2024 - PRESENT
UX Design Internship @Chai
PawPlan
Empowering First-Time Dog Owners Through an AI-Powered Web App
PawPlan is an AI-powered web app designed to support first-time dog owners with personalized routines, real-time guidance, and centralized resources.
As the sole UX designer, I led the entire design process from end-to-end, conducting research, prototyping, and usability testing.The final MVP empowers new owners to build sustainable habits, reduce overwhelm, and confidently navigate early dog ownership.
📍 Harrisburg University: MS Human-Centered Interaction Design capstone project
TIMELINE
Research: Mar - May 2024
Design: Aug - Nov 2024
MY ROLE
UX Designer (solo)
Advised by Professors Richard Wirth, John McKnight, and Nathan Aileo
TOOLS
Figma, Miro, Microsoft Teams, PowerPoint, Excel
96%-100%
Task completion in final usability testing
6.7/7.0
Average ease of use rating across key tasks
"It's straightforward. I'd love to use this if it becomes available"
- Round 3 Testing Participant
The Problem
DISCOVER THE PROBLEM SPACE
During the COVID-19 pandemic, pet adoptions surged—but with that came a steep rise in pet surrenders.
Many first-time dog owners, though well-intentioned, were unprepared for the daily responsibilities of pet care. This mismatch often led to confusion, stress, and overwhelmed shelters.Without follow-up from shelters or consistent guidance, they rely on Reddit threads, YouTube videos, and expensive trainers—adding to their stress and uncertainty.
“I've always wanted a dog, but I've never been an owner before. I don't even know where to begin.”
- Interview Participant 5
UNDERSTAND THE USERS
First-time dog owners vary widely in their motivations, expectations, and everyday lives. Some craved structure. Others didn’t know they needed it.
Designing for that kind of unpredictability meant building flexibility and guidance into every layer of the product.
-
Would users trust AI-generated routines?
-
Would they embrace structure—or find it overwhelming?
-
Could one centralized tool reduce fragmentation without adding complexity?
These were the assumptions I had to test, challenge, and adapt through research.
OTHER CONSIDERATIONS
My Bias and Perspective
As a fellow pet owner, I knew firsthand how unpredictable dog ownership can be—one moment your pup is your best friend, the next they’re chewing through your charger. I deeply empathized with new owners, especially those balancing care with full-time work.
Still, I knew my experience wasn’t universal. Throughout the project, I made a conscious effort to let the research speak louder than my own instincts—and designed based on user patterns, not personal anecdotes.
The Broader Impact
This wasn’t just about designing a pet app. It was about reducing the number of preventable pet surrenders and supporting human–animal relationships.
🐕 Encourages responsible ownership and behavior-based care
🤝 Bridges the post-adoption support gap
💸 Helps users avoid unnecessary spending on fragmented tools
What was the problem?
During the COVID-19 pandemic, a surge in first-time dog adoptions exposed a gap in ongoing support—many owners felt overwhelmed, unprepared, and unsure where to turn. This caused a steep rise in pet surrenders. Many first-time dog owners, though well-intentioned, were unprepared for the daily responsibilities of pet care. This mismatch often led to confusion, stress, and overwhelmed shelters.
Who was it for?
First-time dog owners navigating their first year of pet ownership, often without structured support or prior experience.
Why did it matter?
Shelters and adoption centers are still struggling with the aftermath of impulsive pandemic adoptions. Without the right education or support systems, dogs are returned due to behavioral issues that might have been preventable with proper routines and guidance.
Constraints
Timeline
-
14 weeks of primary and secondary research
-
14 weeks of ideation, testing and iteration
Team
-
Team of one! As the only designer on this capstone, I managed the entire process end-to-end, from research to final UI.
-
Participant outreach was limited to my personal and local network, which shaped the scope of interviews and testing.
-
Since this was not an industry project, I worked without direct developer input or implementation feedback.
Limited Technical Scope
-
As a conceptual prototype, PawPlan was not developed into a live product, so all features had to be clearly communicated through design alone.
Ready for more? Keep scrolling for the full case study!
Part 1: Discovery through Research
The Solution
WHAT I BUILT
PawPlan is an AI-powered web app designed to support first-time dog owners from the very beginning and every step of the way.
