Cloud9

The Problem

Many young adults attempting to quit vaping struggle not only with withdrawal, but with understanding their progress and coping with symptoms over time. Existing quitting tools often focus on streaks or abstinence alone, offering limited reflection, personalization, or social support. This creates a fragmented experience where users lack insight into their patterns and feel isolated during the quitting process.

Phase 1: Research & Discovery

This phase focused on understanding the lived experience of quitting vaping through primary research and synthesis. Over two sprints, we conducted expert interviews and a user survey, then translated findings into personas, user stories, feature epics, and early wireframes. This research fundamentally shaped Cloud9's language, feature direction, and approach to behavior change.

Personas

Based on our research, we created four personas to guide our design decisions:

• Riley the busy college student
• Taylor the high school student with friends who vape
• Matt the parent of a vaping teenager
• Alex the adult who is struggling to quit

As the project progressed, Riley emerged as our primary persona due to strong alignment with our research findings and the team's ability to design and validate features within this context. Her story provided a clear narrative thread that guided the development of Cloud9's core features, ensuring that each experience was grounded in a relatable and empathetic context.

Riley Campbell

Age: 22 | Senior Engineering Student | West Lafayette, IN

Riley is a senior engineering student preparing for graduation and the transition into full-time work. She started vaping as a way to manage stress during demanding semesters and now worries about withdrawal impacting her focus and mental health. As she looks toward her next chapter, she wants to quit vaping without adding more pressure to an already stressful routine.

"I want to quit vaping, but I don’t want it to take over my life."

Goals

• Reduce or quit vaping without disrupting school or daily routines
• Understand what symptoms to expect and how to cope with them
• Enter post-grad life without carrying the habit forward

Pain Points

• Withdrawal symptoms increase anxiety and make it harder to focus
• Stressful academic workload triggers cravings
• All-or-nothing quitting narratives feel discouraging

Design Implications

• Surface symptom-specific insights tied to daily progress
• Support gradual reduction and reflection, not just streaks
• Provide calming, actionable coping strategies during high-stress moments

From Research to Product Direction

Research insights and personas helped us narrow our focus to a small set of problems we could meaningfully address within the scope of the project. Rather than designing for every possible scenario, we prioritized features that supported Riley's day-to-day quitting experience:

• Understanding and coping with withdrawal symptoms
• Gaining insight into one's own vaping habits
• Tracking progress through daily metrics
• Enabling real-time peer support and accountability

Based on our research and prioritized user needs, we organized Cloud9 into a set of feature epics. Each epic represented a core part of the product experience and allowed the team to work in parallel while maintaining a cohesive system.

I served as the primary owner for the Logging and Insights epics, which formed the foundation for Cloud9's personalized experience. These epics translated research insights directly into day-to-day user interactions and informed many downstream design decisions.

Phase 2: Design

In this phase, we translated our research insights into low-fidelity wireframes, high-fidelity prototypes, and database designs. We focused on refining our initial ideas based on our research and began laying the technical foundation for the product. We also developed a project management plan and began to define the technical stack for the project.

Low-Fidelity Wireframes

With core epics defined, we explored low-fidelity wireframes to validate our initial hypotheses and feature ideas before committing to visual design or implementation. These wireframes focus on the core user flows associated with logging vape usage and accessing insights into one's own quitting journey.

Home Page

The home page was designed as a daily check-in point where users can quickly understand their progress, access support, and reflect on their journey without cognitive overload.

• Persistent tracker: Time since last vape is always visible to make progress feel tangible
• Low-friction actions: Commonly used actions like logging, viewing stats, and messaging support groups are accessible immediately upon opening the app

Initial Logging Concepts

Early logging concepts focused on giving users a structured way to record vaping events and the circumstances surrounding them. At this stage, the primary goal was to support reflection and pattern recognition, and data caputred through logs would be used in data visualizations.

• Event-based logging: Users could record vaping events with date, time, and contributing triggers
• Context over judgment: Even early designs emphasized capturing circumstances rather than assigning blame. The goal was to suppport users in understanding their behavior by promoting reflection and self-awareness
• Foundational data capture: Logs were designed to create a data backbone for other features. At this stage, the intention was for the data to be used in data visualizations, but that would be changed as the project progressed

Early Insights Model

Initial insight designs were based on the assumption that time since last vape was the most meaningful indicator of user needs. Insights were delivered at predefined milestones such as one hour, twelve hours, and multiple days without vaping.

• Milestone-driven insights: Guidance surfaced based on elapsed time rather than user input
• Educational focus: Early insights emphasized general information about withdrawal timelines
• Low interaction cost: Users could quickly access insights without logging additional data

While this approach provided helpful baseline education, it treated quitting as a uniform experience and did not account for individual symptoms or emotional states.

As we conducted additional research and usability testing, these assumptions were challenged. Insights later evolved to respond directly to user logs and symptoms, enabling more personalized and empathetic guidance.

Refining the Product Direction

While early wireframes focused on validating core concepts, the transition to high-fidelity design required several foundational decisions. These decisions were informed by research synthesis, instructor feedback, and technical feasibility considerations.

