Example Draft: The Impact of Personalization in Web-Based Credit Card Applications

Generation Statistics

Download Draft

Thesis Topic

The impact of personalisation options in a web-based credit card application - A longitudinal online controlled experiment

This master's thesis presents a longitudinal online controlled experiment examining how personalization options in web-based credit card applications affect user engagement, completion rates, and conversion metrics. The research provides actionable insights for fintech UX design.

Abstract

Digital financial services increasingly rely on personalization to improve user experience and conversion rates. This thesis investigates the causal impact of personalization options in credit card application flows through a rigorous A/B testing framework conducted over a 12-month period.

Our experimental design randomly assigns users to control (standard flow) and treatment (personalized flow) conditions, measuring primary outcomes including application completion rate, time-to-completion, and 30-day activation rates. Secondary outcomes include user satisfaction scores and support ticket volume.

Key Research Questions

• RQ1: Does personalization increase credit card application completion rates?
• RQ2: How does personalization affect time-to-completion and user friction?
• RQ3: Are treatment effects consistent across demographic segments?
• RQ4: What is the long-term impact on card activation and usage?

Methodology Highlights

• Randomized controlled trial with N=50,000+ participants
• Stratified randomization by age, income, and credit score bands
• Intention-to-treat and per-protocol analyses
• Multiple testing correction using Benjamini-Hochberg procedure
• Heterogeneous treatment effect analysis using causal forests

Related Examples

• Example: ESG Self-Financing Portfolio Analysis
• Example: LLM Quantization for Integer-Only Hardware
• Example: GenAI in Software Engineering
• Example: Building Professional Resilience