It acts as a one-stop hub for personalized routines, behavior guidance, and expert-reviewed resources—all tailored to the owner’s lifestyle and the dog’s needs. The platform adapts as the dog grows, helping users build confidence, consistency, and long-term habits.
1/
Homepage: Everything the user needs to know about PawPlan
The homepage introduces PawPlan as a trusted guide for new dog owners. It highlights Fetch, the AI assistant, and gently walks users through what to expect—building trust before signup.
2/
Onboarding: Account Setup and Personalized Intake
The user begins by creating an account and answering a series of tailored questions.
This intake process sets the foundation for a care plan that adapts to both the owner’s lifestyle and the dog’s needs.
3/
Dashboard Overview: From Guidance to Daily Support
The dashboard acts as the user’s command center—combining their personalized schedule, care milestones, insights, and access to Fetch, the AI assistant. It’s built to feel calm, intuitive, and adaptive as both the owner and dog grow together.
The Process
ITERATIVE DESIGN PROCESS
So how did I actually build this?
Behind the final solution was a full-cycle UX process—strategic, iterative, and hands-on.
Research & Analysis
Ideation
Iterative Design
Prototype & Usability Testing
Final Outcome
RESEARCH
To design real support, I needed to understand what breaks down and why through research.
Secondary Research
-
10 literature reviews on pet behavior, adoption trends, and owner psychology
Primary Research
-
10 in-depth interviews with dog owners (from new owners to veterans)
-
2 expert interviews with MSPCA Boston shelter staff
-
4 ethnographic field visits to dog parks
SECONDARY RESEARCH: LITERATURE REVIEW
Through literature review, I explored studies on responsible dog ownership, behavioral training, and adoption trends.
I also reviewed educational materials from MSPCA Boston and training philosophies from leading experts like Cesar Millan, Hyung Wook Kang, and Turid Rugaas. These sources consistently emphasized the importance of clear structure, early intervention, and owner education—shaping my design approach to focus on proactive, trustworthy guidance from the very first interaction.
PRIMARY RESEARCH: USER INTERVIEW
I conducted in-depth interviews with dog owners across varying experience levels, ranging from first-time adopters to long-time owners.
I chose this interview participant group intentionally to reflect a spectrum of preparedness and care strategies. My goal was to understand how different types of owners transition into dog ownership, where they struggle, and how their expectations shape outcomes. I analyzed responses thematically and uncovered patterns around preparedness, daily routines, and shifting responsibilities.
Quote
“(Having a new dog) was a little bit challenging because now I wake up at least two hours earlier than my usual time to walk my dog and give her some attention and feed her before I go to work.”
- Interview Participant 3
Quote
“We watched Cesar Millan and read up on Huskies before adopting. That helped a lot.”
- Interview Participant 4
PRIMARY RESEARCH: expert interviews
At MSPCA Boston, I learned that staff make extra effort to match dogs with suitable adopters, often walking them through specific care needs, behavior expectations and lifestyle fit.
Even with these precautions, staff reported that pet returns still occurred frequently—especially when owners underestimated the time commitment or misunderstood a breed’s needs. Compatibility, they emphasized, was often more important than enthusiasm alone.
PRIMARY RESEARCH: ETHNOGRAPHIC FIELDWORK
At dog parks, I observed clear differences in dog behavior depending on how attentive owners were.
Dog parks
Dogs with attentive, engaged owners appeared more regulated and calm. Others, accompanied by distracted owners, often displayed stress or pulled uncontrollably on leash. This reinforced the importance of owner presence and consistency during outings
ANALYSIS
What I learned from research
By combining literature, interviews, expert insights, and field observation, I uncovered patterns from multiple angles—each revealing a different facet of first-time dog ownership. Together, they formed a fuller picture of real-world challenges. My next step as a UX designer was to synthesize these into clear, actionable direction.
Key Findings
1/
Owners who planned ahead experienced smoother transitions.
Those who researched breed traits and built routines reported feeling more prepared, while impulsive adopters felt overwhelmed and underprepared.
2/
Attentiveness and consistency shaped dog behavior.
Engaged owners—especially in public settings—tended to have calmer, more well-regulated dogs, leading to lower stress and more confidence in daily care.
