Objective Business problem Approach Key findings Recommendations
Purpose
The goal of this showcase is to provide actionable insights to help optimize marketing strategies and improve conversion rates for term deposits. The project consists on: (1) I analyzed data from a bank’s marketing campaigns to reveal factors that drive customer engagement, (2) By exploring patterns in demographics, financial status, and previous campaign outcomes, (3) I aimed to uncover what influences customers to make a deposit. This analysis provides insights that can guide future strategies, helping the bank target the right audiences with the most effective messages. This dataset was more used for Machine Learning but I decided to use it as first data analysis hands-on.
- Understanding the Dataset:
- The dataset comes from UV Urvine repository and the information is from a Portuguese bank’s direct marketing campaigns, where customers were contacted to subscribe to term deposits. The goal of my analysis was to determine which customers were more likely to subscribe and how to improve campaign efficiency.
Data dictionnary
| Column Attributes | Type | Description |
|---|---|---|
| age | numeric | |
| job | categorical | “admin.”,”unknown”,”unemployed”,”management”,”housemaid”,”entrepreneur”, “student”, “blue-collar”,”self-employed”,”retired”,”technician”,”services”) |
| marital | categorical | “married”,”divorced”,”single”; note: “divorced” means divorced or widowed) |
| education | categoricla | “unknown”,”secondary”,”primary”,”tertiary”) |
| default | binary | has credit in default? |
| balance | numeric | average yearly balance, in euros |
| housing | binary | has housing loan? |
| loan | binary | has personal loan? |
| contact | categorical | contact communication type: “unknown”,”telephone”,”cellular”) |
| day | numeric | last contact day of the month |
| month | categorical | last contact month of year |
| duration | numeric | last contact duration, in seconds |
| campaign | item | number of contacts performed during this campaign and for this client (numeric, includes last contact) |
| pdays | item | number of days that passed by after the client was last contacted from a previous campaign (numeric, -1 means client was not previously contacted) |
| previous | item | number of contacts performed before this campaign and for this client (numeric) |
| poutcome | item | outcome of the previous marketing campaign (categorical: “unknown”,”other”,”failure”,”success”) |
No missing values
1. Demographic Insights:
What is the average age of individuals in the dataset?
How does the distribution of age groups look like across different job categories?
What is the percentage of individuals with different education levels (e.g., high school, undergraduate, graduate)?
How does marital status correlate with age and education level?
What is the relationship between job type and marital status?
Key insights:
- (1)
- (2)
2. Loan Defaults and Financial Behavior:
What percentage of individuals have defaulted on a loan?
Is there any correlation between age and loan default rates?
How do loan defaults differ by job category?
Is there any relationship between education level and likelihood of defaulting on a loan?
What percentage of individuals with a housing loan also have a personal loan?
3. Housing and Personal Loans Insights:
How many individuals have both a housing loan and a personal loan?
What is the correlation between marital status and having a housing or personal loan?
How do housing and personal loans correlate with education level?
4. Contact and Campaign Insights:
What is the distribution of contact types (e.g., telephone, email, etc.) used for last contact?
Which contact types are most successful in converting leads to customers (if conversion data is available)?
How does the last contact date impact the likelihood of loan default or acceptance?
What percentage of individuals were contacted during each campaign, and what was the outcome?
5. Customer Segmentation and Targeting:
Which job categories are most likely to have a loan default or a housing loan?
How does the campaign outcome vary across different education levels and marital statuses?
What is the impact of marital status on the decision to take a housing loan or personal loan?
Reference
Moro, R. Laureano and P. Cortez. Using Data Mining for Bank Direct Marketing: An Application of the CRISP-DM Methodology. In P. Novais et al. (Eds.), Proceedings of the European Simulation and Modelling Conference - ESM’2011, pp. 117-121, Guimarães, Portugal, October, 2011. EUROSIS.
https://archive.ics.uci.edu/dataset/222/bank+marketing
License
- This dataset is licensed under a Creative Commons Attribution 4.0 International (CC BY 4.0) license. This allows for the sharing and adaptation of the datasets for any purpose, provided that the appropriate credit is given.