Jr
PROJECTS
Case Studies
Detailed explorations of my graduate-level research projects, demonstrating analytical methodology and measurable business impact.
WELLS FARGO
AI-Powered Brand Reinvention
Developed a comprehensive AI-driven marketing strategy to transform Wells Fargo's customer engagement, leveraging machine learning for churn prediction and behavioral segmentation.
The Challenge
Wells Fargo faced declining customer satisfaction, high churn rates, and lagged behind competitors in AI-powered customer experience. The bank needed to shift from legacy issues to a forward-looking, digitally-enabled financial partner.
Methodology
—Logistic regression model with 10 input features for churn prediction
—K-means clustering to identify four distinct customer personas
—Behavioral data analysis (products, balance, tenure, activity)
—Demographic and financial health segmentation
Key Achievements
81.8% churn prediction accuracy using logistic regression
Identified 4 actionable customer personas via K-means clustering
Developed AI-driven personalization strategy for "Fargo" assistant
Created 24-month implementation roadmap for brand transformation
Quantified potential impact on churn reduction and cross-sell growth
Tools & Technologies
Business Impact
The project delivered a comprehensive framework for Wells Fargo to optimize customer engagement, reduce churn through predictive analytics, and compete with industry leaders like Bank of America's "Erica" AI assistant.
TESLA
Driving Intent: Unlocking EV Adoption
Conducted rigorous market research to identify which consumer perceptions most strongly influence the intention to adopt Tesla vehicles with autonomous features.
The Challenge
Consumers show uncertainty about adopting Tesla's EVs with autonomous features. Key concerns include safety of autonomous driving, familiarity with EV/AV technology, comfort relying on advanced systems, and pricing commitment.
Methodology
—Survey design measuring purchase intent, familiarity, safety perception, and openness to innovation
—Multiple linear regression analysis (PI = β₀ + β₁Familiarity + β₂Safety + β₃Openness)
—Independent samples t-test comparing experienced vs. non-experienced consumers
—Hypothesis testing at 0.05 significance level
Key Achievements
R² = 0.38 with F(3,52) = 10.52, p < 0.00002
Safety perception: strongest predictor (+0.71 points per unit, p = 0.0003)
Familiarity: significant predictor (+0.30 points per unit, p = 0.02)
Experience effect: d = 0.41 effect size (higher intent after test drive)
Developed pricing and bundling strategy recommendations
Tools & Technologies
Business Impact
Research provided Tesla with actionable insights: prioritize safety messaging over innovation, expand test drive programs to build familiarity, and consider flexible subscription pricing to reduce adoption barriers.