3/
Fragmented resources increased confusion.
Many first-time owners relied on scattered tools like Reddit and YouTube, leading to conflicting advice and a lack of reliable structure.
4/
Most pet returns stemmed from unmet expectations.
Experts cited lifestyle mismatches, miscommunication, and underestimating long-term responsibilities as key reasons for surrender.
SYNTHESIZING INSIGHTS
Translating findings into insights
After mapping findings across methods, several patterns stood out—especially around expectations, decision-making, and day-to-day follow-through. While each user story was different, the core pain points consistently pointed to gaps in preparation, routine support, and clarity of guidance.
1/
Preparation Matters More Than Experience
Owners who anticipated lifestyle changes adapted better than those who didn’t plan. Many issues stem not from neglect, but from mismatches between owner hopes and reality.
2/
Fragmented Tools Create Frustration
Most new owners relied on scattered sources like Reddit, blogs, and YouTube—resulting in decision fatigue, stress, and unnecessary costs.
3/
Support Needs to Be Continuous, Not Just Front-Loaded
While initial resources exist, most owners struggled to maintain structure and confidence over time—especially as their dog’s needs evolved.
HOW MIGHT WE STATEMENT
And so how might we proactively support first-time dog owners with continuous, trustworthy guidance throughout the first year of dog ownership?
PawPlan was designed to directly address the core pain points and user needs uncovered during research.
Pain Points
-
Lack of preparation and reliable guidance
-
Difficulty maintaining consistent care routines
-
Confusion about dog behavior and training expectations
-
Reliance on fragmented tools or paid resources
User Needs
-
Get clear, trustworthy guidance early on
-
Stay consistent through changing schedules and responsibilities
-
Access reliable, contextual help without guesswork
-
Simplify support and reduce long-term financial burden
Design features
-
AI-powered, step-by-step onboarding and behavior guidance
-
Smart, lifestyle-adapted routine planner controlled by the user
-
24/7 chatbot with real-time, curated answers personalized to pet stage
-
All-in-one platform that minimizes external costs and tool overload
IDEATION
Exploring where structure, empathy, and guidance could take form.
To move from insights to ideas, I used a mix of collaborative and analytical methods—brainstorming, affinity mapping, card sorting, and the Thinking Hats exercise. These helped me explore different ways a digital tool could offer support that felt timely, trustworthy, and adaptable.
I also mapped key user scenarios across various entry points: adopting from shelters, buying from breeders, or taking in dogs unexpectedly. This exercise revealed shared moments of uncertainty and need—such as the first week at home, training breakdowns, or health concerns—where the product could step in with structure, without being rigid.
Early idea storming—capturing anything that could support first-time owners.
Clustering ideas helped identify where emotional and practical support could overlap.
Mapping different owner journeys revealed shared friction points—especially during onboarding and behavior uncertainty.
I evaluated my early ideas using the Thinking Hats exercise—a technique that allowed me to step into different perspectives such as optimism, caution, logic, emotion, and creative alternatives. This method helped me stay objective, avoid tunnel vision, and refine concepts based on both feasibility and empathy.
ITERATIVE DESIGN
Laying the foundation through quick, low-fidelity exploration.
To define the app’s core structure, I sketched early low-fidelity wireframes focused on key flows: onboarding, profile setup, and daily routines. These screens allowed me to prioritize features, validate core architecture, and prep for early user testing.
Step 1: User Sign Up
A lifestyle-based onboarding flow tailored for different adoption or purchase scenarios: shelters, breeders, or first-time browsers.
User profile setup:
Questions capture owner routines, home environment, and dog traits to inform personalized guidance.
Calendar Dashboard
A centralized space to manage appointments, training, tasks, and vet records—mapped to daily routines.
Journey Map
To explore this, I sketched early flows and created a journey map to provide a holistic view and to map out specific instances of pain points which can be translated into design opportunities.
PrototypING
I skipped the mid-fidelity and tested fast with users.
After early rounds of testing with low-fidelity wireframes, I quickly ran into a challenge: participants focused too heavily on placeholder visuals and layout details, distracting from the core experience I was trying to validate. Because this product relied on building emotional clarity and behavioral trust, I decided to move directly into high-fidelity prototyping.
