Designing for Student Engagement

Around 1 in 5 students fell off track during the first month after enrollment. Later drop-offs varied by course content and difficulty, making the first month the earliest shared opportunity to improve learning momentum across every course.

Our team's key metric was On Track rate: a student was on track when they submitted at least the expected number of weekly assignments.

Onboarding needed to help students navigate the platform, prepare required tools, submit assignments, access mentor and tutor support, and understand how to stay on track. Yet only 13.2% completed all recommended activities, and average confidence in meeting the weekly submission goal was 3.2/5. Based on this evidence, our team pitched improving onboarding as a Q3 initiative.

We needed to help students move from enrollment to active learning as quickly and confidently as possible.

Instead of treating time spent in onboarding as engagement, we defined engagement through the actions that demonstrated students were becoming active participants in their learning experience.

For this project, engagement did not mean keeping students inside onboarding for longer. In fact, the opposite was true. A successful experience should help students leave onboarding and begin learning as quickly as possible. We looked at engagement as a progression:

This meant looking beyond onboarding completion toward behaviors such as:

Together with the Product Owner, I defined three hypotheses to guide our research.

1. Learning by doing would create stronger engagement than passive instruction. Rather than telling students what the platform could do, we believed that guiding them through essential actions would help them experience its value directly.

2. Early commitment would help students turn motivation into a habit. We believed that establishing an achievable learning goal during onboarding could give students an early commitment to work toward and make progress feel more concrete. 

3. Relevant guidance would encourage students to participate. We believed that collecting information about students' backgrounds, motivations, and availability would allow the product to offer more relevant guidance and resources.

Our onboarding survey received 212 responses. It revealed where students needed clearer support before starting their course.

Students were enrolling, but onboarding wasn't preparing them to start learning with confidence.

We interviewed 9 students across different courses. Three recurring struggles emerged:

1. Too much, too early. Students received information before it became relevant, so much of it was forgotten.

2. One onboarding experience didn't fit everyone. Backgrounds, confidence and goals varied widely. Beginners needed more preparation and reassurance, while generic guidance often felt irrelevant.

3. The preset weekly submission goal was hard to judge. New students didn't yet know how long different tasks would take, so they couldn't tell whether the suggested weekly goal was realistic.

Students didn't need more information. They needed the right information at the right moment. Relevant guidance made students feel more prepared to act.

CareerFoundry had identified four broad student personas based on factors such as previous experience, motivation, available study time, and confidence. I mapped the onboarding journey for each persona, including key actions, questions and concerns, emotional highs and lows, product touchpoints, support needs, and potential intervention points.

I facilitated an ideation workshop with the Product Owner, engineers, and content stakeholders. The workshop was structured around the student journeys: team members identified opportunities at each stage and contributed ideas based on their area of expertise.

Then I ran a Crazy Eight brainstorming session with the team, so everyone had the opportunity to sketch their ideal solutions. Here are the 3 How Might We questions that I defined to open our creative thinking.

Opportunity mapping led us to rethink onboarding end-to-end. We prioritized ideas by impact and effort, choosing high-impact directions with realistic implementation effort for the first prototype: a short background survey, an interactive platform tour, a weekly study-time goal, an onboarding checklist, and a tool preparation page.

The experience begins with a short survey that asks students about previous knowledge and professional experience, motivation for taking the course, weekly availability, confidence level, and learning goals.

The information could be used to tailor resources, set more realistic expectations, and provide better support throughout the course. The survey was intentionally concise — its role was not to create an extensive student profile, but to collect enough context to make the next steps more relevant.

Engagement principle When students understand why something is relevant to them, they are more likely to participate.

Instead of asking students to read a long onboarding document, the platform introduces essential features through short coach marks.

The walkthrough helps students become familiar with the platform while actively using it. This reduces the gap between receiving information and applying it.

Engagement principle Action creates understanding and momentum more effectively than passive instruction.

We initially explored a personal weekly study-time goal alongside the preset submission target. The hypothesis was that greater ownership could build confidence and a routine. Testing challenged this: two goals created more mental load, and students wanted control over their progress rather than another goal to manage.

We replaced the two-goal model with one adjustable weekly submission goal. CareerFoundry recommends a target based on the pace needed to stay on track; students can increase or reduce it, with the expected graduation date updating transparently. They can adjust their goal again as circumstances change.

Engagement principle Give students control over the pace — while keeping the consequences clear.

Based on the workshop outcomes, I created a wireframe connecting the background survey, contextual platform tour, tool preparation, goal setting and onboarding checklist.

We tested the logic before polishing the UI through moderated usability sessions via UsabilityHub with 6 current students across UX Design, Web Development and Data Analytics.

1. Personalization felt valuable. The short background survey helped students feel heard before they started learning.

2. Tool preparation reduced anxiety. Setup support was especially helpful for UX Design students new to the field.

3. Two goals created more mental load. A personal study-time goal alongside the preset weekly submission goal made planning more complex. Students wanted more control over their progress, not another goal.

4. Gamification slowed students down. Progress mechanics competed with what students actually wanted: to start learning.

We revised the interaction model, built a high-fidelity interactive prototype and ran 10 moderated usability tests with CareerFoundry students via UsabilityHub. The core onboarding sequence and updated goal-setting model were understood clearly, giving us confidence to move into final design.

Context was shared continuously, not transferred at the end. By build, engineering already understood the problem, evidence, design decisions and trade-offs. Their input helped simplify the logic for self-adjusted weekly submissions, while content and curriculum stakeholders rewrote guidance for different student personas and made the tool preparation section possible.

1. Engagement starts with meaningful action. Students didn't need to spend longer in onboarding. They needed to understand, get ready, start learning, build a routine and stay on track.

2. Better decisions came from different perspectives. Students challenged our assumptions around goals and gamification. Engineering simplified the submission-goal logic, and curriculum experts made preparation content useful for different backgrounds.

3. Give students control with clear consequences. One adjustable submission goal reduced mental load while making the effect on the expected graduation date transparent.

4. Use AI to accelerate learning, not replace product judgment. AI wasn't part of my original workflow. Today, I'd use it to synthesize research while checking raw evidence, challenge hypotheses while prioritizing with evidence, and build functional prototypes to test behavior earlier.