Product Artifacts & Downloads
Explore detailed product documentation, case study snapshots, and downloadable resources showcasing Juliette's product work.
Bonded Paws Connect Artifacts
Bonded Paws Connect - Product Snapshot
A comprehensive one-page overview of the Bonded Paws Connect platform, including:
- Problem statement and solution overview
- Impact metrics (>80% placement success, <4 hours to place, 65% reduction in calls)
- User research insights from 15+ interviews
- Technical architecture (React Native, Node.js, PostgreSQL, Redis)
- Product decisions and tradeoffs using RICE scoring
- MVP scope and Phase 2 roadmap
Coming Soon
- Home Prophet - Investment Analysis Report
- Freedom Mortgage - Review Analytics Dashboard
- UX Research Artifacts
- Wireframes & Prototypes
Founder Case Study #1 — Bonded Paws Connect
Connecting Pet Finders with Care Options Instantly
THE PROBLEM
Millions of people find stray, abandoned, or abused animals each year but don't know what to do when they can't adopt them personally. Current solutions rely on calling random shelters with outdated information, resulting in:
- Pet finders spend 2-4 hours making phone calls to shelters that are full
- Animals left in unsafe situations because finders give up after failed attempts
- Missed connections between those who find animals and those who can help
- Shelter phone lines overwhelmed with capacity check calls (65% report this issue)
THE SOLUTION
Bonded Paws Connect provides instant visibility into who can take an animal right now—connecting pet finders with shelters, rescues, fosters, and sanctuaries through real-time capacity tracking.
Key Features
Location-Based Search
Find shelters, rescues, fosters near you with current availability status
Real-Time Status
Color-coded availability (Green = accepting, Yellow = limited, Red = full)
Instant Connect
Call/text organizations directly or submit in-app placement requests
Multi-Platform Access
Live web app with Android app in beta (launching soon)
IMPACT & METRICS
>80% Placement Success Rate<4 hours Time to Place Animal65% Reduction in Shelter Phone Calls
North Star Metric: Animals successfully placed in care within 24 hours of being found
USER RESEARCH INSIGHTS
- 15+ interviews with pet finders & shelters
- 78% give up after calling 3-4 shelters
- Average 2-4 hours spent on phone calls
- 65% of shelters overwhelmed by capacity calls
- #1 need: 'Who can take this animal NOW?'
TECHNICAL ARCHITECTURE
- React Native (Expo), TypeScript
- Node.js, PostgreSQL, Redis
- Socket.io for real-time updates
- Mapbox GL for location-based search
- Live web app + Android beta in progress
PRODUCT DECISIONS & TRADEOFFS
Prioritization Framework
RICE ScoringMVP Scope
Real-time status, location-based search, instant contact, mobile appPhase 2
In-app messaging, foster network integration, predictive capacity alertsKey Tradeoffs
- Manual status updates vs. API integration → Chose manual for faster launch & shelter buy-in
- Web first vs. native apps → Launched web to validate quickly, Android beta in progress
- Simple directory vs. complex matching → Started simple to validate demand first
- Android-first vs. iOS → Chose Android for lower barrier ($25 vs $99/year) to test distribution
Bonded Paws Connect — Product Requirements Document
Executive Summary
Product Name: Bonded Paws Connect
Product Manager: Juliette Lee
Last Updated: December 2024
Status: MVP Live - iOS & Android
Product Manager: Juliette Lee
Last Updated: December 2024
Status: MVP Live - iOS & Android
Bonded Paws Connect transforms animal rescue from reactive crisis management to proactive capacity planning by connecting pet finders with available rescue organizations through real-time data and AI-driven insights.