This shift allowed me to test with more realism and get richer feedback, evaluating tone, structure, and interaction in context. During testing sessions, I did my best to redirect participants from fizting on visuals and features that were not being tested.
Over three usability testing rounds, I rapidly iterated to validate flows, uncover friction points, and better align the experience with user expectations.
Users misinterpreted layout in low-fi. High-fidelity clarified tone and supported more meaningful feedback.
ACCESSIBILITY CHECK
Designs meet WCAG 2.1 standards.
To ensure readability and inclusivity, I tested color contrast across all UI elements. The updated palette passes both AA and AAA compliance for small text, large text, and interactive components, with a contrast ratio of 9.24:1, exceeding the minimum standard.
For instance, I tested key interface elements like the sidebar navigation for WCAG 2.1 compliance. The contrast between the background #E0EEF9 and foreground #173F62 passed AA and AAA standards for small and large text, with a contrast ratio of 9.24:1—well above the recommended minimum for readability and inclusivity.
Usability Testing
Three rapid testing rounds, each followed by focused iteration.
I ran three usability testing rounds with real users, using a mix of remote screen-sharing sessions, task success tracking, and UX-Lite confidence scoring. My approach emphasized fast iteration: validating one flow, improving it, then testing again.
Round 1
In Round 1, I tested the initial onboarding process and the default dashboard experience with new users.
Goal:
To observe how first-time users navigated account creation, provided information about themselves and their dogs, and understood the dashboard content.
Tasks given to participants:
-
Create an account
-
Go through the onboarding flow
-
Land on the dashboard to complete tasks such as finding their schedule and editing their dog profile.
“Where is the plan? I was expecting to see things about my dog like breed, recommended exercise and training plans.”
- Round 1 participant
This revealed a misalignment in user expectations for the dashboard. Users were expecting to see more. The next step is to add personalized content and plans.
Key Finding
Round 2
In Round 2, I tested revisions to the dashboard layout and onboarding guidance based on insights from the first round.
Goal:
To validate whether the updated structure, revised language, and clearer microcopy helped users better understand the dashboard and complete tasks with confidence.
Tasks given to participants:
-
Create an account
-
Go through the revised onboarding flow
-
Find a specific feature (e.g. training plan) and complete a basic task on the dashboard
“I got confused at first, but once I read the guidance text again, it made more sense. I didn’t realize that the plan would show up based on my answers.”
- Round 2 participant
Guidance language was helpful but not immediately clear. Users needed better indicators that their input directly shaped the dashboard.
Key Finding
Round 3
In Round 3, I tested the final version of the prototype with full onboarding, personalized plan content, and chatbot support.
Goal:
To test the final version of the high-fidelity prototype and assess whether users understood how to use AI-generated support features.
Tasks given to participants:
-
Complete full onboarding
-
Access personalized dog info, training plans, and guidance via a dashboard
-
Use chatbot for a question or check-in
“I love how calming the interface feels. I wish this existed when I got my dog.”
- Round 3 participant
Final version successfully delivered clarity and emotional trust.
Key Finding
Usability Metrics and Outcomes
Did the final design actually work better for users?
To measure impact, I tracked task success and ease of use across three testing rounds. These metrics helped validate whether design changes made things clearer, faster, and more intuitive—especially for first-time dog owners navigating a complex onboarding flow.
-
Task Success Rate: The percentage of users who were able to complete assigned tasks without assistance. A high success rate indicated that layout, labels, and flow were intuitive and goal-oriented.
-
UX-Lite Ease Score: A 7-point scale that measures user confidence and perceived effort. A score of 7 reflects a fully confident, low-friction experience, while anything under usually points to lingering hesitation or friction.
Round 1
-
Task success: 86%
-
UX-Lite Score: N/A
-
Key Change: Overhauled dashboard layout and added personalized content
Round 2
-
Task success: 100%
-
UX-Lite Score: 6.8/7.0
-
Key Change: Simplified navigation and clarified guidance language
Round 3
-
Task success: 98%-100%
-
UX-Lite Score: 7.0/7.0
-
Key Change: Validated final features, adjusted layout based on real usage
My Reflection
With each round of usability testing and design tweaks, the metrics fluctuated as changes were made. At first, some of the adjustments led to unexpected shifts, but as I kept improving the design based on user feedback, the metrics started to bounce back and gradually get better.