High-Fidelity Prototypes

With the product direction refined, we moved on to creating high-fidelity prototypes in Figma. These prototypes were used to validate the design decisions and provide clear direction for the development phase.

Home Page

The home page remained largely unchanged from the low-fidelity prototype, but with more refined visual design and reworked navigation patterns. The primary goal remained to provide clear starting points for the main functions of the app: progress tracking and peer support.

Revised Logging and Insights

This flow demonstrates how Cloud9 responds dynamically to user input. Logging symptoms not only records data, but directly influences the guidance users receive in the moment.

• Separated logging paths: Users can log symptoms or vape usage independently, reducing friction and improving data clarity
• Symptom-driven insights: Logged symptoms trigger relevant guidance, shifting insights from time-based education to personalized support
• Just-in-time delivery: Insight popups appear immediately after logging, offering reassurance when users are most receptive

Phase 3: Development

The development phase focused on translating validated designs into a fully functional web application. Over eight weeks, I implemented core features end-to-end, balancing technical constraints, evolving requirements, and real-world usability considerations.

Adapting to Technical Constraints

One project requirement was the use of two external APIs integrated with our database. When an initially planned API was no longer viable due to technical and scope constraints, I incorporated the Google Maps API into the logging flow to capture location data for vape usage logs.

• Integrated Google Maps API to allow users to associate locations with usage events
• Stored location data alongside log entries for future analysis and insights
• Preserved the original intent of contextual logging while meeting course requirements

Built Product Preview

The following screenshots show the application in a functional state during development, demonstrating key user flows and system behavior.

Home Page During Development

This screenshot shows the home page as implemented midway through development. At this stage, almost all of the core features had been implemented and were ready to be validated through usability testing.

• Functional tracker: Time since last vape updated dynamically based on user logs
• Integrated discussion board: Community content was live and connected to user accounts
• Actionable entry points: Logging symptoms or vape usage remained one click away

Symptom Logging and Insights

This flow demonstrates the system behavior after a user logs symptoms. Logging directly triggers relevant insights, creating a feedback loop between user input and guidance.

• Symptom logging: Users could select symptoms and add contextual notes, which were stored and processed by the system
• Immediate insight delivery: Logging symptoms triggered a relevant insight without requiring additional navigation
• Progressive disclosure: Users could move from a brief popup to a full article with coping strategies to provide more detailed information when users were ready
• Data-driven behavior: Insight content was determined by logged symptoms to provide personalized and relevant guidance

By the end of the build phase, Cloud9 was a fully functional system with end-to-end flows supporting logging, insights, and community interaction. While the core architecture and feature set were in place, this stage also marked the point where assumptions could be validated through real user feedback.

Phase 4: Iteration

With a functional system in place, we entered the iteration phase to evaluate how well Cloud9 supported users in practice. This phase focused on identifying friction across core flows and validating whether design decisions translated into intuitive experiences.

Usability Testing

We conducted moderated usability testing with a mix of mental health professionals, addiction researchers, and vape users. Sessions began with background questions, followed by free exploration using a think-aloud protocol, and concluded with a set of task-based evaluations covering the full system. After each task, participants rated difficulty and shared feedback, followed by exit questions to surface additional insights.

Key Findings

While testing confirmed some known issues, it also revealed more systemic problems with navigation and comprehension. Users consistently struggled to locate core features and understand how insights related to their logged data.

• Navigation friction: Users found the nav bar cluttered and misaligned with their goals
• Log discoverability issues: Participants had difficulty finding and returning to log forms
• Insight confusion: The popup-based insight system felt disconnected and easy to miss

Final Design

Home Page

The home page remained the central hub for the app, but with more refined visual design and clearer navigation. The navigation bar was simplified to focus on primary user tasks, with less relevant pages moved to the footer.

Symptom Logging and Insights

Usability testing showed that users found the insight popup disruptive and disconnected from their logs. In response, the insight system was redesigned to live directly alongside logged data in a dedicated log details page.

• Direct transition: After submitting a symptom log, users are taken directly to a log details page rather than returned to the home page
• All relevant insights: Insights are displayed for every selected symptom, eliminating randomness and information loss
• Clear relationship between data and support: Logs and insights are now visually and conceptually linked

With these changes in place, the link between logged data and supportive insights became more direct and intuitive. Users could now see exactly how their logged symptoms were being used to generate relevant guidance, and could easily return to the log details page to review whenever they needed support.

Phase 5: Delivery

The final phase focused on presenting Cloud9 to a broader audience through visual, interactive, and in-person formats. This stage emphasized clear communication, storytelling, and demonstrating the product’s value beyond the classroom.

Capstone Fair Presentation

Cloud9 was presented at the senior capstone fair, where attendees explored projects through live demos and conversations with project teams.

Capstone Poster

We created a project poster to communicate the problem, solution, and key features of Cloud9 at a glance. The poster was designed for a non-technical audience and served as the primary artifact during the capstone fair.

Video Demo

Outcome & Recognition

Cloud9 was awarded the Students’ Choice Award, selected by fellow capstone students as their favorite project among 53 teams.

Reflection & Takeaways

Cloud9 was my first opportunity to carry a product from initial ideation through final delivery. It challenged me to balance product vision with technical feasibility, time constraints, and user needs.