Situation
This the first part of RevoBank analysis, focused on EDA and A/B testing analysis.
As a Data Analyst on RevoBank’s Card Partnerships team, I was tasked with evaluating and optimizing a 6-month promotional rewards program with RevoShop to increase credit card usage and revenue while reducing program costs. My role was to analyze customer data to provide insights into behaviors that could inform strategies for customizing the rewards by segment to boost engagement in a more cost-effective manner. The goal was to leverage data-driven insights to tailor the program to customer preferences and sensitivities in order to improve partnership value.
Backround
RevoBank implemented an exclusive promotion for its credit card customers in partnership with RevoShop, an ecommerce marketplace. This promotion rewards cardholders with points for shopping at RevoShop to incentivize spending.
After running this promotion for 6 months, RevoBank wants to optimize the program. While it’s driving card usage and merchant revenue, the promotion comes at a high cost to the bank. RevoBank needs data-driven insights into customer spending behaviors to identify ways to refine the program for maximum return.
By analyzing historical customer transaction and demographic data, I can segment users and uncover differences in behaviors and promo sensitivity. These insights can inform strategies to customize the rewards by user group to boost engagement in a more cost-effective manner. My analysis will provide the business intelligence needed to increase card usage and revenue from the partnership while reducing the bank’s promotion expenses.
Problem Definition
The problem was that RevoBank lacked a data-driven understanding of customer behaviors to strategically tailor the RevoShop rewards program for maximum return.
Objective
The objectives of this analysis were to:
- Evaluate the overall performance of the 6-month RevoShop rewards promotion
- Understand customer behaviors through descriptive statistics and trend analysis
- Prove that our promotional usage for several merchant are succeed
Task & Action
| Task | Action | Reason |
|---|---|---|
| Define analysis goals | Wrote out key goals and business questions for exploration | Focused analysis on delivering actionable insights |
| Data Cleaning | Removed irrelevant rows, handled missing values, fixed data types, added promo sensitivity column | Prepared high quality data for accurate analysis |
| Descriptive statistics | Calculated metrics like customer demographics, sales, promo costs and impact | Provided overview of promo performance |
| Exploratory Data Analysis | Analyze more insight about customer demographics, promotional usage, month on book, transaction, and generated cost/revenue | Main activity to capture general insight about RevoBank campaign |
| Trend analysis | Compared metrics across segments and communication levels | Revealed opportunities to optimize promo by segment |
| T-Test | Perform t-test as A/B testing to validate whether RevoBank promotion campaign was successfully generate more income | Make sure our hypothesis are validated objectively by statistical measurement |
| Make recommendations | Suggested data-driven ways to customize promo by segment based on analysis | Informed strategies to boost engagement and reduce costs |
Result - Slide Deck
Attachments
» Dataset Link « | » Data Dictionary Link « | » Colab Link «
I’m grateful you took the time to explore my portfolio site and view my work. Your interest means a lot.
Feel free to provide any feedback or ask questions - I welcome your thoughts. Thank you again for checking out my projects and considering my skills. I appreciate you taking the time to do so.