Improving the application experience on Useful School’s website

Useful School is the world’s first pay-what-you-can online design school for people of color. Our initial design helped us launch a great first cohort, but we knew we had some things to improve before driving traffic to our second cohort.

Here we break down how we refined the web experience to satisfy both business and user needs.

Useful School is the world’s first pay-what-you-can online design school for people of color. The platform launched in December 2021, with our first cohorts running from February to April 2022. We have a large number of applicants on a rolling basis, and a second cohort starting in Summer 2022. Before driving even more traffic to our new program, we wanted to optimize the web experience.

Students in our first cohort had a lot of questions about the class experience, so we realized we could improve the application process by adding more context. Not only would this reduce the time spent answering questions via email and live calls, it would also provide a great first impression and UX to potential new students.

Our discovery process relied on qualitative and quantitative feedback.

Our quantitative feedback came from two of our primary data sources: Google Analytics and Tally, our current application tool. Our main metrics were strong: bounce rate, click-through rate (CTR) from our class pages to the application, and total volume of applications. But we knew we could do even better by giving users more upfront value and a smoother experience.

The majority of our website traffic went to the homepage, class pages, and the student application. Online schools rely on application numbers, so we knew that even a 5% increase in clicks from the class page to the application would lead to a significant increase in growth. We also knew that improving the application experience would reduce our time responding to inquiries.

Qualitative feedback also played a key role in our exploration. I teach, lead curriculum creation, and respond to all emails, so I have a close relationship with our students and prospects. This opened a feedback floodgate from current students, future students, sponsors, and other working designers via emails, video calls, and Slack messages.

People wanted to know more about the class to help them decide if they should apply now or later. They were especially interested in who taught the classes, the required commitment, basic logistics, and if future classes would be offered.

We also conducted a benchmarking analysis of top-converting online programs and other-industry sites, including Netflix and Spotify, to understand best practices and areas of opportunity.

Important footnote on providing a unique experience: we heard from users that it’s important to be inspired by sites that perform well, but it’s equally important that we continue to provide a unique brand and student experience.

Based on our preliminary analysis, we felt there might be an opportunity to focus on improving a key micro-conversion to address both user and business needs—specifically, clicks from the class page to the application.

We gathered each piece of feedback into a FigJam file and started ranking the potential return on investment to design and implement the different suggestions.

We typically use binary questions to understand which changes to make, and the ones with the strongest “yes” get pushed to production. These are questions like: Is this information critical? Is it experienced on high-traffic pages? Is it noticeable in the first 5 seconds of a page visit? Will this take no more than 2 hours for an engineer to implement?

After we ranked and organized all of the feedback, we grouped it into similar themes. A few clear takeaways emerged:

Insight #1: Users needed a clearer CTA on the page. The Apply button in the navigation bar was okay, but people also expected to see a similar button or link on the page itself. And people were used to submitting their email address early in application experiences, so that wasn’t a barrier.

Insight #2: Users wanted to know if they were a good fit for the class before they applied. They also wanted to know the time commitment, required materials and tools (e.g. laptop, software, etc.), application requirements, and so on.

Insight #3: Users wanted to know the results of the program so far. They wanted to see past students’ work, student testimonials, and other success stories. Luckily, we already have lots of this, even though our first two classes haven’t even finished.

Insight #4: Users wanted the overall Useful School vibe. We needed to remain unpretentious and unique in our promise, while balancing fun, wit, and seriousness.

Next, we started brainstorming solutions with our freelance designer/developer and sketching wireframes. We used this key overarching question to help us focus:

How might we add more content and improve the user experience of the class page in a way that feels uniquely Useful School and creates a clearer, self-serve experience?

We started by mapping out a rough information architecture, site map, content and experience ideas, and references from other brands in Figma.

We then created a variety of medium-fidelity wireframes, from more conservative to really “out there” designs. Finally, we reviewed all the ideas and determined which versions best addressed the brief. (spoiler alert: many of the “out there” designs made it to the final. This taught us that it’s important to create a solid brief that also gives designers space to explore more creative designs.)

Here’s what else we learned:

The goal of our Maze test was to validate that the solutions we created addressed the primary feedback. Our designs weren’t final, but we included enough fidelity so that users could grasp the vision and let us know that we were on the right track.

Given our limited resources and the need to announce new classes soon, speed was key.

The test involved showing people a medium-fidelity prototype, and asking for feedback on the experience. Here’s what users had to do in Our Maze test:

Within three days of sending out the maze test, we had 17 valuable responses.

Users preferred or strongly preferred the new design over the currently live website (average score: 7.8 out of 9). A review of the quantitative and qualitative feedback told us that students and prospective students would like to see:

We reviewed the results and discussed how to tackle the changes using what I refer to as the “HiLo method”: putting all the ideas on a Highest Impact vs. Lowest Effort matrix.

Based on user feedback, we saw that we could improve the prototype in the following ways:

The lowest effort, highest impact changes involved the hero section. This made sense, as it’s the highest trafficked section of the page.

We’re currently working to determine which parts to add, remove, and update. We hope to create a final prototype and start web development within a month. We’ll also consider creating the final prototype and asking the same questions from the Maze test to ensure that we’re staying on track and truly assessing our students’ needs.