Final Outcome
What changed for users and why it matters
PawPlan gave new dog owners one calm, centralized place to set up routines, receive behavior-based guidance, and learn without judgment. It offered structure, reassurance, and follow-through—key ingredients for reducing overwhelm, improving care, and lowering the risk of abandonment.
Built to serve both users and the broader pet care ecosystem
The final solution addressed user and business needs by minimizing fragmented support, reducing post-adoption stress, and encouraging long-term, compassionate ownership.
Final MVP Features
After 14 weeks of design and testing, I delivered a high-fidelity MVP (Minimum Viable Product) with:
🧠 AI onboarding and milestone-based scheduling
🗓️ A lifestyle-matching planner tailored to dog and owner
💬 24/7 contextual chatbot support
✨ A calm, emotionally thoughtful UI grounded in real user feedback
Final Thoughts
Designing for trust, care, and future potential
PawPlan has the potential to reduce rates of dog surrender by equipping first-time owners with proactive tools, emotional reassurance, and behavioral clarity—before frustration or miscommunication escalates. The project also serves as a proof-of-concept for how AI can be integrated into emotionally sensitive, behavior-based experiences without compromising user trust.
The final deliverable was presented to academic advisors and design faculty as part of my HCID capstone, where it received strong and positive feedback for its research depth, and real-world application potential.
" My daughter and I would love to use this if it actually becomes available. "
- Round 3 participant
Future Opportunities
Where PawPlan could go next
If developed further, PawPlan could evolve into a more robust, intelligent companion by supporting:
-
Mobile Companion App
Extend usability with real-time reminders, check-ins, and location-aware features on the go.
-
Shared Care Coordination
Enable multiple caregivers (e.g. household members, walkers, sitters) to manage routines, logs, and updates collaboratively.
-
Behavioral Tracking & Alerts
Use voice or image input to detect health or behavioral changes and prompt owners when attention is needed.
-
Third-Party Integrations
Sync with vets, trainers, groomers, or adoption platforms for seamless access to trusted resources and personalized guidance.
Reflection
What I learned through the process
Working solo over two semesters helped me develop stronger judgment—especially in moments where I needed to pivot, reevaluate a hypothesis, or simplify a feature. This project reminded me that thoughtful design isn’t just about usability—it’s about helping people feel capable and supported in moments that matter. Every decision had to balance emotional resonance, technical feasibility, and practical behavior change, and it reinforced how essential adaptability and storytelling are in crafting human-centered products.
What went well?
-
Mixed methods gave me a multi-angle view of the problem space
-
Fast testing cycles helped me catch misalignments early
-
Strong storytelling reinforced the impact during presentation
What didn’t?
-
Early wireframes distracted users from core concepts
-
Designing for AI required simplifying complex logic into digestible flows
What I'd do differently:
-
Explore adaptive automation through Agentic AI
-
Design from the start for multi-caregiver households
-
Prioritize mobile-first flows for users tracking on-the-go
April 4th, 2025
Retrospective
Since completing this project, my understanding of agentic AI and its applications has deepened. Originally, PawPlan’s AI was designed to respond to user inputs with helpful routines, insights, and contextual guidance. While effective, it relied on user prompts to activate its intelligence.
In hindsight, there is a clear opportunity to evolve PawPlan into a more autonomous, agentic system—one that not only responds, but initiates. For example, it could proactively adjust routines based on behavioral data, send gentle nudges when tasks are missed, or recommend changes based on seasonal shifts or dog development milestones.
Agentic AI also presents a chance to reduce cognitive load for overwhelmed users. Instead of asking, "What do I need to do next?" users could simply follow the next best action already queued up by the system.
My Reflection
If I were to revisit this project today, I’d explore how to blend goal-setting, task automation, and adaptive personalization through agentic AI to make PawPlan a more proactive and even more empathetic companion for new dog owners.
What do you want to see next?
Designing a digital alternative for paper brochures and maps.
View the quick study
Designing a digital alternative for paper brochures and maps.
View the quick study