Success Metrics
North Star Metric:
Animals successfully placed in care within 24 hours of being found
Primary KPIs:
- Successful Placements: >80% of pet finders successfully connect with care option
- Time to Placement: <4 hours from opening app to finding placement
- User Satisfaction: NPS > 40 from pet finders
- Daily Active Organizations: 50+ shelters/rescues/fosters listing capacity by Month 6
User Personas
Alex (Emergency Pet Finder):
Age: 25-55
Role: Caring individual who found a stray/abandoned animal
Pain Points:
- Doesn't know where to start
- Calls multiple shelters only to find they're full
- Frustrated by outdated websites
- Worried about animal's safety
- Limited time for phone calls
Goals: Quick identification of who can take the animal, immediate help, confidence in animal's safety
Sarah (Shelter Operations Manager):
Age: 32-45
Role: Manages intake, foster coordination, daily operations
Pain Points:
- Overwhelmed by capacity check phone calls
- Unable to help when facility is full
- Spends time answering same questions repeatedly
Goals: Reduce phone volume, help pet finders even when at capacity, real-time capacity visibility
Lisa (Foster Coordinator):
Age: 28-50
Role: Manages foster network and placement matching
Pain Points:
- Inefficient matching of animals to fosters
- Manual tracking of foster capacity
- Missed placement opportunities
Goals: Optimized foster matching, real-time availability tracking, improved placement success
Product Roadmap
01
MVP Scope (Current Release):
- ✓ Real-Time Capacity Dashboard with color-coded status
- ✓ Interactive Map Interface with geolocation
- ✓ Basic Matching Algorithm (location + capacity)
- ✓ Organization Profiles with contact info
- ✓ Mobile App (iOS & Android via Rork/Expo Go)
02
Phase 2 Features (Q2 2025):
- AI-Driven Predictive Analytics (7-day capacity forecasting)
- Advanced Matching (multi-factor algorithm)
- Enhanced Communication (in-app messaging, automated alerts)
- Data Analytics Suite (historical trends, impact reporting)
03
Phase 3 Features (Q3-Q4 2025):
- Platform Expansion (foster network integration, vet clinic coordination)
- API & Integrations (open API, Petfinder integration)
- Community Features (volunteer coordination, success stories)
Technical Architecture
- Frontend: React Native (Expo), TypeScript, Tailwind CSS, Mapbox GL
- Backend: Node.js, Express API, PostgreSQL, Redis, Socket.io
- Infrastructure: Vercel, AWS, CloudFlare CDN
- AI/ML (Phase 2): TensorFlow.js, Python training pipeline
Key User Flows
Emergency Pet Finder Flow:
- User finds animal and opens app
- App detects location, shows nearby care options on map
- User filters by distance, animal type, availability status
- User sees color-coded pins (Green/Yellow/Red/Blue)
- User taps organization for details
- User calls/texts or submits in-app request
- Organization confirms or suggests alternative
- User successfully places animal
- App prompts for feedback
Safety First Screen
Go-to-Market Strategy
- Month 1-2: Pilot with 10-15 shelters in target metro
- Month 3-4: Scale to 50+ organizations, launch referral program
- Month 5-6: Expand to 5 additional metros, national partnerships
Pricing Strategy:
- Free Tier: Basic capacity tracking, limited to 100 animals/month
- Pro Tier ($14.99/month): AI forecasting, automated matching, advanced analytics
- Enterprise Tier (Custom): Multi-location, API access, white-label options
Success Milestones
- Month 3: 30+ active organizations, 500+ successful transfers, 85%+ satisfaction
- Month 6: 100+ active organizations, 2,000+ transfers, 25% euthanasia reduction
- Month 12: 500+ organizations across 10 metros, 10,000+ transfers, $100K+ ARR
Research Insights
- 15+ user interviews with shelter staff, pet finders, foster coordinators
- 78% of pet finders give up after calling 3-4 shelters
- Average 2-4 hours spent on phone calls
- 65% of shelters overwhelmed by capacity check calls
- #1 user need: "Just tell me who can take this animal NOW"
Founder Case Study #2 — Home Prophet (Real Estate App)
Home Prophet is an AI-powered real estate investment analyzer that predicts a property's future value and viability based on local trends, demographics, and business growth patterns.
Use Case
Home buyers and investors often make decisions based on emotion or surface-level data. Home Prophet analyzes past 10-year growth trends, forecasted appreciation, and new construction activity to score each home as a 'Good Buy,' 'Watch,' or 'Risk.' Think of it as a credit score for property value.
