IronHack Payments Cohort Analysis - Ironhack DSML Bootcamp - Project I

Project Overview

IronHack Payments, a forward-thinking financial services company, offers cash advance solutions. As part of their continuous improvement efforts, IronHack Payments commissioned a cohort analysis project. The goal is to analyze user cohorts based on the month of their first cash advance and track key metrics such as service usage frequency, incident rate, and revenue over time.

Key Objectives:

• Frequency of Service Usage: Analyze how often each cohort uses the cash advance services.
• Incident Rate: Measure payment incidents and track their variation across cohorts.
• Revenue Generation: Calculate the total revenue generated by each cohort over time.
• New Metrics: Propose new metrics to gain further insights into user behavior.

Contributors

• [Annie Meneses] - [GitHub Profile Link] | [LinkedIn Profile Link]
• [Emilia Horton] - [GitHub Profile Link] | [LinkedIn Profile Link]
• [Marc Jahnert] - [GitHub Profile Link] | [LinkedIn Profile Link]

Links to Relevant Resources

• [Presentation Link]: Project presentation on Google Slides.
• [Trello Board Link]: Project task and progress management.

Data Analysis Tools

• Python (Pandas for data manipulation and analysis)
• Tableau (Optional: For visualizing results in a dashboard - Not implemented yet.)
• Streamlit (Optional: for creating an interactive app - Not implemented yet.)

Key Deliverables

1. Python Code: Well-documented code for cohort analysis, including data loading, preprocessing, and visualization.
2. Tableau Dashboard: Visual representations of cohort behavior and insights.
3. Exploratory Data Analysis (EDA) Report: Insights and visualizations from the initial data analysis phase. (Refer to presentation.)
4. Data Quality Report: Documentation of data issues (e.g., missing values, outliers) and how they were addressed. (Refer to presentation.)
5. Presentation: A presentation summarizing key findings from the cohort analysis, the EDA Report and DQ Report.

Dependencies

• pandas - For data manipulation and cohort analysis.
• matplotlib - For plotting visualizations.
• seaborn - For creating attractive statistical plots.
• tableau - For publishing the analysis dashboard.
• streamlit - Optional: For building an interactive web app for the analysis.
• os - For interacting with the operating system, managing paths, etc.
• warnings - For handling and filtering warnings in Python scripts.
• initial_exploration - Custom module for initial exploratory data analysis.
• data_cleaning - Custom module for cleaning and preprocessing the dataset.