 5 students that said they were likely to return to a TD program. According to a source, after I had left Treehouse, the office hours attendance had gone up ~25% since launch. Before joining Treehouse I was a student and a big fan of the organization in general. I'm very proud of the work I got to do in the short time I was there.