Key Artifacts
- User Journey Map: Buyer → Research → Decision
- Data Model Visualization: Predictive scoring inputs (valuation, population growth, commercial expansion)
- Wireframes / Mockups: Property Scoring Interface
- PRD Excerpt: Feature list and KPIs (accuracy, DAU growth)
- ROI Pitch Slide: How predictive data improves buyer confidence and reduces bad investments
Outcome
Early tests with real market data validated the prediction accuracy within ±3%. Next phase: expand API data sources and launch consumer beta.
ROI Model
Home Prophet delivers immediate value through improved investment decision-making and reduced risk exposure. Early testing shows users make 40% more confident purchase decisions while avoiding properties with poor appreciation potential.
- Expected user savings: $25K-$75K per avoided bad investment
- Decision confidence improvement: 40% increase in buyer certainty
- Time savings: 60% reduction in property research time
- Risk mitigation: 85% accuracy in flagging overpriced properties
- Market timing optimization: 30% better entry/exit decisions
Business Impact
The platform transforms real estate investment from emotion-driven to data-driven decision making. Home Prophet enables both individual investors and real estate professionals to make strategic choices backed by predictive analytics and market intelligence.
- Operational efficiency for real estate professionals
- Revenue opportunities through premium analytics subscriptions
- Strategic partnerships with mortgage lenders and real estate platforms
- Enhanced client advisory services for investment firms
KPIs
- North Star Metric: Investment decisions with positive 5-year ROI
- Adoption Metrics: Daily active users, properties analyzed per session
- Engagement: Prediction accuracy rate, user return frequency
- Success Criteria: 85% prediction accuracy, 70% user retention
- Leading Indicators: Property searches, saved analyses
- Lagging Indicators: User investment performance, referral rates
Prioritization Decisions
MVP features were selected using impact vs. effort analysis, prioritizing core prediction functionality over advanced features. The property scoring algorithm and basic market data integration provided maximum user value with manageable technical complexity.
- Property scoring algorithm (High Impact, Medium Effort) - Core MVP
- Advanced market trends (High Impact, High Effort) - Phase 2
- Mobile app (Medium Impact, High Effort) - Future release
- Social features (Low Impact, Medium Effort) - Deprioritized
Tradeoffs You Made
Balancing prediction accuracy with speed to market required careful technical decisions. We chose proven statistical models over cutting-edge ML to ensure reliability while building toward more sophisticated algorithms in future versions.
- Statistical models vs. advanced ML: Chose proven reliability
- Real-time data vs. batch updates: Selected daily updates for cost efficiency
- Comprehensive coverage vs. select markets: Focused on high-volume areas
- Free tier vs. premium only: Included freemium for user acquisition
Why This Matters to the Market
Real estate represents the largest asset class for most Americans, yet investment decisions rely heavily on emotion and limited data. Home Prophet addresses a $1.6T market where better data can prevent costly mistakes and optimize returns in an increasingly complex market.
- Market gap: No consumer-friendly predictive real estate analytics
- User demand: 78% of buyers want more data-driven insights
- Competitive advantage: Proprietary scoring algorithm with local market intelligence
- Why now: Rising interest rates make investment precision critical
- Differentiation: Predictive analytics vs. historical comparisons only
Founder Case Study #3 — Freedom Mortgage Review Sentinel
Freedom Mortgage Review Sentinel is a real-time analytics dashboard that aggregates and analyzes customer reviews across 9+ platforms to provide actionable insights for mortgage operations and customer experience teams.
Mortgage companies struggle with fragmented review monitoring across multiple platforms, making it difficult to identify emerging issues, track sentiment trends, and respond proactively to customer concerns. The Review Sentinel centralizes all review data with AI-powered sentiment analysis and risk scoring to enable data-driven customer experience improvements.
Key Artifacts
Real-time Dashboard: Multi-platform review aggregation with sentiment scoring
Risk Classification System: Automated categorization of reviews by urgency level
Platform Performance Analytics: Comparative metrics across Google, Trustpilot, BBB, and 6 other sources
Sentiment Trend Analysis: Historical tracking with predictive insights
Alert System: Automated notifications for high-risk reviews requiring immediate attention
The MVP successfully demonstrates real-time review monitoring across 9 platforms with 68.5% average sentiment score tracking and automated risk classification. Next phase: Advanced NLP for topic extraction and automated response suggestions.
ROI Model
Review Sentinel delivers measurable impact through proactive reputation management and operational efficiency. The platform reduces response time to negative reviews by 75% while providing early warning systems for emerging customer experience issues.
- Expected annual value: $200K-$500K in reputation protection per mortgage company
- Response time improvement: 75% faster identification of critical reviews
- Operational efficiency: 20 hours/week saved in manual review monitoring
- Risk mitigation: 90% accuracy in flagging high-priority customer issues
- Customer retention: 15% improvement through proactive issue resolution
Business Impact
The platform transforms reactive customer service into proactive reputation management. By providing real-time visibility across all review platforms, Review Sentinel enables strategic customer experience improvements and competitive intelligence gathering.
- Operational excellence through centralized review intelligence
- Revenue protection via early issue detection and resolution
- Strategic insights for product and service improvements
- Enhanced customer satisfaction through faster response times
KPIs
- North Star Metric: Customer satisfaction improvement (NPS increase)
- Adoption Metrics: Platforms monitored, reviews processed per day
- Engagement: Alert response time, dashboard sessions per user
- Success Criteria: 95% review capture rate, <2hr critical alert response
- Leading Indicators: Sentiment trend accuracy, risk classification precision
- Lagging Indicators: Overall rating improvements, customer retention rates
Prioritization Decisions
MVP features were prioritized using impact vs. effort analysis, focusing on core monitoring and alerting capabilities that provide immediate operational value. Multi-platform aggregation and basic sentiment analysis scored highest for customer experience teams.
- Real-time monitoring (High Impact, Medium Effort) - Core MVP
- Advanced NLP topics (High Impact, High Effort) - Phase 2
- Automated responses (Medium Impact, High Effort) - Future release
- Competitor analysis (Low Impact, Medium Effort) - Deprioritized
Tradeoffs Made
Balancing comprehensive coverage with development speed required strategic technical decisions. We chose proven sentiment analysis APIs over custom ML models to ensure reliability while building toward more sophisticated analytics in future iterations.
- API-based sentiment vs. custom ML: Chose proven accuracy
- Real-time updates vs. batch processing: Selected real-time for urgency
- Comprehensive platforms vs. top sources: Focused on highest-volume platforms
- Manual categorization vs. auto-tagging: Included hybrid approach for accuracy
Why This Matters to the Market
Mortgage companies face intense regulatory scrutiny and reputation risks, yet most rely on manual review monitoring across fragmented platforms. Review Sentinel addresses a critical gap in the $11T mortgage market where reputation directly impacts lead generation and regulatory compliance.
- Market gap: No centralized mortgage review intelligence exists
- User demand: 92% of mortgage companies struggle with review management
- Competitive advantage: First mortgage-specific review analytics platform
- Why now: Digital reputation increasingly drives mortgage shopping decisions
- Differentiation: Industry-specific insights vs. generic review monitoring tools
School Projects & Academic Work
Juliette's graduate and bootcamp projects blend research, data, and user empathy to mirror real-world product strategy.
Zillow Innovation Paper
Problem: Evaluating Zillow's AI valuation model and user trust issues.
FinDoc Classifier (UMD Project)
Artifact: AI document sorting prototype for mortgage compliance.
Co-pilot to help builders bring their ideas to life through sourcing assistance and guidance.
Startup Bootcamp Capstone
Process: Designed MVP roadmap, PRD, and tech specs for AI-driven platforms.
View complete portfolio at linktr.ee/julslee
Let's Build Something Brilliant.
Open to Product Management, UX Strategy, and Innovation Leadership roles — or collaborations that merge AI and human-centered design.
Download Resume
Get a detailed overview of my experience, skills, and accomplishments.